introduction: social network analysis applications in complex governance networks complexity, governance & networks-vol 02, issue 01 (2015) 1–4 1 doi: 10.7564/15-cgn29 introduction social network analysis applications in complex governance networks naim kapucu school of public administration, university of central florida e-mail: kapucu@ucf.edu the growth in studying complexity, governance, and networks is a noticeable fact. complex public policy problems require productive collaborations among multiple actors from different sectors. public, private, and not-for-profit organizations work through governance networks and collaborative governance to solve complex public policy problems. there is a strong interest in both the practice and theory of networks in public governance. this interest has given boost to the use of social network analysis (sna) in studies on complex governance networks. this special issue of complexity, governance, & networks, includes theoretically, conceptually, and analytically rigorous papers with social network analysis applications. the papers help our understanding in studying complex governance networks. 1. governance networks networks are composed from interdependent actors (kapucu, 2014) who can create various structures and serve wide range of purposes. in contrast to the policy networks whose focus is on decision-making, and the collaborative networks which emphasize working together in the service delivery process, the governance networks “combine aspects of policy making and service delivery” (lecy, mergel, & schmitz, 2014, p. 648). therefore, governance networks should be distinguished from collaborative networks, policy networks and public management networks due to their emphasis of “the governance process to achieve cross-sector and inter-organizational goals in the public sector” (kapucu, 2014, p. 30). the governance networks acknowledge that the public sector acting alone cannot solve complex problems and it has to work with other actors. these networks are characterized with interdependence among their members and dynamic relationships. importantly, the governance networks are also self-organizing and they challenge the paradigm of markets being the sole nonhierarchical efficient structures. 63037-15-29.indd 1 30/09/15 5:10 pm 2 n. kapucu / introduction: social network analysis applications in complex governance networks 2. complexity of governance networks in addition to the complex and wicked public policy problems, the governance networks could also take a complex structure. as the number of actors involved in the network rises, the complexity of the network increases. moreover, the complexity of the network also increases with the increased functions of the network and increased interdependence among the actors. in addition, existing uncertainties add to the complexity of the governance networks (kapucu, 2014). 3. social network analysis applications in the last decade the research on social network has tripled (borgatti, mehra, brass, & labianca, 2009) and the public administration research focusing on networks has exploded (kapucu, hu, & khosa, 2014; lecy et al., 2014). although in the 1980s the “social network analysis had become an established field within the social sciences “(borgatti et al., 2009, p. 893) a significant portion of the existing public administration literature on networks do not provide proper definition of the term network and has more of a conceptual and theoretical focus (kapucu et al., 2014). in that sense, methodologies used have been “remarkably static” (robinson, 2006, p. 596) and there is evident lack of empirically oriented research (kapucu, 2014). the use of “formal quantitative social network analysis techniques” (lecy et al., 2014, p. 654) and social network analysis (sna) have been limited (kapucu et al., 2014). the social network structure has multiple levels: “single nodes, dyadic ties, and whole networks” (carpenter, li, & jiang, 2012, p. 1336) and sna is sophisticated tool that permits examination of the formal and informal relationships between actors (nodes) within a network. in that light, sna is suitable for analysis of dynamic intraorganizational and inter-organizational relationships (kapucu et al., 2014), as well as measurement of “structural position power” (lecy et al., 2014, p. 644) and its dynamics (isett, mergel, leroux, mischen, & rethemeyer, 2011). social network analysis has been widely used in the field of national security as well as organized crime (borgatti et al., 2009). sna has also been utilized in the field of emergency management, education performance, regional economic development, environmental management, transportation policy, health and social service delivery, urban planning and nonprofit management (kapucu et al., 2014). papers included in this special issue, based on a call for papers and rigorous review processes, cover important theoretical and methodological issues in examining complex governance networks. i will provide brief overview of each paper in the following section. the first article, balancing the quantitative and qualitative aspects of social network analysis to study complex social systems by danny schipper and wouter spekkink, highlights the potential use of sna in investigating complex social systems. the article provides a balanced demonstration of the use of qualitative and quantitative sna approaches. 63037-15-29.indd 2 30/09/15 5:10 pm n. kapucu / introduction: social network analysis applications in complex governance networks 3 the authors claim that combination of qualitative and quantitative sna use is most useful for revealing system-level patterns, and a deeper understanding of the mechanisms that produce these patterns. the second article, network features and processes as determinants of organizational interaction during extreme events by michael d. siciliano & clayton wukich, attempts to statistically model the processes by which governance networks form. the authors explore a range of network features and processes and measure their influence on network formation by using the case of hurricane katrina employing exponential random graph models to identify the drivers of network formation in response to extreme events and disasters. the third article, measuring risks of organizational failure in contract exchange structures by julia l. carboni, uses affiliation network concepts from network analysis to develop a structural index of interdependent relationship between government and other stakeholders or actors in service delivery networks without joint service delivery. network analysis has been criticizers in terms linkages between networks and performance. this index developed in the article can be incorporated into statistical models to analyze contract performance of networked action. the fourth article, the shape of watershed governance: locating network boundaries within multiplex networks by steve scheinert, sarah coleman, christopher koliba, asim zia, & stephanie hurley, approaches governance networks as both nested and interconnected systems and identifies internal boundaries within these networks to differentiate between multiple functional subnetworks using a case study of lake champlain basin (watershed governance networks). the article compares structure of subnetworks and network macrostructure using the quadratic assignment procedure (qap). the fifth and final article, conducting content analysis of documents in network research: a review of recent scholarship by qian hu, examines how researchers have conducted content analysis of different documents including government documents and newspaper news reports to collect and analyze network data in public administration. the article provide useful hints to researchers to gather data from existing documents in conducting network research in analyzing complex intergovernmental and interorganizational research. 4. conclusion notwithstanding the increased number of research on governance networks, the methodological progress lags behind the needs of the field of public administration (kapucu, 2014). acknowledging the development of the network research in the public administration, future research could benefits from conceptual clarity (lecy et al., 2014), methodological rigor and empirical evidence. there is need for large sample size studies which would enable growth of the knowledge on complex governance networks (robinson, 2006). i hope that the special issue will address this important need and contribute to the study of complex governance networks. 63037-15-29.indd 3 30/09/15 5:10 pm 4 n. kapucu / introduction: social network analysis applications in complex governance networks references borgatti, s. p., mehra, a., brass, d. j., & labianca, g. (2009). network analysis in the social sciences. science, new series, 323(5916), 892–895. carpenter, m. a., li, m., & jiang, h. (2012). social network research in organizational contexts: a systematic review of methodological issues and choices. journal of management, 38(4), 1328–1361. isett, k. r., mergel, i. a., leroux, k., mischen, p. a., & rethemeyer, r. k. (2011). networks in public administration scholarship: understanding where we are and where we need to go. journal of public administration research and theory, 21(suppl. 1), i157–i173. kapucu, n. (2014). complexity, governance, and networks: perspectives from public administration. complexity, governance & networks, 1(1), 29–38. kapucu, n., hu, q., & khosa, s. (2014). the state of network research in public administration. administration & society. advance online publication. doi:10.1177/0095399714555752 lecy, j. d., mergel, i. a., & schmitz, h. p. (2014). networks in public administration: current scholarship in review. public management review, 16(5), 643–665. robinson, s. e. (2006). a decade of treating networks seriously. policy studies journal, 34(4), 589–598. 63037-15-29.indd 4 30/09/15 5:10 pm complexity, governance, and networks: introduction to the inaugural issue complexity, governance & networks (2014) 1–4 1 complexity, governance, and networks: introduction to the inaugural issue göktuğ morçöla, geert teismanb, and lasse gerritsb a pennsylvania state university at harrisburg b erasmus university rotterdam 1. yet another journal? why is it necessary to publish a new journal about complexity, governance, and networks? there has been an increasing interest in complexity, governance, and network studies in recent decades. a number of these studies have been published in most prestigious journals in public policy, public administration, and political science. also a few journals that specialize in these areas were launched in recent years. so, why this journal? and why now? the publication of this journal is not only a response to the surging demand for scholarly outlets that publish articles in these areas, but also a response to the need to create a venue to bring together the insights of complexity, governance, and network scholars. complexity, governance & networks will serve as an incubator for conceptual and theory development, methodological advancements in these three areas of study, empirical research, and idea exchanges among them for possible future syntheses. the articles that appear in this inaugural issue of the journal demonstrate how much the thinking and methods in these three areas have advanced. the authors, who are well-known experts in their areas, assess the current state of the advancements and possible directions for future research and conceptual development. 2. how this journal came about complexity, governance & networks has deep intellectual and organizational roots. network studies have a long history in the social sciences; their roots can be traced back to the 1930s. the concept of governance gained popularity more recently, but governance theorists and researchers have already made significant impacts on our understanding of policy and administration processes. complexity sciences have influenced thinking in the natural sciences and mathematics for at least since the 1970s and they were recognized by public policy and administration scholars as early as the 1990s. the authors of the papers published in this issue discuss the intellectual foundations that were laid down in these areas by many scholars. some of those scholars are the contributors to this issue of the journal. 153720_intro.indd 1 03/07/14 1:30 am 2 göktuğ morçöl et al. / complexity, governance, and networks the organizational roots of this journal can be found in the works of a group of scholars who organized panels on “chaos and complexity” at the public administration theory network (pat-net) conferences in the mid-1990s. as the interest in these panels began to wane at the pat-net conferences, it re-surfaced in other venues. an international workshop on complexity and policy analysis was organized by some of the initial members of the pat-net complexity/chaos group in cork, ireland, in 2005. meanwhile a few members of the american society for public administration (aspa) organized complexity panels at the annual conferences of this association and initiated a section on complexity studies (scns). aspa officially recognized the section in 2008. these developments in the united states coincided with the establishment of the research group governance of complex systems (gocs) at the erasmus university rotterdam in 2006. this research group includes scholars who are well-known for their contributions to governance and network studies and complexity sciences. the annual conference of aspa in march 2009 offered an opportunity for some of the members of the aspa sncs group and the gocs group to meet and discuss possibilities of jointly organizing conferences and starting a journal. the first two of these jointly organized conferences were held in rotterdam (compact i, 2011) and los angeles (compact ii, 2013). a third conference will take place in dublin, next year. the long journey of the scholars who organized panels and conferences at both sides of the atlantic and published numerous articles and books since the 1990s has culminated in the publication of complexity, governance & networks. this is only a milestone in a journey that will continue in many decades to come. we, the editors of the journal, would like to thank laurenz baltzer of baltzer science publishers for his support and ever-lasting effort to launch this journal. we intend to make complexity, governance & networks the primary reference for future researchers in these areas. 3. aims and scope of the journal our journal will contribute to philosophical, theoretical, methodological, and empirical developments in complexity, governance, and network studies in public policy, public administration, and politics, and non-governmental organizations. we welcome manuscripts written from a variety of theoretical perspectives (e.g. rational choice theory, game theory, advocacy coalition framework, marxist theory, critical theory, etc.). we welcome theoretical essays and empirical studies that employ quantitative methods (e.g. agent-based simulations, system dynamics, social network analyses), qualitative methods (e.g. single and multiple case studies, content analyses), and mixed methods (e.g. qualitative comparative analysis, process analysis). we also aim to establish a bridge between the worlds of academics and practitioners. therefore we welcome contributions by both. we intend to publish articles that are based on case studies focusing on examples of complex behaviors and network interactions in 153720_intro.indd 2 03/07/14 1:30 am göktuğ morçöl et al. / complexity, governance, and networks 3 public service settings, as well as reflective pieces on the practical values of complexity, governance, and network theories and methods. 4. papers in this issue the papers in this issue take stock of the state of the developments in the studies on complex governance networks. they also point to future directions in theory and research. morçöl makes the case for a complex governance networks conceptualization. after summarizing the contributions in the areas of governance, networks, and complexity studies, he argues that the concepts and methods in them should be synthesized. he also argues that the researchers of complex governance networks should incorporate the insights of well-established theories of policy processes, such as the institutional analysis and development framework and the advocacy coalition framework. teisman and gerrits argue that the complexity of an interconnected society and its governance requires a complexity-informed approach to our domain. using the concept of “emergence”, they articulate a non-decomposable, contingent, non-compressible and timeasymmetric nature of reality. subsequently, they propose three methods that take these aspects into account when putting concepts from the complexity sciences to the test: qualitative comparative analysis, dynamic network analysis, and group model building. kapucu argues that complex public policy problems require productive collaborations among different actors from multiple sectors. to better understand and guide such collaborations, researchers in public administration should use the concepts and methods of networks and network management. he then briefly discusses the research streams on complex networks, network governance, and current research challenges in public administration. mcguire and fyall address more specific issues in the studies on public management networks in contemporary governance. they argue that public management networks do not always possess the capacity to convert collective solutions to formal policy or program adjustments. government or other powerful agencies can often dominate the management of networks, elected officials may make policy decisions that are inconsistent with the recommended action of networks, and assessing the performance of networks is very often a moving target. the authors discuss the administrative and political barriers that can hinder the ability of public management networks to influence policy making and implementation. shrestha, berardo, and feiock’s paper illustrates the levels of sophistication in the conceptualizations and methods used in current network studies. the authors discuss the conceptual issues in applying uniplex network analyses in the studies on coordination and cooperation problems associated with institutional collective action. they make the case for multiplex network analyses (analyses of both the formal and informal relationships formed by policy actors) in studying the inherently complex relations in policy networks. then they propose multiplex versions of “bridging and bonding networks,” which were 153720_intro.indd 3 03/07/14 1:30 am 4 göktuğ morçöl et al. / complexity, governance, and networks found by other researchers to be suitable for solving coordination and cooperation problems in uniplex networks. the complexity of governance, policy, and public management networks is an undercurrent in all the papers in this issue of the journal. some of the papers address more specifically the issues in the conceptualizations of complexity. klijn and koppenjan, for example, discuss their conceptualization of complexity in governance network theory. they observe that there are three types of complexity in governance networks: substantive, strategic, and institutional complexity. they argue that dealing with these types of complexity in networks is essentially a matter of mutual adaption and cooperation. kiel, one of the earliest contributors to the applications of complexity sciences in public administration and policy, traces the evolution of these applications. using a fourstage evolutionary model, he tracks the developments in the pertinent literature. he argues that the evolution of the literature has now reached a “proliferant stage” and the concepts of complexity sciences are ripe for further and more advanced applications. the paper by zia, kauffman, koliba, beckage, vattay, and bomblies is an example of these advanced applications. the authors argue that due to the inherent uncertainty in predicting the evolution of phase-spaces in social-ecological systems (sess), these systems cannot be “optimally” managed through top-down, command and control type of governance designs. they propose a generalized autocatalytic set theory as an alternative. the authors suggest that a bottom-up, emergent and co-evolutionary framework should be used to design the governance regimes of sess. they also suggest that policy and institutional interventions can at best enable the policy-makers to nudge sess towards socially desirable yet ecologically feasible phase-spaces, which in turn are continually revamped as new elements in phase-spaces emerge. 5. invitation to publish the contributions to this issue are good indicators of the state of the advances in governance, networks, and complexity studies. they demonstrate that we have come long way, but we have an even longer way to go. we, the editors and the publisher, invite researchers and theorists around the world to join our journey by contributing to the journal with their most advanced thinking and most sophisticated empirical studies. 153720_intro.indd 4 03/07/14 1:30 am editorial : introduction to the special issue: complexity and time 1 university of bamberg press introduction to the special issue: complexity and time authors: phil haynes and stefan verweij school of humanities and social science, university of brighton, united kingdom; department of spatial planning and environment, university of groningen, the netherlands email: p.haynes@brighton.ac.uk time is a dimension that offers the modelling of complex social systems—or the analysis of these systems in general—a degree of predictability and stability. time is irreversible and causality is emergent, in the sense of the “impossibility of undoing the past” (byrne & callaghan, 2014, p. 114). it structures and organizes the analysis of social systems and therefore our understanding of the social processes analyzed (adam, 1990). in that capacity, time is “external to and abstracted from the processes it measures” (adam et al., 2002, p. 12). researchers conceptualize and measure it in different ways. one distinction is between duration, pace or tempo, trajectory or sequence, and cycles (aminzade, 1992). duration concerns the amount of time that has elapsed for a certain event or sequence of events. this conceptualization of time requires the definition of a start and endpoint of the (sequence of) event(s), and it therefore assumes diachronic emergence (teisman & gerrits, 2014). pace or tempo indicates the rate or speed of movement or progress, i.e., the density of events in a certain amount of time. it might also imply the repetition of events within a certain timespan (aminzade, 1992). trajectory or sequence is akin to abbott’s (1992) idea that processes are structured by events through time and it refers to “the sequential order of events” (aminzade, 1992, p. 462). this is also the concept of time that, for example, pierson (2004) focuses on when talking about path-dependent processes, and it is concerned more with patterns or directions of events. trajectory refers to cumulative events, and researchers may be interested in analyzing the (types of) sequences leading to an outcome of interest (e.g., spekkink, 2016). cycles, finally, are concerned with the repetition of sequences of events (rather than with cumulative events), marked by regularly recurring ascending and descending phases (aminzade, 1992). trajectory and cycles allow for ambiguity about start and endpoints and can therefore be related to synchronic emergence (teisman & gerrits, 2014), i.e., identifiable patterns of emergence. time plays a crucial role in the study of complex social systems (byrne & callaghan, 2014), for instance through the analysis and application of notions such as emergence, self-organization, feedback, co-evolution, and non-linearity (e.g., teisman et al., 2009; de roo et al., 2020). as said, time often features as a structuring principle—e.g., as duration or pace in agent-based models or in the qualitative descriptions of cases over time, where cases are described by trajectories, phases, rounds, or cycles—but it does not take the center stage as a concept in itself. with this special issue, we want to draw attention to time in the analysis of complex social systems. before we provide a short overview of the articles in this special issue and highlight the contributions they make to time in the analysis of complex social systems, we wish to draw attention to some recent methodological advancements that seek to infuse the complexity sciences’ toolkit with more time-sensitive methods. we do not focus here on agent-based modelling—which can be appropriately used to study the emergence (diachronic) of patterns and outcomes, exploring and developing a better understanding of how complex social processes develop and the policies that could influence them and how (e.g., campbell et al., 2015)—but instead want to raise attention to two recent advancements that have received less attention. what unifies these two approaches is an interest with the complexity of cases, as a unit of analysis, rather than a structural method based on aggregate variable measurements. while the cases researched could be human participants, they are also often at the level of comparable policy organizations or even nation states. one aspect of the importance of case diversity in complex social systems, is understanding the trajectory of these cases and their differences and similarities over time. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 1-4 doi: http://dx.doi.org/10.20377/cgn120 mailto:p.haynes@brighton.ac.uk 2 university of bamberg press a first approach concerns qualitative comparative analysis (qca). this is a case-based, comparative, and configurational method, first introduced by charles ragin (1987). since the beginning of the current century, it has been picked up by complexity scholars as a promising method to analyze complex social systems, in particular because of its premise of complex causality (byrne, 2005, 2009; byrne & callaghan, 2014; gerrits & verweij, 2018). a critique on qca has been that its cross-sectional nature diminishes the role of time; cases are frozen in time and then compared to identify complex configurations of conditions that may explain a certain outcome of interest at a certain time (de meur et al., 2009). this critique has led scholars to explore ways to integrate time in qca much more prominently (e.g., caren & panofsky, 2005; hino, 2009; pagliarin & gerrits, 2020; verweij & vis, 2021). although such more timesensitive strategies have yet to prove their merits because their applications are few, they may prove valuable in making qca not only appropriate for the study of static complexity but also of dynamic complexity (verweij & zuidema, 2020). in this special issue, lasse gerrits, robin a. chang, and sofia pagliarin take further the conceptualization of time in case-based methods. a second advancement we wish to give attention to is dynamic pattern synthesis (dps) (haynes, 2017). this is a recent development of the longstanding quantitative and case-based method of cluster analysis. in dps, cluster analysis is adapted and expanded to consider configurational case patterns and trajectories over time. david alemna uses a version of dps in this special issue to compare policy responses in a group of west african states. what then do the articles in this special issue further contribute to the discussion and advancement of time in the analysis of complex social systems? the articles share the discussion and/or application of methodological developments in policy research and their attachment to the metatheoretical framework of complex systems. several articles included here in the special issue—the article by jean boulton and the article by graham room—look at the intersection of such “grand theory” with interdisciplinary policy analysis and consider the importance of identifying patterns across time in an applied and interdisciplinary sense, along with what such patterns might mean for governance leadership and decision-making. jean boulton develops the ontology of “process complexity” to raise attention to the world as complexity-in-process (ever-emerging, becoming). building on the work of prigogine on irreversibility and the thermodynamics of open systems, and by showing the ontological resonances with quantum gravity, daoism, and thinkers such as bergson, whitehead, and james, she argues the case for processes taking center stage in the study of complexity. boulton asserts the need to acknowledge “ontological uncertainty.” policy makers have numerous potential policy futures in front of them. events become more important than things. discerning “patterns of relationships” is argued to be of seminal importance. policies need to be agile and aware of the systemic influences. in a world of “ontological uncertainty,” action is as much informed by ethical choices as it is by “evidence.” graham room has a more policy-conceptual aim and endeavors to marry the policy science literature with the complexity science literature by focusing on “complex timescapes.” timescapes stress the intricate connection between time and complex landscapes. his work is influenced substantially by kauffman’s (1995) scientific concept of “fitness landscapes.” room argues there are “critical junctures” where shifts in the landscape encourage decisive policy action. at some points in time, there is a greater chance of making an influence. he illustrates from political and policy analysis literature how previous seminal thinkers have come close to such a complex policy “metatheory” before, each with their own original concepts and narrative. he is particularly influenced by hirschman’s ideas of “interlocking vicious circles” that prevent development, and “upward spirals” that can bring together opportunities in time when the necessary resources for change are aligned. his analysis is ironically “timely” given the formidable challenges of the covid-19 pandemic and climate change that are knocking at the door of politicians and policy makers the world over. graham room, like jean boulton, also notes the ethical requirement to attempt subjective action, given the never-ending quest for a complexity science of explanation in a world where much will remain unknowable. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 1-4 doi: http://dx.doi.org/10.20377/cgn120 3 university of bamberg press the article by lasse gerrits, robin a. chang, and sofia pagliarin and the article by david alemna are more concerned with the methodological development and empirical application of time within the meta-theoretical framework of complex systems. lasse gerrits and colleagues emphasize the different pacing of events and time pressures within diverse policy circumstances. they stress the importance of “following the dynamics of how time passes” in cases. this leads to the development of the approach of “temporal casing.” this takes further the importance of case-based methods for applied social scientists motivated by complexity theory. “within case variation” becomes central to this new conceptualization and an extension of the case-based method. instead of adopting a priori a certain convention of time in terms of duration, pace, sequence, or cycle (aminzade, 1992; see above)—and some length, speed, or order of events or period—a core point is to take the “developmental stages” of cases as the dictating logic for the reconstruction of cases. their article makes the argument for how a temporal dimension can be an essential and integrated addition to casebased policy research. the authors set out the different steps involved in temporal casing, with applied examples given from previous policy megaprojects research. david alemna’s quantitative study develops case-based research that can capture the political and economic complexity of continental governance and its high dynamics. he adopts the ‘year’ as a conventional measurement of time as duration, and then endeavors to identify stages or phases in countries’ developments characterized by specific configurations. using the method of dps, he illustrates the formidable challenges for the policy aim of achieving economic stability through currency convergence in west africa. because of instabilities, both within the member countries and in the global market external to them, he demonstrates how a wide range of adaptations and interventions are needed if any small policy progress is to be made. the data analysis illustrates the relational complexity in pan-west african governance. there is much diversity in the experience of each country, with only small subgroups of countries forming similar patterns over time. different subgroups and countries have contrasting experiences of convergence. this special issue makes an important contribution to the further integration of complexity science into policy and governance studies. three words stand out more than others across these four articles: cases, dynamics, and patterns. several of the articles take forward the use of case-based methods as a central part of a research strategy to better understand complexity. as a result, the diversity of cases and their experience are amplified, rather than the aggregation of simplified variable relationships. similarly, the application of policy analysis using the complexity framework is increasingly leading to an emphasis on the unstable dynamics of the policy world with its often unpredictable and unknowable futures. nevertheless, the contributing scholars and policy analysts in this special issue illustrate a continuing determination to find appropriate concepts and models, that while respecting the humility required by a complex world, can still uncover the occasional temporal patterns of stability and meaning that will aid policy makers to have the best chance to make favorable decisions going forward. references abbott, a. (1992). from causes to events: notes on narrative positivism. sociological methods & research, 20(4), 428–455. adam, b. (1990). time and social theory. polity press. adam, b., whipp, r., & sabelis, i. (2002). choreographing time and management: traditions, developments, and opportunities. in r. whipp, b. adam, & i. sabelis (red.), making time: time and management in modern organizations (pp. 1–28). oxford university press. aminzade, r. (1992). historical sociology and time. sociological methods & research, 20(4), 456–480. byrne, d. s. (2005). complexity, configurations and cases. theory, culture & society, 22(5), 95–111. byrne, d. s. (2009). complex realist and configurational approaches to cases: a radical synthesis. in d. s. byrne & c. c. ragin (red.), the sage handbook of case-based methods (pp. 101–119). sage. byrne, d. s., & callaghan, g. (2014). complexity theory and the social sciences: the state of the art. routledge. campbell, h. e., kim, y., & eckerd, a. (2015). rethinking environmental justice in sustainable cities: insights from agent-based modeling. routledge. caren, n., & panofsky, a. (2005). tqca: a technique for adding temporality to qualitative comparative analysis. sociological complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 1-4 doi: http://dx.doi.org/10.20377/cgn120 4 university of bamberg press methods & research, 34(2), 147–172. de meur, g., rihoux, b., & yamasaki, s. (2009). addressing the critiques of qca. in b. rihoux & c. c. ragin (red.), configurational comparative methods: qualitative comparative analysis (qca) and related techniques (pp. 147–165). sage. de roo, g., yamu, c., & zuidema, c. (red.). (2020). handbook on planning and complexity. edward elgar. gerrits, l. m., & verweij, s. (2018). the evaluation of complex infrastructure projects: a guide to qualitative comparative analysis. edward elgar. haynes, p. (2017). social synthesis: finding dynamic patterns in complex social systems. routledge. hino, a. (2009). time-series qca: studying temporal change through boolean analysis. sociological theory and methods, 24(2), 247–265. kauffman, s. (1995) at home in the universe: the search for the laws of complexity. viking. pagliarin, s., & gerrits, l. m. (2020). trajectory-based qualitative comparative analysis: accounting for case-based time dynamics. methodological innovations, 13(3), 1–11. pierson, p. (2004). politics in time: history, institutions, and social analysis. princeton university press. ragin, c. c. (1987). the comparative method: moving beyond qualitative and quantitative strategies. university of california press. spekkink, w. a. h. (2016). industrial symbiosis as a social process: developing theory and methods for the longitudinal investigation of social dynamics in the emergence and development of industrial symbiosis. erasmus university rotterdam. teisman, g. r., & gerrits, l. m. (2014). the emergence of complexity in the art and science of governance. complexity, governance & networks, 1(1), 17–27. teisman, g. r., van buuren, a., & gerrits, l. m. (red.). (2009). managing complex governance systems: dynamics, selforganization and coevolution in public investments. routledge. verweij, s., & vis, b. (2021). three strategies to track configurations over time with qualitative comparative analysis. european political science review, 13(1), 95–111. verweij, s. & zuidema, c. (2020). qualitative comparative analysis for analyzing spatial planning processes. in g. de roo, c. zuidema & c. yamu (red.), handbook on planning and complexity (pp. 334-352). edward elgar. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 1-4 doi: http://dx.doi.org/10.20377/cgn120 informed governance: complexity and the commons complexity, governance & networks (2014) 1–8 1 doi: 10.7564/14-cgn15 informed governance: complexity and the commons jack meeka,* and kevin marshallb auniversity of la verne (usa) e-mail: jmeek@laverne.edu buniversity of la verne (usa) e-mail: kmarshall@laverne.edu jack w. meek, ph.d. is a professor of public administration, fellow of the university of la verne academy and director of the master of public administration program. his research focuses on metropolitan governance including the emergence of administrative connections and relationships in local government, regional collaboration and partnerships, policy networks and citizen engagement. jack co-authored governance networks in public administration and public policy (crc press, 2011) and co-edited networked governance: the future of intergovernmental management (cq press, 2012) and business improvement districts: research, theories, and controversies (2008). jack has also edited symposia on complexity theory in emergence: complexity and organization (2010) and co-edited a symposium on integrity in public administration in public integrity (2012). kevin s. marshall, ph.d., j.d., mpa, ba, is a professor of law, economics and finance at the university of la verne and is a fellow of the university of la verne academy. his teaching and research interests include law, economics, finance, corporate finance, public policy, nonprofit governance, as well as quantitative and qualitative research techniques. kevin has published and presented numerous books and articles involving the interdisciplinary workings of law and economics. his work has been cited by both state and federal appellate courts, as well as by numerous other published works and secondary sources. 1. acknowledgement we wish to express our appreciation to all who contributed in the production of this volume, including all those who served as reviewers whose careful and insightful reviews enhanced and improved the articles presented at the symposium and in this volume. we  also want to express our appreciation to dean abe helou, college of business and public management, university of la verne, for his continuing support for continuous dialogue and research regarding the potential of complexity theory and approaches for improving governance. * corresponding author. article_14-15.indd 1 04/03/15 4:42 pm 2 j. meek et al. / informed governance 2. informed governance: complexity and the commons governance challenges are persistent. persistence may mean that meaningful solutions are not feasible within the current frame of thinking or with the current approach or tool we are employing. different lenses – new frames of reference, wider viewpoints, inclusion, – may offer new opportunities of examination of essential patterns of interaction that need attention. as indicated with the articles of this issue, complexity theory offers a different framework for addressing wicked, persistent problems found in highly interdependence spatial domains – the commons. the relevancy of complexity theory continues to reveal itself as academicians and practitioners continue to explore its essence and insights. while complexity theory’s origin is in the physical sciences, it is not new to the social sciences as demonstrated by observations dating back to aristotle (morcol, 2014;), as well as by the more recent observations of the likes of smith, ferguson, von mises, hayek, and ostrom (marshall, 2014; morcol, 2014). renewed attention to complexity theory is rapidly transcending and transforming disciplinary boundaries, establishing cross-boundary linkages between and within the differentiated landscapes of all the physical and social sciences. and perhaps most importantly, at least with respect to the mission of this publication, it is triggering a paradigm shift with respect to exploring, critiquing and designing social system constructs as demonstrated by the following articles. a challenge of the administrative sciences lies in coordinating competing interests in the highly interdependent space of the commons. exploring this challenge and developing coordination and allocation strategies with respect to the many and varied stakeholders and constituents is one of the central interests of policy and administration research. it is central to policy orientation (morcol, 2012), as well as to the management of complex systems and policy (teisman, van buuren, & gerrits, 2009; tait & richardson, 2011). with this collection of essays, we continue this exploration and include both theoretical assessments and empirical interpretations of patterns found within the contested space of the commons. we face enormous challenges in managing the commons, especially in the age of globalization and international linkages. the research presented here is importantly relevant to these challenges. the complexity of the global-sphere is compounded by an ever expanding global commons with an ever expanding need to coordinate and administer the many competing and interdependent interests within multi-dimensional and multijurisdictional spaces. the challenges in water management, in preserving environmental quality, in protecting public health, in resource management, in maintaining institutional integrity is confounded by global-transcendence. “global-transcendence” is a term we use to capture the essence of the emergent properties of the global-sphere’s highly interdependent public spaces in which collective actions and strategies at the local, regional, national or transnational levels can, and often, transcend the dimensional and jurisdictional origins of a specific collective initiative. as globalization and the global-sphere continue to manifest and reveal itself so too does the complexity of the commons. it is this complexity that imposes challenging limitations article_14-15.indd 2 04/03/15 4:42 pm j. meek et al. / informed governance 3 on public administration and the managerial sciences ability to address and implement collective actions and strategies. and it is this complexity that renders the following articles significant and relevant to the administration of governance and networks. in order to construct solutions and strategies for administrating and managing the commons it is important that we recognize the following principles: ● we articulate the nature of the challenge and the role of institutions (ostrom, 1990; young, 2010). ● we recognize conflicting stakeholder interests can be placed along different scales – individual, group and system-wide – at the same time (koliba, meek, & zia, 2011); ● we acknowledge the many different types of actors that influence governance patterns, including non-human actors (gerrits, 2012). ● we recognize that interaction among stakeholders in contested arenas is dynamic and full of uncertainties (koppenjan & klijn, 2004); and ● we seek to understand how our institutions behave within contested arenas. by embracing the above, we are more able to understand patterns of interaction that arise from the many institutions designed to respond to human and environmental interactions within the commons. all of the essays that follow provided insight to the above principles, as well as illustrate their relevance with respect to understanding of and administering to the many contested issues confronted within the commons. complexity theory is utilized to inform the administration of contested public ‘commons’ in terms of allocation strategies, the rules regulating open space, simulated institutional patterns, system intersection analysis, co-evolution of metropolitan space, isomorphic properties of network governance and contributions to self-organization in the theory of collective action. in assessing governance concerns, applications of complexity theory offers opportunities to draw upon differentiated methodologies as a way of conceptualizing the commons and to visualize and examine patterns not previously considered and perhaps not previously observed. the papers collected for this issue each seek to address governance through the application of differentiated methods – economic philosophy, law, simulation, co-evolution, intersection analysis and network analysis. as researchers, we continue to refine our lens in the examination of interdependencies and intersections in search of patterns that inform governance within spatial areas. we seek to understand complex systems. in doing so, the papers presented in this issue demonstrate: 1. current complexity research with previous intellectual efforts; 2. complexity-informed rules that create and nurture innovative capacities and outcomes in the commons; 3. complex behavior of hierarchies under conditions of entropy; 4. complex patterns of metropolitan evolution and dissipation; 5. complex intersection of the delta region and human interests; and 6. isomorphic patterns among watershed networks article_14-15.indd 3 04/03/15 4:42 pm 4 j. meek et al. / informed governance five of the six papers for this volume were presented at the second conference on ‘the challenge of making public administration and complexity theory work’ held at the university of la verne (california) in june 2013. this conference was sponsored by the section for complexity and network studies (scns) of the american society of public administration (aspa), the governance of complex systems group of erasmus university, and the college of business and public management of the university of la  zverne. the sixth paper was selected from a privious aspa/scns conference. all of the papers were subjected to and selected as a result of a two-staged, peer-reviewed process. this selection of papers continues efforts of the scns section of aspa in publishing works presented at national meetings (meek, 2010) and conferences (gerrits & marks, 2012). what follows is a summary of the complexity–related governance themes and papers outlined for this issue of complexity, governance and networks. 2.1. self-organization – governance of the commons we begin this collection of papers with the examination of a central concept in complexity theory – self-organization. goktug morcol, in his paper “self-organization in collective action: elinor ostrom’s contributions and complexity theory” examines the value of connecting current research on self-organization with previous theoretical efforts. morcol notes that the concept of self-organization has deep historical roots and is a central concept in complexity theory as well as in theories of policy processes. for morcol, the work of ostrom instructs our current work on self-organization in how institutions of collective action – including hierarchical structures – emerge through self-organization and how the nature of the environmental characteristics – structured environments – influence collective action processes. 2.2. complexity informed rules for the commons – innovative space kevin marshall in “creating and maintaining innovative space—a framework for unraveling the complexities of entrepreneurial systems” develops a framework, informed by complexity theory, establishing the conditions necessary for the creation and maintenance of innovative spaces. the role of complexity theory, marshall posits, is central to developing public arenas wherein ideas are allowed to develop, evolve and mature. marshall builds and expands on the principle that the structure of the environment or space influences the emergence of self-organization, as well as influences the collective action processes. such spatial structure is chiseled from the many rules implemented within the space, and thus we must be ever so cognizant of the rules that define and form the space. marshall argues that differentiation and integration are essential components of an adaptive, progressive space, and that the administrative and managerial rules implemented and administered within the space not only define the parameters of the space in question, but also influence the innovative, progressive nature of the space. article_14-15.indd 4 04/03/15 4:42 pm j. meek et al. / informed governance 5 2.3. entropy – hierarchical behavior in the commons “defining the entropy of hierarchical organizations,” by david chappell and greg dewey, seeks to understand the entropy of organizational structure given personnel shifts within units and under conditions when organizational units are expanded or contracted. they propose a measure of order within a hierarchical organizational structure based on an analogy drawn from thermodynamic entropy of physical sciences. using computer simulations of hierarchical structures, they reveal the boundaries of both vertical and horizontal entropies and their influence with respect to aggregate, organizational entropy patterns and that each organization is likely to have unique entropic patterns that influence organizational functions. (chappell & dewey, 2014). 2.4. coevolution analysis – metropolitan governance over time in the commons danny schipper and lasse gerrits, in “the emergence of metropolitan governance “draw upon a co-evolutionary framework to analyze governance patterns constructed in the city of amsterdam over almost three decades. schipper and gerrits argue that, while institutional design and network approaches each have their theoretical merits, they fail “to provide an explanation for the highly metropolitan governance arrangements that have emerged, with within countries.” (schipper & gerrits, 2014.) they further argue that “a  co-evolutionary analysis of metropolitan governance is necessary to understand the emergence of metropolitan governance systems as a complex process of reciprocal selection.” (schipper & gerrits, 2014). through the lens of interpreting environmental conditions and selection mechanisms, their paper interprets policy documents and newspaper accounts during the time in question to assess the development of governance strategies. the authors find that dynamic environments challenge the fitness of existing structures and, as a result, metropolitan governance evolves in a punctuated fashion. 2.5. intersection analysis – delta governance in the commons delta areas are commons areas full of contending actors and interests, including those from agriculture, industry, shipping and energy. appropriate concerns are raised for delta areas faced with flood risks and ecological decline. delta areas are also full of overlapping jurisdictional interests challenging the ability to coordinate competing interests. the work of bonno pel, jitske van popering-verkerk, arwin van buuren, & jurian edelenbos, “intersections in delta development; analyzing actors for complexity-sensitive spatial concepts” seeks to develop an integrative solution to delta governance. in the examination of the dutch southwest delta, the authors conduct stakeholder analysis to assess differentiated interests and co-evolving properties within the system. the authors develop subsystem configurations based on a ‘synthesis of intersections’ among stakeholders for governance consideration. in this work, the authors demonstrate how a complex adaptive system framework can inform the development of ‘complexity-sensitive spatial concepts.’ article_14-15.indd 5 04/03/15 4:42 pm 6 j. meek et al. / informed governance 2.6. isomorphic properties – watershed governance in the commons “isomorphic properties of network governance: comparing two watershed governance initiatives in the lake champlain basin using institutional network analysis” by christopher koliba, adam reynolds, asim zia, steven scheinert examines two governance networks – one regulatory and the other a collaborative partnership network – established in watershed planning documents for the lake champlain basin. institutional network analysis was conducted, measuring network centrality, structures and clusters and a statistical comparison of the task structures of the two networks. the authors examined the degree these two planning networks – in structure and function – reify themselves in the plans they create. the authors used institutional isomorphism theory “to anticipate and explain any mirroring effects observed in the data.” (koliba, reynolds, zia, & scheinert, 2014). the authors found that, while isomorphic tendencies were found in structural design, these tendencies did not appear in policy tool selection. the latter outcome lead the authors to offer pattern outcomes attributed to different kinds of isomorphism. 2.7. summary these papers extend the dialogue the governance utilizing complexity and network theory while drawing upon a diversity of methodological approaches in exploring complex systems. through the lens of complexity, these papers inform the art of collaboration in developing forms of collective action, innovation spaces; these papers inform our understanding of institutional evolution, design and adaptive institutional properties. one promising tenant of complexity theory may be that the examination allows for a wide variety of viewpoints to be included. this approach likely enhances the prospect of better solutions. references chappell, d., & dewey, g. (2014). defining the entropy of hierarchical organizations. complexity, governance and networks. gerrits, l.m. (2012). punching clouds: an introduction to the complexity of public decision-making. litchfield park, az: isce publishing. gerrits, l.m., & marks, p. (eds.). (2012). public administration in complexity. litchfield park, az: isce publishing. göktuğ morçöl, lorlene hoyt, jack w. meek and ulf zimmerman, eds. (2008) business improvement districts: research, theories, and controversies. public administration and public policy series, auerbach publications of taylor & francis. hayek, f. a. (1960). the constitution of liberty. chicago, il: university of chicago press. hayek, f. a. (1973). law, legislation and liberty, volume 1: rules and order. chicago, il: university of chicago press. hayek, f. a. (1976). law, legislation and liberty, volume 2: the mirage of social justice. chicago, il: university of chicago press. meek, j. w. and kurt thurmaier, eds. (2012) networked governance: the future of intergovernmental management. cq press. isbn 9781452203256 meek, j. w. guest editor (2010) symposium complexity theory for public administration and policy, vol. 12, issue 1, for the journal emergence: complexity and organization. article_14-15.indd 6 04/03/15 4:42 pm j. meek et al. / informed governance 7 keil, d. l. (1994). managing chaos and complexity in government: a new paradigm for managing change, innovation, and renewal. chichester, england: wiley. kickert, w. j. m., klijn, e.-h., & koppenjan, j. (1997). managing complex networks: strategies for the public sector. thousand oaks, ca: sage. koliba, c., meek, j. w., & zia, a. (2011). governance networks in public administration and public policy. new york, ny: crc press. koliba, c., reynolds a., zia, a., & scheinert, s. (2014). isomorphic properties of network governance: comparing two watershed governance initiatives in the lake champlain basin using institutional network analysis. complexity, governance and networks. koppenjan, j. f. m., & klijn, e. h. (2004). managing uncertainties in networks: a network approach to problem solving and decision making. london, england: routledge. marshall, k. s. (2014). creating and maintaining innovative space—a framework for unraveling the complexities of entrepreneurial systems. complexity, governance and networks. matthew t. witt, jack w. meek and suzanne beaumaster, co-editors (2012), “symposium: integrity in public administration,” journal of public integrity, vol. 14, no. 3, summer 2012, pp. 225–298. meek, j. w. (2010). complexity theory for public administration and policy. an international transdisciplinary journal of complex social systems, 12(1): 1–4. mises, l. v. (2007). introduction: economics and praxeology (vol. 1). in b. b. greaves (ed.), human action, a treatise on economics (vols. 1–4). indianapolis, in: liberty fund. (reprinted from foundations for economic education, 4th rev. ed., 1996, new york, ny: irvington-on-hudson. originally published 1949) göktuğ morçöl (2012). complexity theory for public policy. london, england: routledge. göktuğ morçöl (2014). self-organization in collective action: elinor ostrom’s contributions and complexity theory. complexity, governance and networks. ostrom, e. (1990). governing the commons: the evolution of institutions for collective action. cambridge, england: cambridge university. pel, b., van buuren, a., & edelenbos, j. (2014). intersections in delta development: analyzing actors for complexity-sensitive spatial design. complexity, governance & networks. schipper, d., & gerrits, l.m. (2014). the emergence of metropolitan governance. complexity, governance & complexity. smith, a. (1910). the wealth of nations. london, england: j. m. dent (originally published 1776) smith, a. (1982). lectures on jurisprudence (report dated 1766). indianapolis, in: liberty fund. (reprinted from the glasgow edition of the works and correspondence of adam smith vols. 1–5, by r. l. meek, d. d. raphael, & p. steinet, eds., 1978, oxford, england: clarendon press) tait, a., & richardson, k. a. (eds.). (2011). moving forward with complexity: proceedings of the 1st international workshop on complex systems thinking and real world applications. litchfield park, az: isce publishing. tait, a., & richardson, k. a. (eds.). (2011). moving forward with complexity: proceedings of the 1st international workshop on complex systems thinking and real world applications. litchfield park, az: isce publishing. teisman, g., van buuren, a., & gerrits, l. m. (eds.). (2009). managing complex governance systems: dynamics, self-organization and coevolution in public investments. london, england: routledge. young, o. r. (2010). institutional dynamics: emergent patterns in international environmental governance. cambridge, ma: mit press. article_14-15.indd 7 04/03/15 4:42 pm article_14-15.indd 8 04/03/15 4:42 pm complexity theory and its evolution in public administration and policy studies complexity, governance & networks (2014) 71–78 71 doi: 10.7564/14-cgn9 complexity theory and its evolution in public administration and policy studies l. douglas kiel university of texas at dallas, school of economic, political and policy sciences 800 w. campbell road, richardson, tx 75080 usa, 972-883-2019 e-mail: dkiel@utdallas.edu this paper traces the evolution of the application of the complexity sciences in the literature of public administration and public policy. a four stage evolutionary model is used to track the development of this literature. the evolution of the literature has now reached a third evolutionary stage. this third stage is a proliferant stage in the development of the literature in which applications and knowledge production has increased dramatically. keywords: complexity sciences, evolution, emergence, convergence, proliference, divergence, public administration and policy studies. 1. introduction we are fortunate to live in a historical period of both rapid and substantial change. this is a period in which human knowledge accumulates at rates that exacerbate efforts to accommodate and store this knowledge. keeping up with the accumulated knowledge requires a singular focus that is rare in a world of continuous partial attention. the acceleration of knowledge as a global resource is also reflected in the acceleration of knowledge regarding complexity studies in public administration and public policy. the period of early research in this area has evolved, over the last quarter century, into a period of proliferation in which scholars have applied an increasing array of methods to examine the behaviors of complex governance, administrative, and policy systems. in this paper i strive to explore the evolution of complexity studies in public administration and public policy. in particular, i view the knowledge accumulation in this area as an evolutionary process. tracing the evolution of what and how we know does not reduce the uncertainties, errors, or omissions in knowledge production processes, but it does provide a means of exploring the field of complexity studies as an evolving enterprise. this paper is comprised of four sections. first, i present schumacher’s (1986) evolutionary model as a means for appreciating the birth and maturation of complexity studies in public administration and policy. next, the literature of the complexity sciences in public administration and public policy is placed within the evolutionary sequence of article_14-9.indd 71 26-06-2014 17:32:40 72 l. douglas kiel / complexity theory and its evolution schumacher’s model. this analysis provides means for assessing the relative maturity of this academic enterprise and for assessing its current evolutionary stage. any review of the current state of an academic field requires some sense of the evolution of that field. complexity studies in public administration can be traced back approximately a quarter of century, to kiel’s early studies (e.g., kiel, 1989, 1993, 1994). the sources cited in this paper are a small sample of the growing and evolving literature on the applications of complexity theory in public administration and policy. the realization of this very journal, complexity, governance & networks, clearly represents the evolving maturity of the field. how can we best understand the evolution of the applications of complexity theory in public administration and policy? the scholarly literature provides some frameworks for understanding the evolution of an academic field of study. kuhn (1970) suggests that scientific paradigms are challenged from within themselves, through a hegelian conflict of ideas, which first instigates incremental change, but at one point the entire paradigm shifts and a new paradigm achieves dominance. if complexity theorists in public administration and policy seek dominance, it is likely to be a mild dominance based on a few important insights, rather than an effort to alter the entire landscape of public administration and policy studies. important insights such as the pervasive nature of nonlinear and dynamic phenomena in our field do not necessitate a revolution. these insights do suggest, however, that the simplifying requirements of the dominant linear model may fail to capture essential parts of the puzzle that constitute the reality of public administration and policy. if complexity studies are seen as an extension of established open systems models then complexity is simply an enhancement to existing knowledge, rather than a seismic shift in world views. however, the recognition that the truly interesting elements of reality are nonlinear and non-gaussian implies a more general critique of the value of using standard statistical techniques to understand complex dynamic phenomena. theories of innovation may also help to add insight into the evolution of complexity studies in public administration and policy. for example, roger’s (1983) oft-cited model of innovation reveals a standard sequence of behavior but also reflects a flattened normal distribution as a majority of followers allow early adopters to take on the risk of innovation. while such dynamics are likely to occur, roger’s model does not inform us of the continued process of the evolution of a scientific or technological innovation. an evolutionary model of change may, however, provide insights into the change in an academic discipline. the model of evolutionary change offered by schumacher (1986) may provide some insight into the evolution of complexity studies in public administration and policy. schumacher’s (1986) model of evolutionary change provides a means for assessing the relative maturity of the theory and methods used to study complex systems in public administration/policy. schumacher’s model includes a four-stage process that is intended to describe evolutionary processes in all living systems, including human sociotechnical systems. these four stages progress in a circular and sequential fashion. these four sequential stages are emergence, convergence, proliferance, and divergence. article_14-9.indd 72 26-06-2014 17:32:40 l. douglas kiel / complexity theory and its evolution 73 the first stage in schumacher’s model is emergence. emergence refers to the creation or production of a “new state of matter” (p. 10). such new matter, representing novelty, evolves as attractive forces “experiment” with environmental conditions to determine adaptive fit. emergent phenomena may continue to evolve along uncertain pathways or may die out if the emergent “mutation” cannot survive in its environment. high levels of risk exist for emergent entities and survival is not ensured. as an emergent entity evolves, forces of repulsion and attraction serve to determine whether the entity will die or converge into a coherent form capable of further reproduction. this second evolutionary stage is labeled convergence. at this stage of evolution, there is coherence with the mutation now being clearly distinct in its environment. a critical mass of the entity now exists providing the chance of further replication. the converged entity, now with some solid grounding in its environment, may reach a stage of environmental fit in which it proliferates. the entity thus matures to a level that it fills multiple niches in its environment and may dominate its environment. this third evolutionary stage, a period of proliferance, allows the possibility for increased system complexity that may generate novelty and eventually new forms of the entity. proliferance expedites complexity and further allows recombinations that may produce further complexity. a fourth state in this evolutionary paradigm is divergence. the divergent stage evidences novel forms of the proliferant entity as it seeks new forms of adaptive fit. divergent forms may die out with high frequency given the dominance of the previous form. eventually, though, a divergent form creates adequate attraction so that the cycle continues and a new emergent form of matter is produced. an example of this evolutionary model using technological innovation will serve to emphasize the relevance of this model. consider the case of the evolution of modern computing technology in the developed world. the emergence of the mainframe (vacuum tube technology) in the late 1930’s resulted in the convergence of the technology and its organizational applications in the 1950s–1970s. the rise of the microcomputer and networks engendered a proliferant era in which computers proliferated in the environment. a new divergent stage has emerged with wearable computing and sensors that may lead to new yet unknown emergent technologies. as a further example of general applicability of this evolutionary model, let us examine its relevance to human cultural innovation. consider the behavioral revolution in the social sciences as an evolving phenomenon. many of the statistical tools supporting this “revolution” emerged in the early 1900’s (ziliak & mccloskey, 2008). the post-world war ii era showed general convergence, as more social science disciplines relied on statistical analyses for scientific understanding. the proliferant period of the 1980s to the present shows an expanding array of statistical tools and techniques available to the social scientists. predicting the forthcoming divergent stage may not be wise in a nonlinear world, but such a stage may include elements of capturing big data on social behavior combined with individual level means for assessing behavior and health (pentland, 2008). article_14-9.indd 73 26-06-2014 17:32:40 74 l. douglas kiel / complexity theory and its evolution 2. emergence: 1989–1998 the emergent stage of the application of the complexity sciences to public administration and policy studies was in the decade of 1989–1998. these earliest efforts can be viewed as testing the relevance of the complexity sciences to the field of public administration and policy studies. these early publications were generally focused on applications of dissipative structures kiel (1989) and notions of self-organization (comfort, 1994) to public organizations. the majority of these early publications is quite general and definitional in nature and intended to acquaint the scholarly community with the terms and concepts of the complexity sciences. thus these efforts tested whether the larger public administration and policy community would accept the relevance of the basic concepts of the complexity sciences. the challenge of identifying the proper language is evidenced in works that examined topics ranging from chaos theory to quantum mechanics (overman, 1996). this body of literature was metaphorical in tone and content. in the literature public organizations were viewed as dissipative self-organizing structures. in the emergent stage there were some early time-series analyses as well. kiel and elliott (1992) used phase diagrams to examine the nature of change in the u.s. federal budget. kiel (1993) applied phase diagrams to spending and work activity dynamics in a state government agency. the most sophisticated time-series analysis during this period is kiel and seldon’s (1998) discovery of nonlinearity and randomness in time series data from a police organization. the emergent stage of applications of the complexity sciences to public administration and policy studies was sustained as top journals such as public administration review and the journal of public administration research and theory showed willingness to publish works incorporating the novelty of the complexity perspective. kiel’s work (1994) was the first book-length effort applying the complexity sciences, at that time generally labelled chaos theory, to public administration. the risk, inherent in all novelty, was reduced as the primary sources of knowledge dissemination accepted the value of the complexity sciences to our field. 3. convergence: 1999–2002 in the convergent stage, once emergent phenomenon can now maintain a distinct and coherent niche in its environment. such coherence is exemplified in two seminal works that evidenced the convergent stage of evolution. these two seminal works are comfort’s, shared risk (1999) and morçöl’s, a new mind for policy analysis (2002). louise comfort’s book, shared risk, is an underappreciated work. this was the first book to apply the complexity sciences in a comprehensive and methodologically rigorous way to an important public administration and policy phenomenon. comfort’s highly detailed analysis used considerable data from actual seismic events to explore the dynamics of organizational responses to earthquakes. comfort showed that statistical tools such as nonlinear logistical regression are useful in understanding responses to extreme events. article_14-9.indd 74 26-06-2014 17:32:40 l. douglas kiel / complexity theory and its evolution 75 comfort’s study of organizational responses to extreme events is also important because it laid the groundwork for the larger recognition of the importance of networks when exploring complex governmental phenomena. comfort’s work (1999) remains required reading in the field today. göktuğ morçöl’s a new mind for policy analysis further evidences the convergent stage of the evolution of complexity studies in public administration and policy studies. morçöl’s work reveals that the vast scope of the field of policy studies was also amenable to the epistemology and methods of the complexity sciences. a new mind for policy analysis was the first work of philosophical depth to explore the relevance and need for applications of complexity to policy studies. his book also fit complexity studies into the larger socio-historical body of scientific explanation. he also examined a large scope of issues ranging from the relevance of governmental action in a complex economy to the relevance of cognitive science to studying complex policies. i see these two seminal works as the archetypes of the convergent stage of an academic field. morçöl (2002) places the complexity sciences within the larger evolution of scientific paradigms. this book provides the theoretical and philosophical foundations that provide future scholars the foundations for both applications and further theory development. comfort’s work (1999) confirmed that the complexity sciences could enhance our understanding of complex “real-world” phenomena. the theory and methodological foundations represented in these two works showed that the field was adequately stable to expedite a succeeding stage of proliferance. 4. proliferance: 2003 to the present the proliferating stage of an evolving phenomenon represents maturity and fecundity. the convergent stage resolves the risk of emergence, allowing more attraction and a proliferating body of knowledge and tools. the proliferant stage represents a stage of increasing production. applications of network theory and the methods of network analysis have proliferated in the current era (kapucu, 2006; kapucu, arslan, & collins, 2009; kapucu & garayev, 2011; koliba, meek, & zia, 2011). the uncertainties, nonlinearities and instabilities often associated with networks coincide with the behavior of complex systems. thus the complexity sciences provide a strong backdrop for further exploration of public sector networks. it is not possible here to provide a comprehensive picture of the scope and depth of the literature of network analysis applying complexity principles to public administration and policy studies. this proliferant literature though is likely to maintain its current upward trajectory as we learn more about the challenges and prospects of complex governance networks. another proliferating method that is associated with the complexity sciences is agent-based modeling (abm). numerous applications of agent-based modeling in public administration and policy now exist. these applications range from models of lottery article_14-9.indd 75 26-06-2014 17:32:40 76 l. douglas kiel / complexity theory and its evolution markets (chen & chie, 2008) to models of humanitarian interventions in war torn nations (mccaskill, 2012). the enormity of the potential applications of abm to public administration and policy studies is clearly evident in this period of proliferation (see,  dennard, richardson, & morçöl, 2008; kim & lee, 2007). the fact that mccaskill’s (2012) abmbased dissertation received the best dissertation award of the national association of schools of public affairs and administration is clear evidence that the public administration scholarly community is prepared to incorporate abm into the body of accepted research methods. further evidence of the proliferation of the applications of the complexity studies to public administration and policy is the obvious increase in the number of scholars contributing to this literature. edited volumes by teisman and klijn (2008) and teisman, van buuren, and gerrits (2009) are evidence of this growing body of scholars capable of applying complexity concepts to issues of public concern and governance. dennard, richardson and morçöl’s (2008) edited volume also reveals the expanding body of methods and scholars capable of applying these methods to the many complex challenges of governance. three recent works also validate the view that the current period represents a period of proliferating knowledge production. morçöl’s (2012) recent book represents a new level of maturity in the field as knowledge of both theory and method proliferate to ask new questions and perhaps settle old questions concerning the potential for building theory of relevance to complex governance networks. scholars have recognized for two decades that when contending with complexity practitioners and analysts may, at best, receive guidance only from heuristics (kiel, 1994). the proliferant period of knowledge production reveals the production of such heuristics. room (2011) now provides such a roster of heuristics applicable to policy guidance and implementation in our “turbulent world.” most recently, ramalingam’s (2014) application of the complexity sciences to the challenges of foreign aid reinforce the dynamics of a proliferating enterprise as the number of applicable phenomena of relevance expands. finally, it is worthy of note that evidence of a proliferating academic enterprise is the reality of the increasing difficulty in keeping up with the growing body of literature. 5. divergence: the future? the divergent stage of an evolving phenomena represents a decoupling, a bifurcation perhaps, from the proliferant form (schumacher, 1986). this decoupling may lead to novelty that creates new emergent forms that again initiate the process of emergence, convergence, proliferance, and divergence that constitute the cycle of evolving complexity. the knowledge artifacts of the divergent stage are generally difficult to determine as to specifics since the proliferant period does not provide all of the information required to determine which mutations in the process of the knowledge “speciation” may diverge to create novel forms. attempting to define the divergent stage of evolving nonlinear and dynamic phenomena, such as knowledge production, may prove fanciful. however, the article_14-9.indd 76 26-06-2014 17:32:40 l. douglas kiel / complexity theory and its evolution 77 proliferation of knowledge in several academic fields may add insights into future developments in our field. the vast majority of the literature of the complexity sciences applied to public administration and public policy is focused on the organizational or institutional level of analysis. advances in the cognitive sciences and the neurosciences, however, may add to our knowledge of the challenges of complexity at the individual level. this new knowledge is salient, as it may enhance our knowledge of the individual level functions and dysfunctions of coping with rapidly evolving complexity. improved understanding of the individual level cognitive challenges of evolving complexity, combined with our knowledge of organizations and institutions, may lead to new means for building human systems that are both resilient and appreciative of human capabilities and limitations. the rise of “big data” and the ability to capture increased detail in related longitudinal data may also add to our understanding of the temporal dynamics of complex governance systems. emerging mechanisms for tracking individual level behavior (pentland, 2008) may also allow studies of human behavior within organizational and institutional contexts that provide greater levels of detail and complexity than have previously been available to scholars. 6. conclusion a distinct theoretical and methodological foundation for the study of complex governance systems now exists. the theory of the endeavor is now well established (koliba, meek, & zia, 2011; morçöl, 2012) and the methodological mechanics (comfort, 1999; dennard, richardson, & morçöl, 2008) are well defined. scholars interested in expanding the body of knowledge now possess the means for moving the entire field beyond a stage of growth and expansion to a stage of great relevance and of greater support for understanding and managing complex human systems. the success of any academic endeavor is subject to probability distributions of many interacting variables, such as timing, environmental circumstances, and validation. the trajectory of complexity studies in public administration and policy engenders confidence that the confluence of these probability distributions will result in achieving the goals of informing practice and in developing interdisciplinary social and administrative sciences that further knowledge production. in a historical period typified by increasing complexity, knowledge of the processes of complexity generation and the handling of that complexity are increasingly central to the democratic project, to the lives of individuals and to humanity’s future. references chen, s., & chie, b. (2008). agent-based modeling of lottery markets: policy-making from the bottom up. in l. dennard, k. richardson, & g. morçöl (eds.), complexity and policy analysis: tools and concepts for designing robust policies in a complex world (pp. 231–248). goodyear, az: isce publishing. article_14-9.indd 77 26-06-2014 17:32:40 78 l. douglas kiel / complexity theory and its evolution comfort, l. k. (1994). self-organization in complex systems. journal of public administration research and theory, 4(3), 393–410. comfort, l. k. (1999). shared risk: complex systems in seismic response. new york, ny: pergamon. dennard, l., richardson, k., & morçöl, g. (eds.). (2008). complexity and policy analysis: tools and concepts for designing robust policies in a complex world. goodyear, az: isce publishing. kapucu, n. (2006). interagency communication networks during emergencies: boundary spanners in multiagency coordination. the american review of public administration, 36(2), 207–225. kapucu, n., arslan, t., & collins, m. l. (2009). examining intergovernmental and interorganizational response to catastrophic disasters: toward a network—centered approach. administration & society, 42(1), 222–247. kapucu, n., & garayev, v. (2011). collaborative decision-making in emergency and crisis management. international journal of public administration, 34(6), 366–375. kiel, l. d. (1989). nonequilibrium theory and its implications for public administration. public administration review, 49(6), 544–551. kiel, l. d. (1993). nonlinear dynamical analysis: assessing systems concepts in a government agency. public administration review, 53(2), 143–153. kiel, l. d. (1994). managing chaos and complexity in government: a new paradigm for managing change, innovation and organizational renewal. san francisco, ca: jossey-bass. kiel, l. d., & elliott, e. (1992). budgets as dynamic systems: change, variation, time and budgetary heuristics. journal of public administration research and theory, 2(2), 139–156. kiel, l., & seldon, b. (1998). measuring temporal complexity in the external environment: nonlinearity and the bounds of rational action. american review of public administration, 28(3), 246–265. kim, y., & lee m. (2007). agent-based models as a modeling tool for complex policy and managerial problems. korea journal of public administration, 45(2), 25–50. koliba, c., meek, j., & zia, a. (2011). governance networks in public administration and public policy. new york, ny: taylor & francis. kuhn, t. (1970). the structure of scientific revolutions (2nd ed.). chicago, il: university of chicago press. mccaskill, j. (2012). the effectiveness of coordinating military and nongovernmental organization activities in complex humanitarian interventions: testing policy options with agent-based modeling (doctoral dissertation). university of texas at dallas. morçöl, g. (2002). a new mind for policy analysis: toward a post-newtonian and postpositivist epistemology and methodology. westport, ct: praeger. morçöl, g. (2012). a complexity theory for public policy. new york, ny: routledge. overman, e. s. (1996). the new science of management: chaos and quantum theory and method. journal of public administration research and theory, 6(1), 75–89. pentland, a. (2008). honest signals: how they shape our world. cambridge, ma: mit press. ramalingam, b. (2014). aid on the edge of chaos. new york, ny: oxford university press. rogers, e. (1983). diffusion of innovations (2nd ed.). new york, ny: free press. room, g. (2011). complexity, institutions and public policy: agile decision-making in a turbulent world. cheltenham, england: edward elgar. schumacher, b. g. (1986). on the origin and nature of management. norman, ok: eugnosis press. teisman, g., & klijn, e.-h. (2008). complexity theory and public management: an introduction. public management review, 10(3), 287–297. teisman, g., van buuren, a., & gerrits, l. (2009). an introduction to understanding and managing complex process systems. in g. teisman, a. van buuren, & l. gerrits (eds.), managing complex governance systems: dynamics, self-organization and coevolution in public investments (pp. 1–16). new york, ny: routledge. ziliak, s., & mccloskey, d. (2008). the cult of statistical significance: how the standard error costs us jobs, justice and lives. ann arbor, mi: university of michigan press. article_14-9.indd 78 26-06-2014 17:32:40 policy energy and public management networks complexity, governance & networks (2014) 39–48 39 doi: 10.7564/14-cgn6 policy energy and public management networks michael mcguirea,b*, rachel fyallc a indiana university b kyung hee university e-mail: mcguirem@indiana.edu c university of washington e-mail: fyall@u.washington.edu public management networks are fixtures of contemporary governance. they play an important role in planning for and delivering public goods and services. however, public management networks do not always possess the capacity to convert collective solutions to formal policy or program adjustments. despite the very broad range of activities in them, the ability of networks to create this “policy energy” is limited. government or other powerful agencies can often dominate the management of networks, elected officials may make policy decisions that are inconsistent with the recommended action of networks, and assessing the performance of networks is very often a moving target. this article discusses these and other types of administrative and political barriers that can hinder the ability of public management networks to influence policy making and implementation. keywords: public management networks, network limitations. 1. introduction the scope of attention given to managing networks has increased over the past few decades. as more practitioners work across organizational boundaries (kettl, 2006), more scholars have focused their attention to networks by developing frameworks for managing network processes (ansell & gash, 2008; bryson, crosby, & stone, 2006; emerson, nabatchi, & balogh, 2012; mcguire, 2002; purdy, 2012; rethemeyer & hatmaker, 2008). the underlying assumption of these frameworks is that management enables multiple organizational representatives to create multi-agency solutions to public problems and to achieve the collective goal(s) of the network. policy design and implementation are portrayed merely as direct products of effective network administration. while such atmospheres do prevail sometimes, the opposite also exists in network operations: “acrimony, power determination, disagreement over problems and aims, impossibility in reaching agreement, and lack of implementation ability” (mcguire & agranoff, 2011, p. 266). these obstacles are rarely discussed in the network and collaboration literatures (vangen & huxham, 2014). * corresponding author. article_14-6.indd 39 03/07/14 1:34 am 40 m. mcguire et al / policy energy and public management networks the types of networks portrayed in these and other frameworks are what we refer to as “public management networks,” which include agencies involved in public policy making and administrative structures through which public goods and services are planned, designed, produced, and delivered (and any or all of the activities). such networks are typically intersectoral, intergovernmental, and based functionally in a specific policy or policy area. that is, officials from government organizations and agencies at federal, state, and local levels operate in structures of exchange and production with representatives from profit-making and not-for-profit organizations. we differentiate between public management networks and social networks. the former are goal-directed, whereas the latter are serendipitous networks comprised of individuals who happen to connect. social networks are not managed in the sense that direction is applied to structuring and guiding the relationships. social networks do include specific, but sometimes unintentional, patterns of connectivity (cross & parker, 2004). these social networks in the workplace, for example, can be important mechanisms for improving managerial decisions and they are not devoid of goals, but they “do not need to be governed as a whole, since they do not aim to achieve a preconceived shared objective” (saz-carranza, 2012, p. 7). the public management networks discussed here are governed to some degree by network members. while recent studies elucidate the types of managerial skills that are (or can be) utilized in network settings, there has not been enough focus on the problem of whether all public management networks are strong enough to make a substantive difference in determining the design and outcomes of public policy. take, for example, the u.s. national rural development partnership (nrdp) that was created in 1990. nrdp was a collaborative effort involving representatives from all three levels of governments, tribal governments, the private sector, and the nonprofit sector to promote rural development across the nation. the primary vehicles through which the partnership sought to achieve its goals were state rural development councils (srdcs). these state-level councils, still in existence today in a handful of states, are designed to advance rural interests through collaboration, partnerships, and helping rural communities to access resources they need to accomplish their community-level goals, and are responsible for creating their own mission, structure, and operating guidelines, and implementing action plans. the ability of these public management networks to affect change on a regional or national level, however, was constrained by governor’s offices, state economic development agencies, state legislatures, and the state-based offices of the largest funder, the u.s. department of agriculture (radin et al., 1996). federal funding for the srdcs was eliminated within the last decade for most of the 40 councils and those that remain are funded largely through foundation grants. while heralded in the 1990s as one response to the problems of rural communities, these substantive policy barriers were not easily overcome by street-level, network action. the network system of tribes, community providers, industry associations, and state and local implementers could not turn around this reluctance to continue the program. there are other examples of the limitations of public management networks in spain (blanco & goma, 2002), the united kingdom (newman, 2005; rhodes, 1997), cities in the midwestern article_14-6.indd 40 03/07/14 1:34 am m. mcguire et al / policy energy and public management networks 41 united states (agranoff & mcguire, 2003), and scotland (huxham & vangen, 2005), but these studies are often outnumbered by the rosier frameworks dotting the public management research landscape. our central argument is that network deliberations and the solutions networks propose for implementing policies often fail to replace government decisions, even when implementation is shared with government agencies. as a blocking force, agency (nonnetwork) power is very real when lead organizations, often government agencies, serve to keep certain problems off the agenda, withhold support for key network strategies or decisions, or withhold required agency-controlled resources (mcguire & agranoff, 2011). in these situations, public management networks may have low capacity for policy (formulation and/or implementation) influence or policy change, which constitutes low levels of what we call policy “energy.” policy energy refers to the capacity of any organization, agency, or other structural arrangement (in this case, public management networks) to affect the design and outcomes of a public policy or program. too often, networks devise reasonable solutions, but then run into administrative and political barriers that prevent the next action steps. and public management networks rarely have the legitimate authority to enforce compliance with their own preferred solutions. as o’toole (1997) argues, administrators working within networks “should not assume that they possess authority” (p. 48). when managing networks, goals can be shared, objectives can be achieved, and solutions can be offered, but the government agency’s representative(s), even if s/he is a member of the network, usually makes the final policy decision. lacking policy energy, public management networks will recommend action more than they decide policy. this article focuses on the administrative and political barriers that public management networks face in trying to generate policy energy. 2. administrative barriers to generating policy energy much has been learned by scholars who have investigated the activities of public management networks. the “black box” of the workings of such networks is becoming clearer, or at least less muddled. for example, research indicates that governing authority in public management networks is not always conferred or assigned to a single government or non-governmental agency (weber & khademian, 2008). more than one manager or leader may exist in a single network, and that managerial or leadership role may not be defined as clearly as that of a supervisor in hierarchical organizations. also, scholars acknowledge that public management networks are governed largely by network members’ ability to facilitate collaboration through relational skills such as patience, empathy, honesty, and deference (vangen & huxham, 2003). administrative tasks such as motivating network members, creating trust, treating others as equals, maintaining a close knit group, sharing information freely amongst networks members, and other people-oriented behaviors are important to ensure network effectiveness (silvia & mcguire, 2010). research also suggests that the typical network member (or manager) operates in multiple networks. although a “home” network may be the primary responsibility of a manager(s), other article_14-6.indd 41 03/07/14 1:34 am 42 m. mcguire et al / policy energy and public management networks public management networks intersect or are adjacent to the home network, resulting in a system of networks that can lead to overlapping responsibilities and roles for network members (rethemeyer & hatmaker, 2008). the black box of networks also contains the ingredients for administrative obstacles to generating policy energy. one such obstacle can be found in the different, sometimes conflicting methods for a manager to facilitate network activity. milward and provan (2006) build a framework based on the idea that there may be a designated, legitimate steerer of the network, but there may also be managers in the network, who can have split missions and sometimes split loyalties. this distinction between steerers and managers is consistent with the ideas proposed by those who differentiate between governance of a network (exercising control, regulation, inducement, incentive or persuasive influence over the whole network), and management within a network (activities carried out by individuals who are themselves actors in the network) (6, goodwin, peck, & freeman, 2006). the role of a manager “is not given a priori to one actor” (koppenjan & klijn, 2004, p. 11). network management can be undertaken by several actors, with managerial activities sometimes conducted simultaneously by more than one actor. since network members represent the values and preferences of their home agencies as well as those of the network, the members can be placed in a conundrum: managers may have to choose between serving their individual agencies’ missions and the mission of the network (thomson & perry, 2006). additionally, few government agency administrators work full-time in networks. with the exception of certain “boundary spanners” who work full time to fulfill crossagency tasks (leroux, 2014), many administrators spend as little as 15 to 20 percent of their total time in collaborative activities, including participating in networks (agranoff & mcguire, 2003). there actually is little time for a public management network to develop policy energy due to the other overwhelming demands on network members. there are situations in which the different managers in the network enact their roles differently. one network member might approach the role as a facilitator who is willing and able to move the collaborative agenda forward by allowing all members to have a voice and seeking consensus. others may depend on “collaborative thuggery,” which is perhaps a more pragmatic but less relational approach to managing networks that generally is not consistent with the spirit of collaboration (vangen & huxham, 2003). reconciling these dual roles is not easy, and leads to public management networks trying to ameliorate conflicting agendas rather than advancing the effective design and implementation of policy. policy energy is thus sapped by managerial conflict. the process of addressing such internal conflict results in little time for exerting effort into managing external stakeholders. these external stakeholders, whose support is needed to formulate and implement policy (silvia, 2011), and therefore generate policy energy, include the recipients of the services provided by the network, the home agency that the members of the network represent, as well as those home agencies’ stakeholders. perhaps most important, the stakeholders who are rarely touched by managerial activity are elected officials and top agency leaders. internal relationships can trump the maintenance of external relationships, making it difficult to cope with the power of external article_14-6.indd 42 03/07/14 1:34 am m. mcguire et al / policy energy and public management networks 43 forces such as key agencies, critical policymakers, and other important “movers” in the policy world (mcguire & agranoff, 2011). while in the midst of moving the collaboration agenda forward, there are processing costs that can overwhelm public management networks. network participants are concerned about the time and opportunity costs taken away from the management of their home organization (agranoff, 2007). graddy and chen (2006) point to the transaction costs involved in network-based decision making. many hours are spent in task forces or work groups that occur outside of the more formal partner meetings, resulting in a type of “collaborative inertia” that is marked by sluggish progress, lack of achievements, and sometimes network collapse (huxham & vangen, 2005). even when this inertia is somehow overcome, it nevertheless comes at the expense of prolonged consensus building and relationship management. such relationship management is often in a state of flux, requiring the “constant interplay between process design and process management’ (edelenbos & klijn, 2006, p. 426), and resulting in hyperactivity and inefficient overprocessing. 3. political barriers to generating policy energy how do political barriers and power imbalances prevent public management networks from influencing the design and outcomes of public policy? it has become a truism in research that power and influence may be unequally distributed in networks (mcguire, 2011). the more power possessed by an organization (or the organization’s representative), the more influence it has to determine the nature of the interorganizational exchange (cook, 1977). a scholarly focus on management based in consensus building and relationship management masks the potential for coercive behavior in networks. the more powerful members of the network can shape network activities through such means as controlling the agenda and even withholding needed policy making resources (e.g. finances, human capital, political support, information, expertise). powerful members can resist collaboration because of their concerns for protecting their agencies’ “turf,” whereby they defend against other network members encroaching on their autonomy. power imbalances also can exist in networks because some organizations or even a single organization bring to the network an inordinate amount of resources; thus the organization’s representative is able to exert undue influence over the other network members. such power asymmetries suggest “the inevitability that some partners will be more central to the enactment of the collaborative agenda than are others . . .” (vangen & huxham, 2014, p. 58). these network partners may be in opportunistic positions (burt, 2009) that create unequal opportunity among network members, while others may be less willing or able members of a network. it is also true that in addition to experiencing power asymmetries, power can be and often is redistributed within the network over time. gray (1989) acknowledges that at least a temporary redistribution of power occurs regularly in collaborative settings. in these situations, the design of public policy can be dominated either by a single network member or by members whose policy influence waxes and wanes. article_14-6.indd 43 03/07/14 1:34 am 44 m. mcguire et al / policy energy and public management networks powerful governmental actors shape and determine public policy outcomes through the instrumental use of networks (klijn & skelcher, 2007). one example of this can be found in the increasing use—or misuse—of performance measures, which are often determined by political considerations (radin, 2006). as koliba (2014) argues regarding networks, “[q]uestions of performance are eminently informed by who has power” (p. 88). various stakeholders involved in the network evaluate its effectiveness using multiple and sometimes conflicting criteria, and different constituencies expect different outcomes. these constituents are principals “who monitor and fund the network; agents, who work in the networks both as administrators and service-level professionals; and clients, who actually receive the services provided by the network” (provan & milward, 2001, p. 416). the relationships between principals, agents, and clients, result in differing assessments of effectiveness by the community, the network itself, and the organizations that are part of the network. in addition to the potentially conflicting assessments of performance by various stakeholders, network objectives vary over time, thus making it difficult to establish a consistent and clear policy goal. public management networks themselves can be so complex that their impact on performance is somewhat unpredictable for all involved (o’toole, 1997), so evaluating a network’s performance based on ex ante objectives is often unrealistic (klijn & koppenjan, 2000). public management networks have no central, coordinating actor to design and monitor goals, therefore measures based in goal achievement are less applicable. because of the various stakeholders involved, it is a challenge just to determine whose objectives or perspectives should be prioritized when trying to evaluate network performance. if certain parties do not participate in the interaction process, the chances are high that their interests and preferences will not be represented in any network-derived solution. even when a representative of an agency is supportive of a network solution, it does not mean that the entire agency and its sponsors/funders will be supportive. the ability of public management networks to create policy energy is limited when a policy outcome is a continuously moving target. networks have not eclipsed the centrality of government agencies in designing and implementing public policy, which is another reason why networks are limited in generating policy energy. even in public management networks comprised of representatives from many different organizations and across sectors, it is the network member(s) from government agencies who often pull the various governmental and non-governmental forces together and distribute power and responsibilities; it is also the focus of political identity and the main instrument of political legitimacy (hirst, 2000). a public organization is created by law to serve a particular purpose, whereas many public management networks are cooperative ventures whose members must continually work to convince their stakeholders that their work in the larger network is valuable and worthwhile (milward & provan, 2006). there still are legal mandates and financial resources that government agencies control, and the legitimacy that government participation brings to the network is sometimes necessary to garner and maintain the support of both network members and external stakeholders. for these reasons, the predominant nature of the government agency can article_14-6.indd 44 03/07/14 1:34 am m. mcguire et al / policy energy and public management networks 45 virtually overlay or even overshadow the operations of networks, which is perhaps the ultimate political barrier to creating policy energy (mcguire & agranoff, 2010). 4. conclusion we have argued that some public management networks lack substantial levels of policy energy. government agencies are changing to accommodate collaborative, networked activity as their representatives increasingly work with organizations external to the agency, but networks still face administrative and political obstacles that constrain their capacity to make policy changes and program adjustments. a government manager may be convinced that s/he knows best and seek to control network processes and performance, which results in the network as a whole possessing only minimal influence on the policy process. the transaction costs of managing public management networks can be high and sometimes insurmountable due to the multiple and conflicting agendas with which each network member must contend. performance measures for networked programs can be “gamed” and external political considerations of the effectiveness of networks can trump the demands of managing the internal processes of networks. these obstacles are not easy to overcome. we do not wish to overstate the argument. public management networks are far from powerless and they certainly can, and do, influence the design and outcomes of public policy. the increase of research on public management networks confirms their prevalence and relevance to public management in general. rarely, however, is there an acknowledgement by researchers that there are limitations to public management networks, particularly in their capacity for generating policy energy. scholars should turn their attention to understanding this important phenomenon. references 6, p., goodwin, n., peck, e., & freeman, t. (2006). managing networks of twenty-first century organisations. new york, ny: palgrave macmillan. agranoff, r. (2007). managing within networks: adding value to public organizations. washington, dc: georgetown university press. agranoff, r., & mcguire, m. (2003). collaborative public management: new strategies for local governments. washington, dc: georgetown university press. ansell, c., & gash, a. (2008). collaborative governance in theory and practice. journal of public administration research and theory, 18, 543–571. blanco, i., & goma, r. (2002). proximidad y participacion: marco conceptual y presentacion de experiences. in i. blanco, & r. goma (eds.), gubiernos locales y redes participativas. barcelona, spain: ariel. bryson, j. m., crosby, b. c., & stone, m. m. (2006). the design and implementation of cross-sector collaborations: propositions from the literature. public administration review, 66, 44–55. burt, r. s. (2009). structural holes: the social structure of competition. cambridge, ma: harvard university press. cook, k. (1977). exchange and power in networks of interorganizational relations. the sociological quarterly, 16, 62–82. article_14-6.indd 45 03/07/14 1:34 am 46 m. mcguire et al / policy energy and public management networks cross, r., & parker, a. (2004). the hidden power of social networks. boston, ma: harvard business school press. edelenbos, j., & klijn, e. -h. (2006). managing stakeholder involvement in decision-making: a comparative analysis of six interactive processes in the netherlands. journal of public administration research and theory, 16, 417–446. emerson, k., nabatchi, t., & balogh, s. (2012). an integrative framework for collaborative governance. journal of public administration research and theory, 22, 1–29. graddy, e. a., & chen, b. (2006). influences on the size and scope of networks for social services delivery. journal of public administration research and theory, 16, 533–552. gray, b. (1989). collaborating: finding common ground for multiparty problems. san francisco, ca: jossey-bass. hirst, p. (2000). democracy and governance. in j. pierre (ed.), debating governance. oxford, england: oxford university press. huxham, c., & vangen, s. (2005). managing to collaborate: the theory and practice of collaborative advantage. london, england: routledge. kettl, d. f. (2006). managing boundaries in american administration: the collaboration imperative. public administration review, 66, 10–19. klijn, e. -h., & koppenjan, j. f. m. (2000). public management and policy networks: foundation of a network approach to governance. public management, 2, 135–158. klijn, e-h., & skelcher, c. (2007). democracy and governance networks: compatible or not? public administration, 85, 587–608. koliba, c. (2014). governance network performance: a complex adaptive systems approach. in r. keast, m. mandell, & r. agranoff (eds.), network theory in the public sector: building new theoretical perspectives (pp. 84–102). new york, ny: routledge. koppenjan, j. f. m., & klijn, e. -h. (2004). managing uncertainties in networks. london, england: routledge. leroux, k. (2014). the boundary spanning solution? crossing boundaries in the united states. in j. o’flynn, d. blackman, & j. halligan (eds.), crossing boundaries in public management and policy: the international experience (pp. 64–77). new york, ny: routledge. mcguire, m. (2002). managing networks: propositions on what managers do and why they do it. public administration review, 62, 599–609. mcguire, m. (2011). network management. in m. bevir (ed.), sage handbook of governance (pp. 436–453). thousand oaks, ca: sage. mcguire, m., & agranoff, r. (2010). networking in the shadow of bureaucracy. in r. durant (ed.), the oxford handbook of american bureaucracy (pp. 372–395). oxford, uk: oxford university press. mcguire, m., & agranoff, r. (2011). the limitations of public management networks. public administration, 89, 265–284. milward, h. b., & provan, k. g. (2006). a manager’s guide to choosing and using collaborative networks. washington, dc: ibm center for the business of government. newman, j. (2005). enter the transformational leader: network governance and the micropolitics of modernization. sociology, 39, 717–734. o’toole, l. j., jr. (1997). treating networks seriously: practical and research-based agendas in public administration. public administration review, 57, 45–52. provan, k. g., & milward, h. b. (2001). do networks really work? a framework for evaluating public sector organizational networks. public administration review, 61, 414–423. purdy, j. m. (2012). a framework for assessing power in collaborative governance processes. public administration review, 72, 409–417. radin, b. a. (2006). challenging the performance movement: accountability, complexity, and democratic values. washington, dc: georgetown university press. radin, b. a., agranoff, r., bowman, a., buntz, g. c., ott, s. j., romzek, b. s., & wilson, r. h. (1996). new governance for rural america: creating intergovernmental partnerships.lawrence, ks: university press of kansas. article_14-6.indd 46 03/07/14 1:34 am m. mcguire et al / policy energy and public management networks 47 rethemeyer, r. k., & hatmaker, d. (2008). network management reconsidered: an inquiry into management of network structures in public service sector provision. journal of public administration research and theory, 18, 617–646. rhodes, r. a. w. (1997). understanding governance: policy networks, governance, reflexivity and accountability. buckingham, uk: open university press. saz-carranza, a. (2012). uniting diverse organizations: managing goal-oriented advocacy networks. new york, ny: routledge. silvia, c. (2011). collaborative governance concepts for successful network leadership. state and local government review, 43, 66–71. silvia, c., & mcguire, m. (2010). leading public sector networks: an empirical examination of integrative leadership behaviors. the leadership quarterly, 21, 264–277. thomson, a. m., & perry, j. l. (2006). collaboration processes: inside the black box. public administration review, 66, 20–32. vangen, s., & huxham, c. (2003). enacting leadership for collaborative advantage: dilemmas of ideology and pragmatism in the activities of partnership managers. british journal of management, 14, s61–s76. vangen, s., & huxham, c. (2014). building and using the theory of collaborative advantage. in r. keast, m. mandell, & r. agranoff (eds.), network theory in the public sector: building new theoretical perspectives (pp. 51–67). new york, ny: routledge. weber, e. p., & khademian, a. (2008). wicked problems, knowledge challenges, and collaborative capacity builders in network settings. public administration review, 68, 334–349. article_14-6.indd 47 03/07/14 1:34 am article_14-6.indd 48 03/07/14 1:34 am complexity, governance, and networks: perspectives from public administration complexity, governance & networks (2014) 29–38 29 doi: 10.7564/14-cgn3 complexity, governance, and networks: perspectives from public administration naim kapucu university of central florida e-mail: kapucu@ucf.edu complex public policy problems require a productive collaboration among different actors from multiple sectors. networks are widely applied as a public management tool and strategy. this warrantrs a deeper analysis of networks and network management in public administration. there is a strong interest in both in practice and theory of networks in public administration. this requires an analysis of complex networks within public governance settings. in this this essay i briefly discuss research streams on complex networks, network governance, and current research challenges in public administration. keywords: complexity, networks, network governance, network analysis. 1. introduction complex public problems require a productive collaboration across different organizations, professions, and sectors. network governance refers to inclusive and participatory approaches to policy-making, collective action, and implementation. in addressing the complex policy problems, public managers, civic leaders, and professionals are often being expected to cooperate with each other. in difficult times, collaborative strategies and cross-boundary institutions gain importance for sharing resources, solving complex policy problems, and leveraging experience and knowledge. devolution, privatization, and regulation are social, political, and economic trends that have resulted in an increased use of networks in the u.s. and affected their structures and functions. the move to privatize public services involves transfers of ownership to increase efficiency in them. as the popularity of the new public management approach waned (polidano, 2001), network governance and collaborative management approaches in administering programs to provide public services have gained popularity (mcguire, 2006; provan & kenis, 2007). the network governance approach has become influential also in the literature on public policy deliberation, design, and implementation, as policy processes have involved increasingly multiple stakeholders and deregulation. the new forms of governance have been depicted in terms of network relations in the literatures of public administration and public policy. it is argued that governance, policy, or public management networks help article_14-3.indd 29 03/07/14 1:31 am 30 n. kapucu / complexity, governance, and networks the public sector mobilize resources from non-state actors in order to deliver effective services and formulate and implement public policies (frederickson, 1999; ingold & varone, 2012; kapucu, 2012; powell, 1990; rhodes & williams, 1996). the literature has identified various types of networks such as collaborative networks, policy networks, public management networks, and governance networks (heaney, 2006; isett, mergel, leroux, mischen, & rethemeyer, 2011). the literature on policy networks is the oldest; the focus of this literature is mainly on public policy decision-making and involves policy making actors. the term collaborative networks is commonly used to depict the networks of governmental and non-governmental organizations in the administration of public programs and the delivery of public services. the term governance networks emphasize the governance process to achieve cross-sector and inter-organizational goals in the public sector (isett et al., 2011). governance networks conceptualization call to the attention the fact that government agencies, or the public sector in general, cannot solve today’s complex policy problems by themselves and they need the help of non-state actors (lecy, mergel, & schmitz, 2013). in the public administration and policy literature, often the terms policy, collaborative, and governance networks are used interchangeably. 2. complexity, governance, and networks networks can offer the complex, interdependent responses to complex policy problems that have uncertain solutions and can lead to political disagreements (provan & lemaire, 2012). complex problems “lack a definite formulation” and a “stopping rule”; “[e]numerable sets of potential solutions” do not exist for these wicked problems, and the solutions are just “better or worse,” not right or wrong (koliba, meek, & zia, 2011a, p. 15). networks are also necessary when political processes complicate administrative strategies (kettl, 2006). networks can combine human capital in order to participate in collaborative problem-solving activities. the use of network concept can be traced back to the beginning of the studies on federalism in the u.s. (kettl, 2006; koliba et al., 2011a). scholars connected network governance to intergovernmental cooperation in the 1960s (emerson, nabatchi, & balogh, 2012). as public problems have become more and more complex, there has been a realization that individual government agencies working alone can no longer handle them. instead, there has been a growing emphasis on replacing hierarchical bureaucratic structures with more integrated horizontal networks. these networks have been the focus of recent works on collaborative public management involving a variety of network arrangements. although these types of collaborative efforts are increasing in numbers, a related growing concern is whether such arrangements have been any more effective than those involving single-agency or hierarchical structures. the measures that are often used to answer this question are those that apply to individual organizations, rather than network arrangements. networks are complex systems that have emergent qualities, they are adaptive to changing conditions, and they have the capability to self-organize. in both systems theory article_14-3.indd 30 03/07/14 1:31 am n. kapucu / complexity, governance, and networks 31 and complexity theory, the roles of feedback and interactions play a central role in understanding society (koliba et al., 2011a). as complex systems, networks change according to the “relative strength of feedback loops” (koliba et al., 2011a, p. 92). negative feedback are “controls placed over the system that come into effect when the system deviates from a goal” (koliba et al., 2011a, p. 175). positive feedback are rewards to the system that accentuates as a result of the two forces of mimicking and attention shifting, which together lead to saliency of previously ignored policy issues in a debate. the medium of feedback in governance networks may be explicit or tacit in the following possible forms: policy tools – grants, contract agreements, and regulations, representation and interest group competition, acts of administration, accountability, and performance measurement (koliba et al., 2011b). 3. complexity and networks networks involve a number of actors that are not only connected to each other, but also interdependent (agranoff, 2007; agranoff & mcguire, 2001). according to sørensen (2006) the term governance implies “a complex governing process in which a multitude of public and private actors interact to govern society” (p. 99). governance networks are characterized by the complexity of the interactions and decision-making process as well as the policy development and implementation (klijn, edelenbos, & steijn, 2010). complexity is described as the challenges and uncertainties that arise when managing and governing through networks, making policy decisions, and policy implementations. the high level of interdependence, number of functions, and number of actors involved in networks increase their complexity. sørensen (2006) specifically states that the multitude of actors (and expected functions and tasks) involved in the process of governance increases the complexity of network governance. the different perceptions, cultures, goal convergence, institutional frameworks, and power structures of actors involved in the networks also cause complexity (o’leary & vij, 2012). the existence of uncertainties contributes to complexity of networks in addition to the other factors highlighted above. bueren, klijn, and koppenjan (2003) identify three types of uncertainties that typify policy problems: cognitive, strategic, and institutional uncertainty. cognitive uncertainty, as the name implies deals with the lack of a clear understanding of the causal relationships that underlie the problem and are a result of a lack of scientific knowledge on the subject matter. strategic uncertainty results from the multiple perspectives coming from multiple actors trying to address the wicked problem. conflicting strategies and diverging perceptions may result in a lack of progress in formulating solutions and also unexpected consequences. institutional uncertainty is the result of the nature of decision-making that takes place through complex policy networks that are polycentric in nature, fragmented and belong to multiple arenas and levels (local, regional, country-level, global). some in the literature highlight the complexity and uncertainty in governing networks and focus on how to deal with policy problems by utilizing networks and governance article_14-3.indd 31 03/07/14 1:31 am 32 n. kapucu / complexity, governance, and networks effectively. acknowledging complexity of network governance is one thing; developing capacities in effective network governance is another. for example, termeer, dewulf, breeman, and stiller (2012) identify four governance capabilities that are important for addressing these potential problems in managing complex networks. these capabilities are: (1) reflexivity (capability to prevent tunnel vision and appreciate different perspectives and frames); (2) resilience (capability to adjust actions to unpredictable consequences and changing conditions); (3) responsiveness (capability to respond to diverse demands, changing expectations and issues); (4) revitalizing (capability to overcome stagnations and recuperate policy processes). complexity is not only a characteristic of policy problems that networks aim to solve, but it is a characteristic of the governance of the networks as well. collaboration between actors in the network involves a struggle for advantage and compromise to determine the real impact and value of managing in networks (meier & o’toole, 2007). thus, although actors and agencies are working towards achieving collaborative advantage, they also seek to maintain their institutional identity, influence and power (koliba et al., 2011a). according to sørensen and torfing (2009), governance networks aim to solve policy problems and “enhance democratic participation in public policy-making, but it may also create conflicts and deadlocks and make public governance less transparent and accountable” (p. 234). in their comparative analysis of the governance networks literature, klijn and edelenbos (2013) observe that most of the literature in the us has focused on service delivery and policy implementation networks, while the european literature utilizes networks to describe policy creation and decision-making in dealing with complex problems. 4. network governance this section briefly highlights network governance perspectives in dealing complex policy issues and implementation. network governance includes non-state actors; it is a participatory approach to policy-making, collective action, and implementation ( bingham, nabatchi, & o’leary, 2005; provan & huang, 2012). the emphasis is on the complex interactions between different entities to manage governance processes and provision of public services. “networked governance is a particular framing of collective decision making that is characterized by a trend for a wider range of participants to be seen as legitimate members of the decision-making process in the context of considerable uncertainty and complexity” (stoker, 2006, p. 41). the term network governance is commonly substituted with network management in the public administration literature. this literature has focused on how management in networks is different or similar compared to managing in more traditional settings of the hierarchy or a market-based system (agranoff & mcguire, 2001; fung, 2006; meier & o’toole, 2001). agranoff and mcguire (2001) have identified key network management behaviors and strategies in network governance. another major stream of literature concerns the effectiveness and the performance of network governance. some scholars study the performance of local public networks article_14-3.indd 32 03/07/14 1:31 am n. kapucu / complexity, governance, and networks 33 through empirical analysis (span, luijkx, schols, & schalk, 2012), while most others rely on conceptual and theoretical discussions on the outcomes of collaborative arrangements (rogers & weber, 2010), network management strategies and performance measurements of networks (herranz, 2008, 2010; provan & milward, 2001), and network performance and effectiveness (provan & lemaire, 2012; provan & milward, 1995). there is also a literature on the implications of network governance for democracy (bogason & musso, 2006; heikkila & isett, 2007; klijn & edelenbos, 2013; mathur & skelcher, 2007; sørensen, 2006; sørensen & torfing, 2009; zeemering, 2012). there is an emphasis on enhancing accountability and promoting democratic values in networks by the involvement of citizens (bogason & musso, 2006; fung, 2006). klijn and skelcher (2007) suggest that more recently scholars have started studying the relationship between representative democracy and governance networks, or as klijn and skelcher (2007) call it the “democratic dimension of governance networks.” other recent topics and areas of study concern the influence of trust in the performance of governance networks (klijn et al., 2010), accountability of networks ( koliba, mills, &  zia, 2011b), economic development networks (feiock, lee, & park, 2012), and environmental management (robins, bates, & pattison, 2011). some recent studies applied new innovative methods of analyzing performance of networks, such as comparing effectiveness of disaster response networks by comparing planned networks with emergent/actual networks during disasters (brooks, bodeau, & fedorowicz, 2012; choi & brower, 2006; kapucu & demiroz, 2011; nohrstedt, 2013). 5. methodological issues as governing of complex networks provides several challenges, so does studying of the complex network governance. recently the concept of network and network governance have received increasing attention in public administration and policy circles. i personally see network analysis is one of the appropriate methods in studying network governance. in the study of network governance, questions arise about whether boundaries (mission, resources, capacity, responsibility, and accountability) should exist, how they are drawn, and how to deal with their trade-offs. the mission of a network is to address a specific complex problem by allocating resources and a large budget while managing its interorganizational capacity and expertise. each actor in the network must have defined roles and responsibilities to ensure proper contribution to the mission, but the inability to hold any one actor accountable amidst shared responsibility is a potential issue (kettl, 2006). most of the literature that exists on network governance is conceptual and theoretical in nature and there are only few occasions in which the concepts were tested empirically. most of the researchers did not operationalize the term network; they rather used it as a metaphor. the most popular method used in studying networks in public administration is case studies. this is somehow expected since the contextual nature of networks impacts the characteristics and structure of governance networks. the context is also integral to article_14-3.indd 33 03/07/14 1:31 am 34 n. kapucu / complexity, governance, and networks understanding the complexities and processes involved in the interactions and exchanges that make up a governance network. other researchers used comparative case studies (herranz, 2010; nowell, 2009; rogers & weber, 2010). most studies on networks either relied on interviews of network actors, or some document or content analysis. elite interviews are a common source of data for research on governance networks (e.g. sørensen, 2006; zeemering, 2012). sørensen (2006) relied on elite interviews to understand the outcomes of network governance in four danish municipalities, for example. even though limited, some of the articles used social network analysis (sna) as a method. among the measures used, whole network such as density, visual maps, and centralization measures along with ego network measures, such as degree centrality and betweenness centrality were most common. sna is utilized as an analysis method in analyzing theoretical constructs that are defined as relational. these constructs can be processes and outcomes (knox, savage, & harvey, 2006; wasserman & faust, 1994). sna analyzes ties among actors or nodes with in a network. actors within a network can represent individuals, organizations, and communities. these ties can indicate communication, information exchange, formal contractual relations, or, informal friendship ties between nodes. sna provides ways to analyze dynamic relationships between various actors and examining complex processes and various types of interactions within network governance systems. both intra organizational and inter-organizational relationships can be studied using sna. 6. conclusion the ability for organizations to collaboratively work across sectors in order to solve complex public policy problems exists through networks, governance networks, collaborative governance, and collaborative public management. several principles from economics, organization behavior, political science and policy studies, sociology, and human behavior were utilized in addressing complexity networks across individuals, organizations, and sectors. even though the essay used limited number of journals in the field of public administration for a limited time period and books, the growth in studying complexity, governance, and networks in public administration is a noticeable fact. a major body of research on network governance in public administration concerns the performance of networks or the effectiveness of managing through networks. some progress has been made in using different methods, such as case study method, different tools such as social network analysis, as recommended by berry et al. (2004), to advance research on network governance in public administration. yet, significant challenge remains to be addressed by the researchers in the future. the number of network research has significantly increased since o’toole’s (1997) call for more rigorous and systemic research on networks. yet methodological advancements of network research is still behind the need in public administration. compared with the rapidly growing interest in collaborative governance and network research, empirical article_14-3.indd 34 03/07/14 1:31 am n. kapucu / complexity, governance, and networks 35 studies on networks remain limited. i hope that complexity, governance, and networks will provide a platform for researchers from interdisciplinary perspectives, including public administration, to advance research on complex governance networks. references agranoff, r. (2007). managing within networks: adding value to public organizations. washington, dc: georgetown university press. agranoff, r., & mcguire, m. (2001). big questions in public network management research. journal of public administration research and theory, 11, 295–326. berry, f. s., brower, r. s., choi, s. o., goa, w. x., jang, h.s., kwon, m., & word, j. (2004). three traditions of network research: what the public management research agenda can learn from other research communities. public administration review, 64(5), 539–552. bingham, l. b., nabatchi, t., & o’leary, r. (2005). the new governance: practices and processes for stakeholder and citizen participation in the work of government. public administration review, 65(5), 547–558. bogason, p., & musso, j. a. (2006). the democratic prospects of network governance. the american review of public administration, 36(1), 3–18. brooks, j., bodeau, d., & fedorowicz, j. (2012). network management in emergency response: articulation practices of state-level managers – interweaving up, down and sideways. administration & society, 45(8), 911–948. bueren, e. m. v, klijn, e. h., & koppenjan, j. f. m. (2003). dealing with wicked problems in networks: analyzing an environmental debate from a network perspective.journal of public administration research and theory, 13(2), 193–212. choi, s. o., & brower, r. s. (2006). when practice matters more than government plans: a network analysis of local emergency management. administration and society, 37(6), 651–678. emerson, k., nabatchi, t., & balogh, s. (2012). an integrative framework for collaborative governance. journal of public administration research and theory, 22(1), 1–29. feiock, r. c., lee, i. w., & park, h. j. (2012). administrators’ and elected officials’ collaboration networks: selecting partners to reduce risk in economic development. public administration review, 72(s1), 58–68. fung, a. (2006). varieties of participation in complex governance. public administration review, 66(s1), 66–75. frederickson, h. g. (1999). the repositioning of american public administration. ps: political science and politics, 32(4), 701–711. heaney, m. t. (2006). brokering health policy: coalitions, parties, and interest group influence. journal of health politics, policy and law, 31(5), 887–944. heikkila, t., & isett, k. r. (2007). citizen involvement and performance management in special-purpose governments. public administration review, 67(2), 238–248. herranz, j., jr. (2008). the multisectoral trilemma of network management. journal of public administration research and theory, 18(1), 1–31. herranz, j., jr. (2010). multilevel performance indicators for multisectoral networks and management. the american review of public administration, 40(4), 445–460. ingold, k., & varone, f. (2012). treating policy brokers seriously: evidence from the climate policy. journal of public administration research and theory, 22(2), 319–346. isett, k. r., mergel, i. a., leroux, k., mischen, p. a., & rethemeyer, r. k. (2011). networks in public administration scholarship: understanding where we are and where we need to go. journal of public administration research and theory, 21(s1), i157–i173. kapucu, n. (2012). the network governance in response to acts of terrorism: comparative analyses. new york, ny: routledge. kapucu, n., & demiroz, f. (2011). measuring performance for collaborative public management: using network analysis methods and tools. public performance and management review, 34(4), 551–581. article_14-3.indd 35 03/07/14 1:31 am 36 n. kapucu / complexity, governance, and networks kettl, d. f. (2006). managing boundaries in american administration: the collaboration imperative. public administration review, 66(1), 10–19. klijn, e., & edelenbos, j. (2013). the influence of democratic legitimacy on outcomes in governance networks. administration & society, 45(6), 627–650. klijn, e., edelenbos, j., & steijn, b. (2010). trust in governance networks: its impacts on outcomes. administration & society, 42(2), 193–221. klijn, e. h., & skelcher, c. (2007). democracy and governance networks: compatible or not? public administration, 85(3), 587–608. knox, h., savage, m., & harvey, p. (2006). social networks and the study of relations: networks as method, metaphor and form. economy and society, 35(1), 113–140. koliba, c., meek, j. c., & zia, a. (2011a). governance networks in public administration and public policy. new york, ny: crc press. koliba, c., mills, r. m., & zia, a. (2011b). accountability in governance networks: an assessment of public, private, and nonprofit emergency management practices following hurricane katrina. public administration review, 71(2), 210–220. lecy, j. d., mergel, i. a., & schmitz, h. p. (2013). networks in public administration. public management review, 15(1), 1–23. mathur, n., & skelcher, c. (2007). evaluating democratic performance: methodologies for assessing the relationship between network governance and citizens. public administration review, 67(2), 228–237. mcguire, m. (2006). collaborative public management: assessing what we know and how we know it. public administration review, 66(s1), 33–43. meier, k. j., & o’toole, l. j. (2001). managerial strategies and behavior in networks: a model with evidence from u.s. public education. journal of public administration and theory, 11(3), 271–293. meier, k. j., & o’toole, l. j. (2007). modeling public management: empirical analysis of the managementperformance nexus. public administration review, 9(4), 503–527. nohrstedt, d. (2013). explaining mobilization and performance of collaborations in routine emergency management. administration & society, 1–28. doi:10.1177/0095399712473983. nowell, b. (2009). out of sync and unaware? exploring the effects of problem frame alignment and discordance in community collaboratives. journal of public administration research and theory, 20(1), 91–116. o’leary, r., & vij, n. (2012). collaborative public management where have we been and where are we going? the american review of public administration, 42(5), 507–522. o’toole, l. j. (1997). treating networks seriously: practical and research-based agendas in public administration. public administration review, 57(1), 45–52. polidano, c. (2001). why civil service reforms fail. public management review, 3(3), 345–362. powell, w. w. (1990). neither market nor hierarchy: network forms of organization. research in organizational behavior, 12, 295–336. provan, k. g., & huang, k. (2012). resource tangibility and the evolution of a publicly funded health and human services network. public administration review, 72(3), 366–375. provan, k.g., & kenis, p. (2007). modes of network governance: structure, management, and effectiveness. journal of public administration research and theory, 18(2), 229–252. provan, k. g., & lemaire, r. h. (2012). core concepts and key ideas for understanding public sector organizational networks: using research to inform scholarship and practice. public administration review, 72(5), 638–648. provan, k. g., & milward, h. b. (2001). do networks really work? a framework for evaluating public sector organizational networks. public administration review, 61(3), 414–423. provan, k. g., & milward, h. b. (1995). a preliminary theory of interorganizational network effectiveness: a comparative study of four community mental health systems. administrative science quarterly, 40(1), 1–33. rhodes, r., & williams, a. (1996). the new governance: governing without government. political studies, 44(4), 652–667. robins, g., bates, l., & pattison, p. (2011). network governance and environmental management: conflict and cooperation. public administration, 89(4), 1293–1313. article_14-3.indd 36 03/07/14 1:31 am n. kapucu / complexity, governance, and networks 37 rogers, e., & weber, e. p. (2010). thinking harder about outcomes for collaborative governance arrangements. the american review of public administration, 40(5), 546–567. sørensen, e. (2006). metagovernance: the changing role of politicians in processes of democratic governance. american review of public administration, 36(1), 98–114. sørensen, e., & torfing, j. (2009). making governance networks effective and democratic through metagovernance. public administration, 87(2), 234–258. span, k. c., luijkx, k. g., schols, j. m., & schalk, r. (2012). the relationship between governance roles and performance in local public interorganizational networks: a conceptual analysis. the american review of public administration, 42(2), 186–201. stoker, g. (2006). public value management a new narrative for networked governance? the american review of public administration, 36(1), 41–57. termeer, c. j. a. m., dewulf, a., breeman, g., & stiller, s. j. (2012). governance capabilities for dealing wisely with wicked problems. administration & society, 1–31. doi:10.1177/0095399712469195. wasserman, s., & faust, k. (1994). social network analysis: methods and applications. new york, ny: cambridge university press. zeemering, e. s. (2012). the problem of democratic anchorage for interlocal agreements. the american review of public administration, 42(1), 87–103. article_14-3.indd 37 03/07/14 1:31 am article_14-3.indd 38 03/07/14 1:31 am complexity in governance network theory complexity, governance & networks (2014) 61–70 61 doi: 10.7564/14-cgn8 complexity in governance network theory e. h. klijn* and j. f. m. koppenjan erasmus university rotterdam e-mail: klijn@fsw.eur.nl in this article, we discuss how complexity is viewed in governance network theory. the article provides a systematic elaboration of the notion of complexity, distinguishing three types: substantive, strategic , and institutional complexity. we argue that dealing with these types of complexity in networks is essentially a matter of mutual adaption and cooperation. an important explanation for the occurrence of deadlocks, breakthroughs and outcomes is the presence and the quality of attempts to manage complex interaction processes in networks. keywords: governance networks, substantive complexity, strategic complexity, institutional complexity, network management. 1. complexity in governance network theory one aspect that is clearly emphasized in most of the literature on governance networks is the complexity of the governance processes within these networks. we define governance networks as sets of autonomous yet interdependent actors (individuals, groups, organizations) that have developed enduring relationships in governing specific public problems or policy programs. the complexity of these networks is implied in one of the most important books on networks: hanf and scharpf’s interorganizational policy making: limits to coordination and central control (1978). the main argument throughout this seminal work, which is more frequently cited in europe than in the us, is that although the actors who deal with policy problems are interdependent to each other for resources, there usually are no governance structures set up that deal with these interdependencies. network theory did not start with the publication of hanf and scharpf’s book; it has a long tradition in both political science and (inter-)organizational science, which goes back to the early 1960s (for an overview, see klijn & koppenjan, 2012). hanf and scharpf were, however, among the first to use a network perspective to address the “wicked” character of policy problems—a term that was first used by rittel and webber (1973) and has been the most prominent concept in the network studies in the field of public administration. hanf and scharpf direct attention to where network theory, which at that time * corresponding author. article_14-8.indd 61 03/07/14 1:36 am 62 e. h. klijn et al. / complexity in governance network theory differed significantly from many other perspectives in public administration: that network theory tends to downplay the central control possibilities of policy and decision-making processes and instead emphasizes the complexity of these processes. in this article, we explore how complexity is conceptualized and analyzed in the network perspective and how it is used to understand and explain decision-making processes within networks. three types of complexity in governance networks complexity, most of the network literature emphasizes, is not simply caused by the fact that multiple actors are present within governance, although this is an important condition. the presence of diverse components in a system makes a system complicated, but not complex (gerrits, 2012). complicatedness can be tamed by the development of knowledge on these components and their relationships. complexity cannot be tamed, however. complexity reflects the dynamic nature of a system’s components and their relationships, making it very hard to predict how the system will behave and which outcomes will be produced (kickert, klijn, & koppenjan, 1997; koliba, meek, & zia, 2010; morçöl, 2012;, van buuren, & gerrits, 2009). governance networks are multi-actor systems that are not simply complicated, but complex. if we look at the wide literature on networks, and the literature that strongly influenced the network literature (for an extensive elaboration of the latter literature, see klijn & koppenjan, 2012), we can identify three types of complexity in these networks: substantive complexity, strategic complexity, and institutional complexity (koppenjan & klijn, 2004). 2. substantive complexity substantive complexity is about the content of the problem addressed and the nature of solutions under consideration. in most of the mainstream public administration literature, substantive complexity is first of all attributed to the lack of knowledge and information. complexity then is considered to be the result of the absence of data, or the absence of research or access thereto. this, so is often argued, is a result of the state of the scientific knowledge available at the time, and will be solved by further scientific research and development. the increase in the levels of co2 in the atmosphere causes the temperature on earth to rise (the greenhouse effect) but, as long as it is not known what specific concentrations have what kinds of effects, it is impossible to assess the magnitude of this rise in temperatures. substantive complexity, however, is not caused only by the absence of information and knowledge. what is often more important is that information is available, but its validity is contested. since actors have different perceptions of problems and view them from different frames of reference, they interpret the available information differently (schön & rein, 1994; fischer, 2003). the presence of different perceptions is really what makes policy problems wicked (rittel & webber, 1973; head, 2008; weber & khademian, 2008). “wicked problems” are article_14-8.indd 62 03/07/14 1:36 am e. h. klijn et al. / complexity in governance network theory 63 complex, not only because they are technical in nature, or involve many components or actors, but more so because the actors involved in them have different perceptions of their nature and its solutions. often wicked problems can be found in the areas of physical planning, environmental issues, and social problems (e.g. elderly care, social welfare issues, crime). the value differences among actors and the need for coordinated action among them make the policy problem complex (head, 2008; provan & kenis, 2008). the substantive complexity of wicked problems cannot be resolved by collecting more information, because this complexity is not caused by information shortages, but by the lack of a joint frame of reference and shared meaning among actors. as a result, new information can be interpreted in different ways. also, since all actors may engage in information gathering in their own way, information may be diverse, conflicting, or hard to understand, which may result in information overload and the articulation of conflicting truths. information gathering in order to deal with wicked problems in networks will not reduce complexity, but on the contrary, will contribute to it (klijn & koppenjan, 2004). 3. strategic complexity most network theorists would argue that governance networks are characterized by strategic complexity, in addition to substantive complexity (e.g. scharpf, 1978, 1997; agranoff & mcguire, 2003). strategic complexity is a result of the strategic choices actors make when they articulate complex problems (allison, 1971; kingdon, 1984). because actors are autonomous and networks lack clear hierarchical control forms, each actor choses his/her own strategy. as a result, various or even conflicting, strategies may develop around a complex issue (koppenjan & klijn, 2004). furthermore, actors anticipate each other’s strategic moves and respond to them (scharpf, 1997). because of these interactions, it is difficult to predict what strategies actors will choose, how strategies will evolve during the process, and how the interactions of these strategies will influence the process of problem-solving. complexity and indeterminacy characterize the interactions within governance networks. it is not easy to reduce or eliminate the strategic complexity that is created by these interactions. in a complex society characterized by interdependencies (castells, 1997), actors have discretion to make their own choices. unexpected strategic turns thus are an intrinsic characteristic of processes in networks. 4. institutional complexity governance networks are also characterized by institutional complexity. institutions can be defined as sets of rules regulating behavior (scharpf 1997; ostrom, 1990). networks are enduring relationships between actors that have resulted in the emergence of sets of rules that characterize these relationships. each network will have a unique set of rules. network rules may reduce complexity and enhance cooperation, since they article_14-8.indd 63 03/07/14 1:36 am 64 e. h. klijn et al. / complexity in governance network theory make the behaviors of actors more predictable. however network rules may compete with other sets of rules stemming from informal groups, specific professional roles, organizations, national laws, etc. (march & olsen, 1989; scott, 1995). if the number of rules grow, become inconsistent, opaque, and not well understood, they may generate complexity instead of predictability. what is more, wicked problems cut across existing demarcations between organizations, administrative levels, and networks. as a result, interactions become more difficult because the behavior of actors representing various networks will be guided by the different rules and frames of reference; they will have other routines and speak another professional language (ostrom, 2005; baumgartner & jones, 2009). 5. coping with complexity: the need for interaction network theorists suggest that public policymaking and service delivery in governance networks require coping with complexity, because problem definitions, solutions, and knowledge are contested, strategic interactions of actors are hard to anticipate. all of these factors lead to unpredictable outcomes, and different institutional regimes produce ambiguity about which rules to follow. the assumption that underlies network theory is that handling the complexity of difficult societal problems requires mutual adaption and cooperation among network actors (scharpf, 1978, 1997; marin & mayntz, 1991; rhodes, 1997; mandell, 2001). as a result of the resource dependencies among network actors, some actors may have opportunities to “veto” the decisions made by others. if actors do not negotiate their strategies with others and focus solely on the accomplishment of their own goals, by using so-called goalone strategies, these strategies are likely to lead to blockades and stagnation and hence to inefficient and ineffective decision making (koppenjan & klijn, 2004; agranoff & mcguire, 2003). go-alone strategies in networks generally lead to substantively poor and sub-optimal problem solving. they lead to poor solutions because knowledge is dispersed across many actors (head, 2007; hess & ostrom, 2006) and therefore these strategies result in solutions that fall substantively short in tackling complex problems (fischer, 2003). they lead to sub-optimal solutions because these strategies usually start with the goal of optimizing particular values, but often, multiple values are involved in decision making, and solutions must reflect this multiplicity of values (koppenjan & klijn, 2004). go-alone strategies tend to optimize particular values at the expense of others in the decision-making process and outcome and therefore contradict this multiplicity of values. cooperation is necessary in networks, but it is not easy and it comes at a cost. an interaction process may stagnate or be blocked because actors have different perceptions, they pursue conflicting interests or strategies, they lose interest in an issue, or the transaction costs of cooperation are too high (williamson, 1996). yet, because of their resource dependencies, actors who deal with wicked problems may not have any other choice but to keep on looking for ways to cooperate. article_14-8.indd 64 03/07/14 1:36 am e. h. klijn et al. / complexity in governance network theory 65 explanations of the processes and outcomes of interactions within governance networks due to their complexity, processes within networks aimed at dealing with wicked problems are characterized by stagnation and blockades. an important question, therefore, is how stagnations and blockage on the one hand, and breakthroughs and outcomes on the other, can be explained. in this section we group the explanations provided by the network literature into three categories: cognitive, social, and institutional explanations. in addition, we introduce a fourth explanation: the presence and quality of network management. 6. cognitive explanations: substantive complexity as cause the stagnation in a network may originate from the varying perceptions or frames about the nature, causes, and effects of problems and their solutions (fischer, 2003; hajer & wagenaar, 2003). there may be differences of opinion about the nature of a problem and the quality of the available knowledge and solutions. if actors are not aware of the fact that they have different perceptions , this may result in misinterpretations and discussion in which they talk past each other, without reaching an agreement. in extreme cases, this may result in an enduring “dialogue of the deaf” (wildavsky & tenenbaum, 1981; van eeten, 1999). if actors try to convince one another, and involve experts or commission research in order to strengthen their arguments, the cognitive differences may be deepened. under these conditions, debates on environmental problems or the construction of infrastructural projects, for instance, may result in a war of reports. scientific knowledge may serve to enhance knowledge conflicts and thus substantive complexity, instead of decreasing them (nowotny, scott, & gibbons, 2001; bijker, bal, & hendriks, 2009). a substantive breakthrough is required to break cognitive impasses. overcoming ambiguity, misunderstandings, and differences of opinion requires a convergence of ideas and perceptions and the development of a mutual understanding of situations and events. this calls for frame reflection and a cross-frame debate, in which problems and solutions are formulated anew (fischer, 2003; schön & rein, 1994; hajer & wagenaar, 2003). cognitive variety in terms of solutions, problem definitions, and the scope within which solutions are sought is an important precondition for frame reflection and learning. the turnover of key persons in networks and changes in actor constellations may result in social variation with similar effects (sabatier & jenkins-smith, 1993). another important factor that will help frame reflection and learning is the degree to which actors are able to include expertise and organize research in a way that helps to identify joint knowledge questions and supports the process of joint sense making and learning (hajer & wagenaar, 2003; head, 2008). this section shows that substantive complexity is not the same as the presence of a variety of perceptions. cognitive variety is actually a condition for substantive breakthroughs. it is the absence of a joint frame of reference and a mutual understanding that causes stagnations and deadlocks. article_14-8.indd 65 03/07/14 1:36 am 66 e. h. klijn et al. / complexity in governance network theory 7. social explanations: strategic complexity as cause deadlocks and stagnation emerge in networks when the strategies of those actors whose resources are indispensable for dealing with the problem are uncoordinated or in conflict, or when there is no interaction among actors. actors are often insufficiently aware of their mutual dependencies or they fail to discover a mutual interest. as a result, they choose go-alone strategies that bring them into conflict with one another. governance networks are not “cozy places of harmony”; they can be characterized by a high degree of conflict. a lack of dedication to solving a problem may also underlie stagnation. actors may simply not be interested in investing their resources. conflicts and stagnation may result in transaction costs and make it difficult to achieve coordination (williamson, 1996; huxham & vangen, 2005). a breakthrough can emerge if parties are able to align their strategies. for this to occur, strategic uncertainties must be reduced. this may be achieved by the formulation of process agreements with ad hoc rules that help to make the behavior of actors more predictables. furthermore, when one or more actors operate as brokers, facilitators, conflict managers, or arbiters, there is an increased chance of preventing, or at least limiting, the destructive influence of deadlocks, realizing breakthroughs, and making decisions (agranoff & mcguire, 2003; mandell, 2001). if actors succeed in formulating a proposal for a solution that links various objectives and offers the perspective of an improved situation for most of the actors, a strong incentive may be created to cooperate. a well-known example of such a win-win-solution in the netherlands is the breakthrough reached in an enduring conflict on the construction of a road within an urban area: the case of the sijtwende tunnel. the municipality opposed the road that was planned by the central government, since it wanted to develop real estate in the area. a private developer suggested that it be allowed to build a tunnel. in this case the extra costs for tunnel construction would be partly covered by a contribution paid from the profits of real estate development (koppenjan, 2005). our discussion in this section shows that the presence of many actors is not the same as social complexity. it is the lack of coordination mechanisms that causes stagnation and deadlocks. the presence of social variety is one of the conditions to realize breakthroughs. 8. institutional explanations: institutional complexity as cause stagnation or deadlocks in network processes can be caused by a weakly developed institutional structure, i.e. the absence of a clear set of mutually shared rules. mutually shared rules help to reduce the risks involved in participating in interactions in networks. they also often have a mitigating effect upon conflicts, and they provide procedures for enhancing interaction and managing conflict. a weakly developed institutional structure does not imply that there are no institutional rules. the problem is more one of institutional complexity, i.e. many different rules coming from various institutional backgrounds. for instance, attempts at building new integrated health care networks (like initiatives to reduce alcohol use among young people, article_14-8.indd 66 03/07/14 1:36 am e. h. klijn et al. / complexity in governance network theory 67 or to reduce the problem of overweight) encounter problems because the initiators have to deal with institutional rules from various policy sectors. the incompatibility of orientations, rules, and languages that guide the parties’ behaviors makes the process of dealing with wicked problems something like building the tower of babel (march & olsen, 1989; ostrom, 2005). institutional characteristics, such as the nature of rules and trust among actors, can affect stagnation and cooperation as well. cooperation is more difficult in networks with rigid rules than in networks with less rigid rules. various empirical studies have shown that a high level of trust in networks is related to better network performance, indicating that trust is a very important for cooperation in networks (provan, huang, & milward, 2009; klijn, edelenbos, & steijn, 2010). networks with a strong institutional structure, such as recognizable rules and relatively strong trust relations between actors, may result in lower transaction costs because provisions that further cooperation do not need to be developed from scratch, and parties can rely on existing arrangements. institutional breakthroughs occur when dysfunctional rules change or are replaced. as a result, new behaviors may emerge, which result in breakthroughs in interaction processes. also, the creation of new organizational structures and formal rules can enhance the cooperation between parties involved and thus, indirectly, help to realize a breakthrough. the creation of new rules that guide actors’ behaviors or the creation of more trust is not something that can be accomplished in the short run, however. institutional change is often the result of unintended developments or events. (ostrom, 2005; koppenjan & klijn, 2004). the uncertainty on institutional changes is a source of institutional complexity and a cause for stagnations in network processes itself. 9. network management as an explanation we propose an additional explanation for the occurrence of deadlocks, breakthroughs, and the emergence of policy outcomes in governance networks: the presence or absence of attempts to manage complex interaction processes in networks and the quality of these management efforts. network management can be seen as a way to address the complexity in networks. it may be focused on substantive complexity by enhancing variety, supporting learning processes, and arranging joint research. network management can address strategic complexity by initiating and strengthening interactions, arranging relationships, and mediating conflicts. network management may be used to deal with institutional complexity by engaging in institutional redesign (changing rules, resource distributions, and organizational relationships). because these three forms of complexity are interrelated, the efforts to manage them cannot be isolated from each other. a dialogue of the deaf (cognitive complexity) may be address by introducing a new actor (addressing strategic complexity). institutional complexity may be overcome by initiating process agreements (addressing strategic complexity). network management activities may have an incidental, ad hoc nature. it may be a onetime response of one of the actors to a deadlock. network management may also be a article_14-8.indd 67 03/07/14 1:36 am 68 e. h. klijn et al. / complexity in governance network theory continuous and planned activity. furthermore, network management may be in the hands of one actor, a governmental agency or another organization, but it may also be a role that alternates among actors. and, last but not least, network management may be contested— both in terms of who performs the function and what strategies are performed—thus adding to the complexity that characterizes governance networks, rather than reducing it. the number of the empirical studies on network management has increased significantly in the last 15 years. many cases studies explored network management strategies (mandell, 2001; sørensen &torfing, 2007). recently published survey studies show that the employment of network management strategies contributes to better network performance (meier & o’toole, 2007; provan, et al., 2009; klijn, steijn, & edelenbos, 2010; akkerman & torenvliet, 2011). 10. conclusions in this article, we have elaborated on the network perspective of complexity. in particular we identified three types of complexity in networks. we also showed how these types of complexities can be used as explanations for stagnations and breakthroughs in network processes. attempts at managing complexity do not automatically lead to solving wicked problems. they are, rather, focused on achieving conditions under which it becomes possible to deal with wicked problems. it involves enhancing learning processes between parties aimed at substance, process and institutions. network management is focused on improving cognitive learning processes, on joint image building, enrichment and goal entanglement, and on strategic and institutional learning processes, aimed at enhancing cooperation between parties with diverging interests, perceptions and objectives, roles (experts, citizens, users, civil servants, entrepreneurs, politicians) and diverging institutional backgrounds. references agranoff, r., & mcguire, m. (2003). collaborative public management: new strategies for local governments. washington, dc: georgetown university press. akkerman, a., & torenvlied, r. (2011). managing the environment effects of network ambition on agency performance. public management review, 13(1), 159–174. allison, g. t. (1971). essence of decision. boston, ma: little brown. ansell, c., & gash, a. (2008). collaborative governance in theory and practice. journal of public administration research and theory, 18(4), 543–571. baumgartner, f. r., & jones, b. (2009). agendas and instability in american politics 2nd ed. chicago, il: university of chicago press. bijker, w. e., bal, r., & hendriks, r. (2009). the paradox of scientific authority: the role of scientific advice in democracies. cambridge, ma: the mit press. fischer, f. (2003). reframing public policy: discursive politics and deliberative practices. oxford, uk: oxford university press. gerrits, l. (2012). punching clouds: an introduction to the complexity of public decision-making. litchfield park, az: emergent. article_14-8.indd 68 03/07/14 1:36 am e. h. klijn et al. / complexity in governance network theory 69 hajer, m., & wagenaar, h. (eds.). (2003). deliberative policy analysis: understanding governance in the network society. cambridge, ca: cambridge university press. hanf, k. i., & scharpf, f. w. (eds.). (1978). interorganizational policy making. london, england: sage. head, b. 2008. wicked problems in public policy. public policy, 3(2), 110–118. hess c., & ostrom, e. (eds.). (2006). understanding knowledge as a commons: from theory to practice. cambridge, ma: mit press. huxham, c., & vangen, s. (2005). managing to collaborate: the theory and practice of collaborative advantage. london, england: routledge. hwang, s., & moon, i. c. (2009). are we treating networks seriously? the growth of network research in public administration & public policy. connections, 29(2), 4–17. kickert, w. j. m., klijn, e. h., & koppenjan, j. f. m. (eds.). (1997). managing complex networks: strategies for the public sector. london, england: sage. kingdon, j. w. (1984). agendas, alternatives and public policies. toronto/boston, ma: little brown. klijn, e. h., & koppenjan, j. f. m. (2006). institutional design: changing features of networks. public management review, 8(1), 141–161. klijn, e. h., & koppenjan, j. f. m. (2012). governance network theory: past, present and future. policy and politics, 40(4), 187–206. klijn, e. h., edelenbos, j., & steijn, b. (2010). trust in governance networks: its impact and outcomes. administration and society, 42(2), 193–221. klijn, e. h., steijn, b., & edelenbos, j. (2010). the impact of network management strategies on the outcomes in governance networks. public administration, 88(4), 1063–1082. koliba, c., meek, j., & zia, a. (2010). governance networks in public administration and public policy. boca raton, fl: taylor & francis/crc press. koppenjan, j. f. m., & klijn, e. h. (2004). managing uncertainties in networks: a network approach to problem solving and decision making. london, england: routledge. koppenjan, j. f. m., & koliba, c. (eds.). (2013). transformations towards new public governance: can the new paradigm handle complexity? international review of public administration, 18(2), 1–84. koppenjan, j. f. m. (2005). the formation of public-private partnerships: lessons from nine transport infrastructure projects in the netherlands. public administration, 83(1), 135–157. kort, m., & klijn, e. h. (2011). public–private partnerships in urban renewal: organizational form or managerial capacity. public administration review, 71(4), 618–626. mandell, m. p. (ed.). (2001). getting results through collaboration: networks and network structures for public policy and management. westport, ct: quorum books. march, j. g., & olsen, j. p. (1989). rediscovering institutions: the organizational basis of politics. new york, ny: free press. marin, b., & mayntz, r. (eds.). (1991). policy networks: empirical evidence and theoretical considerations. boulder, co: westview press. meier, k., & o’toole, l. j. (2007). modelling public management: empirical analysis of the management– performance nexus. public administration review, 9(4), 503–527. morçöl, g. (2012). a complexity theory for public policy. new york, ny: routledge. nowotny, h., scott, p., & gibbons, m. (2001). rethinking science: knowledge and the public in an age of uncertainty. cambridge, uk: polity press. ostrom, e. (2005). understanding institutional diversity. princeton, nj: princeton university press. provan, k. g., & kenis, p. (2008). modes of network governance: structure, management and effectiveness. journal of public administration research and theory, 18, 229–252. provan, k. g., huang, k., & milward, b. h. (2009). the evolution of structural embeddedness and organizational social outcomes in a centrally governed health and human service network. journal of public administration research and theory, 19, 873–893. rhodes, r. a. w. (1997). understanding governance. buckingham, uk: open university press. rittel, h. j. w., & webber, m. m. (1973). dilemmas in a general theory of planning. policy sciences, 4, 155–169. article_14-8.indd 69 03/07/14 1:36 am 70 e. h. klijn et al. / complexity in governance network theory sabatier, p. a., & jenkins-smith, h. c. (1993). policy change and learning: an advocacy coalition approach. boulder, co: westview press. scharpf, f. w. (1978). interorganizational policy studies: issues, concepts and perspectives. in k. i. hanf & f. w. scharpf (eds.), interorganizational policy making (pp. 345–370). london, england: sage. scharpf, f. w. (1997). games real actors play. actor-centered institutionalism in policy research, boulder, co: westview press. schön, d. a., & rein, m. (1994). frame reflection: toward the resolution of intractable policy controversies. new york, ny: basic books. scott, w. r. (1995). institutions and organizations. thousand oaks, ca: sage. sørensen, e., & torfing, j. (eds.). (2007). theories of democratic network governance. cheltenham, uk: edward elgar. teisman, g. r., van buuren, a., & gerrits, l. (2009). managing complex governance systems. london, england: routledge. turrini, a., cristofoli, d., frosini, f., & nasi, g. (2010). networking literature about determinants of network effectiveness. public administration, 88(2), 528–550. van eeten, m. (1999) dialogues of the deaf: defining new agendas for environmental deadlocks. delft, the netherlands: eburon. weber, e., & khademian, a. (2008). wicked problems, knowledge challenges and collaborative builders in network settings. public administration review, 68(2), 334–349. wildavsky, a., & tenenbaum, e. (1981). the politics of mistrust: estimating american oil and gas resources. beverly hills, ca: sage. williamson, o. e. (1996). the mechanisms of governance. oxford, uk: oxford university press. article_14-8.indd 70 03/07/14 1:36 am from the habit of control to institutional enablement: re-envisioning the governance of social-ecological systems from the perspective of complexity sciences complexity, governance & networks (2014) 79–88 79 doi: 10.7564/14-cgn4 from the habit of control to institutional enablement: re-envisioning the governance of social-ecological systems from the perspective of complexity sciences asim ziaa*, stuart kauffmanb,c, chris kolibaa, brian beckaged, gabor vattaye, arne bombliesf a department of community development and applied economics, university of vermont, usa e-mail: asim.zia@uvm.edu b departments of mathematics and biochemistry, university of vermont, usa c department of signal processing, tampere university of technology, finland d department of plant biology, university of vermont, usa e department of physics of complex systems, eotvos university, hungary f department of civil and environmental engineering, university of vermont, usa due to the inherent uncertainty in predicting the evolution of phase-spaces in socialecological systems (sess), these systems cannot be “optimally” managed through top-down, command and control type of governance designs. instead, generalized autocatalytic set theory, a type of network and complexity theory with foundations in mathematical graph theory, may be used as a bottom-up, emergent and co-evolutionary framework to design the governance regimes of sess. under this theoretical re-conceptualization, the policy and institutional interventions can at best “enable” the policy-makers to nudge sess towards socially desirable yet ecologically feasible phase-spaces, which in turn are continually revamped as new elements in phase-spaces emerge. keywords: autocatalytic sets; adjacent possible; phase-spaces; co-evolution; governance design; policy regimes; global governance. 1. introduction the persistence of wicked policy problems (rittel & webber, 1973), coupled with minor and major instances of market failure, pose particular challenges to those looking to employ complexity science concepts to study and steer social, economic, and environmental systems (broadly defined as social-ecological systems sess in the tradition of ostrom, 2007). since the end of second world war, the top-down command and * corresponding author. article_14-4.indd 79 03/07/14 1:32 am 80 a. zia et al. / from the habit of control to institutional enablement control style management of global sess by international organizations has placed human civilization on an un-sustainable path. the unparalleled loss of biodiversity (sala  et  al., 2000), growing food insecurity (bohle, downing, & watts, 1994), rising in-equity in global distribution of wealth (palpacuer, 2008), and runaway human-induced climate change (stocker et al., 2013) are but few symptoms of this unsustainable pathway. it could be argued that since the dawn of the industrial revolution, sess have been mismanaged under the garbs of free market capitalism, benevolent colonialism, and the scientific management of global economic, ecological, and social capitals (zia, 2013). the current configuration of international organizations, which we define as a global governance regime, further reinforces this unsustainable pathway. worse, the newtonian conceptions of political economy (for examples, see romer, 1990) have incentivized the creation and sustenance of controversial international organizations such as world trade organization, world bank and international monetary fund, which perpetuate an “ optimization-envy” in the management of inherently complex sess. this optimization-envy is predicated on the newtonian conceptions of equilibrium in dynamic systems, which pivot the design of global governance regimes and international organizational structures to the sustenance of free markets; these conceptions maximize endogenous economic growth while neglecting mediumto long-term response of ecological and natural systems, such as global scale food insecurity induced by anthropogenic climate change and biodiversity loss. governance of sess is ultimately a complex game of balancing market values, such as maximizing global social welfare function through the promotion of free trade, with public values, such as ecological conservation. promoting shortto medium-term unfettered economic growth over long-term ecological conservation is an example of the “habits of control” that are widely observed in the current global governance and policy regimes. an unintended consequence of this top-down global governance design perspective is that global-scale ecological crises, such as biodiversity loss and climate change, have emerged as existential risks for human civilization; they cannot be effectively mitigated without regulating free markets that cause the ecological crises in the first place. sociologist max weber (1947) observed that with newton we became disenchanted and entered modernity. the “habit of control” is part of this newtonian perspective we still live with, and it fails us in a world where we cannot know ahead of time a sufficient understanding of what is probable or what is even possible (e.g. see kauffman, 1999, for a critique of the newtonian paradigm). we argue in this article that complexity science offers some critical insights into understanding the operating dynamics of sess and intra-societal dynamics shaping economic and political systems. those readers who are less familiar with the advances in complexity science are encouraged to read some of the major works in the field drawn from physics, biology, and computer science. recently complexity science has been applied to the study of markets (axtell, 2001), societies (epstein, 2006), public policy (morçöl, 2012), and governance networks (koliba, meek, & zia, 2010; teisman, van buuren, & gerrits, 2009). a variety of trade publications provided essential concepts, computational underpinnings, and major characteristics of complexity science (emergence, self-organization, article_14-4.indd 80 03/07/14 1:32 am a. zia et al. / from the habit of control to institutional enablement 81 etc.). of these essential concepts, we discuss one that became prevalent in recent decades: the theory of generalized autocatalytic sets (gacs). the gacs has been used to better understand the frontiers of certainty in complex systems. some of the key concepts from gacs theory are phase-spaces, the adjacent possible, co-evolution, and general autocatalytic sets. according to longo and montévil (2013), a “phase-space is the space of the pertinent observables and parameters in which the theoretical determination of the system takes place” (p. 64). changes in phase-spaces move systems through generative iterations of “adjacent possible.” kauffman (1999) defines “adjacent possible” as the successive evolutionary pathway in which autonomous agents move through a temporal sequence, reproducing themselves with mutations enabling their adaptation to the changing environments. in this context, some may be disappointed to learn that too little is known about the evolution of phase-spaces in the sess, primarily due to scientific inability to predict the nature of mutations and adaptations in the adjacent possible. this is a very humbling rebuke to the air of certainty that prevails in much newtonian science and the habits of control that has evolved with it. under the newtonian paradigm, the chains of causality, such as that c follows from b which follows from a, have been used to predict future phase-spaces of a system. in contrast, biological systems, including natural ecosystems, and social systems are relatively less predictable than physics-based systems, a case that has been made extensively elsewhere (beckage, gross, & kauffman, 2011; beckage, kauffman, zia, koliba, & gross, 2013; longo & montévil, 2011, 2013; zia, kauffman, & niiranen, 2012). the capacity of researchers to predict the future states of biological systems, or their sub-components, is diminished. this is because the nested hierarchical structure of biological functions— nucleotides forming dna, which form genes, which form cells, and so on—provides a virtually endless combinatorial variations that are sensitive to environmental conditions and respond to autonomous decision making by higher order organisms in the nested hierarchies of ecological systems. in social systems, the behaviors of autonomous agents, individual members of societies, are governed by constellations of rules, norms and shared strategies, which evolve over time (ostrom, 2005). we call this system of explicit and tacit rules, norms, and strategies that forge and sustain the material bonds that form families, communities, and organizations over multiple generations “institutions.” in modern societies, economic markets are examples of established institutions that govern the exchanges of goods, services, and resources. institutions evolve and the evolutionary patterns of the phase-spaces in these social systems are essentially stochastic and often unpredictable. in other words, they are “unprestatable” at longer timescales (longo, montévil, & kauffman, 2012). the unprestatability here refers to the theoretical inability to accurately identify the elements (observables and parameters) of future phase-spaces. the unpredictability and unprestatability in social systems arises not just from the internal dynamics of social systems (e.g. revolutions and technological advances), but also from the mutual interdependence of social systems with article_14-4.indd 81 03/07/14 1:32 am 82 a. zia et al. / from the habit of control to institutional enablement ecological systems (e.g. provision and regulation of food, air, and water based ecosystem services from ecosystems to sustain the populations, as defined in millennium ecosystem assessment (2005)). predicting the evolution of phase-spaces in sess is inherently beyond the capacity of the newtonian paradigm. in this article, we argue that the habit of top-down control in governing sess often fails because top-down management is predicated on the assumption of complete knowledge of phase-spaces. we posit that governance theorists need to look for bottom-up emergence; they should acknowledge the unpredictability in governance systems because of their co-evolutionary nature (kauffman, 1993; morçöl, 2012). co-evolution in complex systems typically involves mutual interdependence among their elements. a complexity science inspired form of governance could aim at enabling and coordinating generative co-evolutionary processes in sess, bearing in mind that when we enable by laws and institutional mechanisms, we cannot entirely pre-state the evolution of phase-spaces in social systems. hence, as new phase-spaces emerge, adaptive interventions to continuously refine and revise institutional mechanisms are always needed by societies. in this paper, we present a framework that is informed by complexity science. in this framework, we utilize the generalized autocatalytic set (gacs) theory to model the co-evolution of phase-spaces in complex sess (kauffman, 1986). within the gacs conceptualization, sess cannot be “optimally” managed due to the lack of determinism in predicting the evolution of their phase-spaces; instead, the policy and governance interventions can at best “enable” the policy-makers in sess to nudge the system towards socially desirable, yet ecologically feasible phase-spaces, which in turn are continually revamped as new elements emerge in phase-spaces. in the next section we briefly describe formal gacs theory. in the following section, we review recent studies and modeling efforts in applying gacs to model the evolution of phase-spaces in social, economic, and political systems. we conclude the paper by exploring the potential role of gacs theory in investigating the design and structure of governance regimes of sess from a complexity science perspective. 2. formal gacs theory arguably, gacs has played an important role in understanding the origin of life in this universe (see kauffman, 1993). gacs has its critics in the fields of molecular biology and genetics (e.g. lifson, 1997; maynard smith & szathmary, 1995). the critics did not accept the hypothesis that gacs can sufficiently solve the origin of life problem in the physical universe. this origin of life problem might not be resolved conclusively anytime soon; we leave it to biologists to debate. our interest in gacs is to apply it explaining the “emergence” of phase-spaces in the “adjacent possible” in sess. the formal mathematical and graph-theoretical concepts of gacs were developed by hordijk and steel (2004), hordijk, kauffman and steel (2011), kauffman (1986, 1993), mossel and steel (2005), and steel (2000). here we will present the formal definition of gacs from a biochemistry perspective (adapted from steel, 2000) and then describe the adaptations of gacs methodology in ses modeling applications. article_14-4.indd 82 03/07/14 1:32 am a. zia et al. / from the habit of control to institutional enablement 83 steel’s (2000, p. 92) version of gacs can be summarized as follows. let x denote a set of molecules. a reaction r will denote a pair r 5 ({a,b}, c), a,b,c px) which represents an allowable chemical reaction in both forward and backward directions: a1b ↔ c. let f (for “food”) denote a distinguished subset of x. let r be the set of allowable reactions. a catalyzation is a pair (x, r) where x p x, r p r, denoting that molecule x catalyzes reaction r. let c # x x r be a set of catalyzations. given the quadruple (x, f, r, c), a subset r` of r is reflexively autocatalytic (ra), if for all r p r`, there exists an s p supp (r`): (s, r) p c, connected to f if supp (r`) 5 clr`(f), and connected, reflexively autocatalytic (cra) if r` is both ra and connected to f. the clr`(f) represents the closure of x`(a subset of x) with respect to r` that satisfies the condition that for each reaction a1b ↔ c in r`: a,b p x` u w ⇒ c p w, and c p w ⇒ a, b p w, where w is a unique minimal subset of x. the formal gacs system described above captures “the abstract idea of “life” as a self-catalyzing system that is able to sustain itself by using a stable food source” (steel, 2000, p. 92). hordijk et al. (2011, p. 3) present a visual representation (reproduced here as figure 1) of the formal gacs system defined above for a simple example of gacs with seven molecule types {a, b, c, d, e, f, g} (solid nodes) and four reactions {r1, r2, r3, r4} (open nodes). the food set is f 5 {a, b}. solid arrows represent reactants going into and products coming out of a reaction, and dashed arrows represent catalysis. the subset r 5 {r1, r2}, shown with bold arrows in figure 1, is a cra (connected, reflexively autocatalytic) set. a generalized autocatalytic set (gacs) is a whole system that “gets to exist” in the emergent universe above the level of atoms, precisely because it is a self-reproducing whole. in self-reproduction processes, the generalized autocatalytic sets exhibit an important dynamic property. if we consider catalyzing a reaction a “catalytic task,” then the set as a whole achieves “task closure,” as formally defined above, which in turn enables the emergence of new properties of the system’s phase-space in the adjacent possible. this dynamic property of the gacs has been reproduced in earlier applications, such as caminati and stabile (2010) and padgett, lee and collier (2003)in social and economic systems, dittrich and winter (2008) in political systems, gabora (2004, 2011) in anthropology, and marion and uhl-bien (2001, 2003), and reschke and kraus (2009) in management sciences. a b d g e f c r1 r2 r3 r4 figure 1. a simple example of a gacs (reproduced from hordijk et al. (2011, p. 3)). article_14-4.indd 83 03/07/14 1:32 am 84 a. zia et al. / from the habit of control to institutional enablement 3. applications of gacs in social ecological systems padgett et al. (2003) developed an agent-based model (abm) called “hypercycle,” which used gacs to model the emergence of complex market structures from simple firm interactions, representing market competition, and their coevolution in the technology sector. the hypercycle abm has three components: rules (“skills”), balls (“products”), and bins (“firms”). inside the model, the rules/firms transform balls/products into other balls/ products. while the rules/skills are contained in bins/firms, they learn from each other the set of “transformative” rules/skills, also known as “technologies” in the hypercycle abm. products successfully transformed within the firm are passed randomly to one of the firm’s eight possible trading partners. if that firm possesses a compatible skill, it transforms the product further and passes that along in a random direction. firms continue passing around transformed products among themselves until the product lands on a firm that does not possess a compatible skill to transform it further. at that point, the product is ejected further into the environment, and a new iterative ball is selected to begin the iterative process again, representing the autocatalytic process of economic production and trading of goods, enabled by the coevolution of technologies and firms. this simple model reproduces some interesting properties of the complexity observed in real-world economic evolution of goods and services over time; however, the model has not been applied or calibrated to a specific market evolution yet. caminati and stabile (2010) took a data-mining approach to explore the notion of “modularity” and autocatalytic sets and “to identify the functional and structural units of an empirical knowledge pattern that defines the strongest systematic and self-sustaining mechanisms of knowledge transfer and accumulation within the network” (p. 365). caminati and stabile (2010) reconstructed the architecture of the empirical knowledge pattern based on uspto patent citation data from 1975-1999 and discovered that the recent progress towards catalyzing the information and communication technology revolution is marked by innovative solutions to “complementarities based on strong and mutual knowledge interactions between the different application-oriented and base components of the technology” (p. 394). dittrich and winter (2008) developed a toy model to simulate a hypothetical political system that is driven by autocatalytic set theoretical dynamics, as formally described in the previous section. the political system “consists of a list of molecules and a list of reaction rules (i.e. production rules). a molecule represents a specific communication, which in the political system is a decision” (p. 620). in dittrich and winter’s toy model, there are 13 different decisions and 20 reaction rules. the simulation model reproduces the evolving network of political organizations given the assumed decisions and reaction rules. they argue that the simulation methodology they propose permits mapping of real data to the set of political organizations and can help us understand the dynamic properties of the evolving political networks in future applications of gacs. gabora (2004) argues that ideas are not replicators in human cultural evolution, but minds are. she further argues that human culture is an “associatively-structured network” of ideas that together form an internal model of the world, or “world-view.” a world-view article_14-4.indd 84 03/07/14 1:32 am a. zia et al. / from the habit of control to institutional enablement 85 is a “primitive uncoded replicator, like the autocatalytic sets of polymers widely believed to be the earliest form of life” (p. 128). building on this work, gabora (2011) suggests that cultural evolution cannot be strictly explained by darwinian principles of natural selection and replicators, as dawkins (1976, 1982) did. instead, “what evolves through culture is worldviews [original italics], the integrated web of ideas, beliefs and so forth, that constitute our internal models of the world, and they evolve, as did early life, not through competition and survival of the fittest but through transformation of all” (p. 16). according to marion and uhl-bien (2001), autocatalytic interaction is the key to understanding effective complex organizational behavior, because “autocatalysis depends upon emergent distributed intelligence. . . . which cannot be directed but can be enabled by leaders” (p. 398). marion and uhl-bien (2003) applied the insights from the gacs theory to describe the emergence of al-qaeda leadership as an outcome of social dynamics in muslim societies. reschke and kraus (2009) apply gacs theory to analyze stability and change in social and cultural systems. for reschke and kraus (2009, p. 263), the relations in autocatalytic networks lead to “constrained probabilities” of interactions, which are more probable than the average unconstrained probability over all possible interactions. according to reschke and kraus (2009) “institutions came to be seen as sets of habits which form as autocatalytic sets of functional interactions. those elements that perform relatively better will be selected by inclusion, while others will fade into relative or absolute oblivion. stability and change depend on the network connections among entities in social systems, such as peer groups, organizations, and societies” (p. 266). 4. implications and conclusions the underlying graph theoretical and mathematical foundations of gacs theory provide powerful tools to formally study the governance of sess from the perspective of complexity science. if the governance of market and societal conflicts is mediated by nested hierarchies of institutionalized action arenas, also known as “polycentric governance regimes” (ostrom, 2005), a tweak in institutional rules in any node of this vast network of action arenas essentially catalyzes and enables new set of constraints and opportunities that ultimately affect the autocatalytic cycles of evolution in nested social ecological complexity of our planet. in the context of global governance of sess, the prevailing habits of control observed in the strategic and tactical behaviors of international organizations need to be re-considered. this reconsideration of the global governance regime from a complexity science perspective can potentially provide an alternative bottom-up pathway for enabling coevolution among various components of sess as new phase-spaces emerge over time. in the broad complexity science-informed context of governing sess, we propose articulation of a new research program to study institutional enablement that is predicated on an institutional framework facilitated by a gacs theoretical perspective on designing governance regimes. institutional enablement applies a complexity science approach to article_14-4.indd 85 03/07/14 1:32 am 86 a. zia et al. / from the habit of control to institutional enablement innovation and systems change. by viewing global sess as coupled human, biological and physical systems, we plan to explore the opportunities that complexity science and gacs theory could potentially contribute to the reconceptualization of a vision for understanding and steering governance regimes of sess. with this view, we seek to explore how policy and governance design changes could be treated as triggers. these triggers can in turn enable or disable the emergence of innovation in societies by affecting the enabling or governing variables that drive a system to stability or instability. the gacs theoretical perspective, with continual refinement, can open up new vistas of understanding in public management and policy sciences. it will require a sustained research program for many years and decades to test and demonstrate the viability of this theoretical perspective. the new journal, complexity, governance and networks, can be a great venue to advance the proposed research program, generating novel ways to understand governance networks from gacs and other generative theories of complexity science. acknowledgment we gratefully acknowledge funding from the national science foundation eps-1101317. references axtell, r. l. (2001). zipf distribution of us firm sizes. science, 293(5536), 1818–1820. beckage, b., gross, l., & kauffman, s. (2011). the limits to prediction in ecological systems. ecosphere, 2(11), 125. beckage, b., kauffman, s., zia, a., koliba, c., & gross, l. (2013). more complex complexity: exploring the nature of computational irreducibility across physical, biological, and human social systems. in h. zenil (ed.), irreducibility and computational equivalence: 10 years after the publication of wolfram’s a new kind of science (pp. 79–88). heidelberg, germany: springer verlag. bohle, h. g., downing, t. e., & watts, m. j. (1994). climate change and social vulnerability: toward a sociology and geography of food insecurity. global environmental change, 4(1), 37–48. caminati, m., & stabile, a. (2010). the pattern of knowledge flows between technology fields. metroeconomica, 61(2), 364–397. dawkins, r. (1976). the selfish gene. oxford, england: oxford university press. dawkins, r. (1982). the extended phenotype. oxford, england: oxford university press. dittrich, p., & winter, l. (2008). chemical organizations in a toy model of the political system. advances in complex systems, 11(4), 609–627. epstein, j. m. (ed.). (2006). generative social science: studies in agent-based computational modeling. princeton, nj: princeton university press. gabora, l. (2004). ideas are not replicators but minds are. biology and philosophy, 19(1), 127–143. gabora, l. (2011). five clarifications about cultural evolution. journal of cognition and culture, 11, 61–83. hordijk, w., kauffman, s. a., & steel, m. (2011). required levels of catalysis for emergence of autocatalytic sets in models of chemical reaction systems. international journal of molecular sciences, 12(5), 3085–3101. doi:10.3390/ijms12053085. hordijk, w., & steel, m. (2004). detecting autocatalytic, self-sustaining sets in chemical reaction systems. journal of theoretical biology, 227(4), 451–461. kauffman, s. a. (1986). autocatalytic sets of proteins. journal of theoretical biology, 119, 1–24. article_14-4.indd 86 03/07/14 1:32 am a. zia et al. / from the habit of control to institutional enablement 87 kauffman, s. a. (1993). the origins of order. new york, ny: oxford university press. kauffman, s. a. (1999). investigations. new york, ny: oxford university press. koliba, c., meek, j., & zia, a. (2010). governance networks in public administration and public policy. boca raton, fl: crc press. lifson, s. (1997). on the crucial stages in the origin of animate matter. journal of molecular evolution, 44, 1–8. longo, g., & montévil, m. (2011). from physics to biology by extending criticality and symmetry breakings. progress in biophysics and molecular biology, 106(2), 340–347. longo, g., & montévil, m. (2013). extended criticality, phase spaces and enablement in biology. chaos, solutions & fractals, 55, 64–79. longo, g., montévil, m., & kauffman, s. (2012). no entailing laws, but enablement in the evolution of the biosphere. proceedings of the fourteenth international conference on genetic and evolutionary computation conference companion, usa, 1379–1392. doi:10.1145/2330784/2330946. marion, r., & uhl-bien, m. (2001). leadership in complex organizations. the leadership quarterly, 12(4), 389–418. marion, r., & uhl-bien, m. (2003). complexity theory and al-qaeda: examining complex leadership. emergence: complexity and organization, 5(1), 54–76. maynard smith, j., & szathmary, e. (1995). the major transitions in evolution. oxford, england: oxford university press. millennium ecosystem assessment. (2005). ecosystems and human well-being. washington, dc: island press. morçöl, g. (2012). a complexity theory for public policy. london, england: routledge. mossel, e., & steel, m. (2005). random biochemical networks: the probability of self-sustaining autocatalysis. journal of theoretical biology, 233, 327–336. ostrom, e. (2005). understanding institutional diversity. princeton, nj: princeton university press. ostrom, e. (2007). a diagnostic approach for going beyond panaceas. proceedings of the national academy of sciences, 104(39), 15181–15187. padgett, j. f., lee, d., & collier, n. (2003). economic production as chemistry. industrial and corporate change, 12(4), 843–877. palpacuer, f. (2008). bringing the social context back in: governance and wealth distribution in global commodity chains. economy and society, 37(3), 393–419. reschke, c. h., & kraus, s. (2009). an evolutionary perspective on the management of stability and change. evolutionary and institutional economic review, 5(2), 259–278. rittel, h. w. j., & webber, m. (1973). dilemmas in a general theory of planning.policy sciences, 4, 155–169. romer, p. m., (1990) endogenous technological change. journal of political economy, 98(5), s71–s102. sala, o. e., chapin, f. s., armesto, j. j., berlow, e., bloomfield, j., dirzo, r., . . . wall, d. h. (2000). global biodiversity scenarios for the year 2100. science, 287(5459), 1770–1774. steel, m. (2000). the emergence of a self-catalysing structure in abstract origin-of-life models. applied mathematics letters, 13(3), 91–95. stocker, t., qin, d., plattner, g. k., tignor, m., allen, s. k., boschung, j., . . . midgley, p. m. (2013). ipcc 2013: summary for policy makers. climate change. new york, ny: cambridge university press. teisman, g., van buuren, a., & gerrits, l. m. (2009). managing complex governance systems. london, england: rutledge. weber, m. (1947). the theory of social and economic organization. london, england: routledge & kegan paul. zia, a. (2013). post-kyoto climate governance:confronting the politics of scale, ideology and knowledge. london, england: routledge. zia, a., kauffman, s., & niiranen, s. (2012). the prospects and limits of algorithms in simulating creative decision making. emergence: complexity and organization, 14(3), 89–109. article_14-4.indd 87 03/07/14 1:32 am article_14-4.indd 88 03/07/14 1:32 am solving institutional collective action problems in multiplex networks complexity, governance & networks (2014) 49–60 49 doi: 10.7564/14-cgn7 solving institutional collective action problems in multiplex networks manoj shresthaa, ramiro berardob, richard c. feiockc* a university of idaho b university of wisconsin milwaukee c florida state university askew school, 626 bellamy building, tallahassee fl, 32306-2250 e-mail: rfeiock@fsu.edu this essay extends theoretical arguments pertaining to single (uniplex) networks on how to solve coordination and cooperation problems associated with institutional collective action to multiplex networks constituting both formal and informal relationships formed by policy actors. while coordination problems reflect difficulties for actors in arriving at jointly desired policy outcomes, cooperation problems mean that actors have conflicting interests and, thus, face incentives to defect on each other. we propose multiplex versions of bridging and bonding networks, which have been found suitable for solving coordination and cooperation problems in single networks. although our approach is limited to the simultaneous analysis of formal and informal relationships in policy networks, the arguments we present should aid researchers interested in analyzing policy networks beyond these manifestations of inherently complex relationships. keywords: multiplex networks, institutional collective action problems, collaborative governance. 1. introduction scholars have long recognized the multidimensionality of social systems in which large numbers of social agents are interconnected by a wide range of social relationships. these relationships include friendship, professional ties, exchanges of resources, such as information, goods and services, etc. while each type of relationship in these multiplex networks can be studied in isolation, a comprehensive analysis of a social system is incomplete without examining how the different types of relationships affect each other (robins & pattison, 2006; bae & feiock, 2012). studying multiplex networks are particularly important for solving institutional collective action dilemmas arising from the fragmentation of political and administrative authority in contemporary societies (feiock, 2013). fragmentation makes governance * corresponding author. article_14-7.indd 49 03/07/14 2:20 am 50 m. shrestha et al. / solving institutional collective action problems inherently complex, since a broad range of public, non-profit, and private organizations create a variety of formal and informal relationships in the management and delivery of public goods and services (kettle, 2000; feiock, 2009; feiock & scholz, 2010). interdependencies among these organizations produce the need for coordinated behavior, which can be more easily accomplished in networks with certain features facilitating such behavior. formal relationships can be defined in statutes, which prescribe what a policy network should look like, or in contractual agreements that define the terms of relationships between two or more institutions (edelenbos & klijn, 2006; klijn & koppenjan, 2000; mcguire, 2002; meier & o’toole, 2003; milward & provan, 2006). but beyond the formal relationships, there are also informal links that may be formed between institutions for a variety of reasons, including similar ideological positions or policy views and agreements on organizational goals. network structures emerge from these relationships that can shape the performance and stability of decision-making systems thus affecting the design and implementation of public policies (feiock & scholz, 2010). for example, in metropolitan service provision, local governments maintain contractual relationships with their peers for multiple services (shrestha & feiock, 2009; shrestha, 2010; andrew, 2009), but they may also be connected to each other through informal ties when they exchange information or meet one another in regional decision-making venues. considerable progress has been made in the policy sciences in understanding single types of relationships among actors in isolation, either formal (andrew, 2009; minkoff, 2012, 2013; post, 2004; shrestha, 2010;) or informal (andrew & carr, 2012; feiock, 2009, 2013; feiock, lee, & park, 2012; lazer, 2011; lee, feiock, & lee, 2012; schneider, scholz, lubell, mindruta, & edwardsen, 2003; scholz, berardo, & kile, 2008). much less progress has been made in understanding the inherently multiplex character of these networks as they operate together. as a result, important questions remain unanswered. perhaps the most basic one is: how do actors shape their formal and informal relationships with each other when they face different types of institutional collective action (ica) dilemmas? our aim in this essay is to start a scholarly conversation that can lead to finding precise answers to this question. we believe this is an important endeavor because of the pervasiveness of ica dilemmas and the ever-growing need to better understand the rather complex ways in which policy actors relate to each other when trying to solve those problems. this essay proceeds first with a description of institutional collective action (ica) problems. then it discusses how those problems can be tackled by utilizing bridging and/ or bonding capital in multiplex networks. next, it introduces propositions for specific types of bonding and bridging configurations that one should expect to see when policy actors face coordination and cooperation problems. finally, the essay highlights the limitations of and opportunities for analyzing multiplex networks. 2. ica dilemmas and networks institutional collective action dilemmas arise from the fragmentation of authority in governance systems where policy decisions made by one jurisdiction to pursue its goals article_14-7.indd 50 03/07/14 2:20 am m. shrestha et al. / solving institutional collective action problems 51 are at odds with the activities or the collective benefits of the affected jurisdictions. these dilemmas are likely to manifest in the form of either coordination or cooperation problems. the former occurs when actors such as local governments have similar goals but disagree on how to reach them. for example, in metropolitan areas where multiple local governments coexist, shared interests in managing traffic flow to avoid congestion and gridlock lead cities to coordinate road construction, lane closures and the timing of traffic signals. cooperation problems, on the other hand, take place when actors have conflicting goals and they are likely to defect on each other in order to reach those goals. for instance, cities trying to increase their tax base by attracting new businesses and investments may engage in offering competitive concessions that may lead to collective loss to the cities in the region. studies show that policy actors develop both formal and informal relationships to mitigate these types of problems (feiock, 2013; feiock & scholz, 2010; lubell, schneider, scholz, & mete, 2002). for instance, local jurisdictions can become members of regional economic development partnerships in order to coordinate their economic development decisions (feiock, chen, & hseih, 2016). similarly, local governments can create networks of formal service contracts to minimize the risk of potential defection associated with the difficulty of monitoring contracts (andrew, 2009; shrestha, 2008; shrestha & feiock, 2013). in addition, representatives of the jurisdictions (either elected or appointed) can build mutual trust through informal exchange of information or meeting at shared venues in order to prevent potential non-compliance of the mutually agreed terms of contracts that are not easily verifiable. unfortunately, the specialized literature in public administration and policy has lagged behind in examining how these formal and informal relationships among policy actors help solve ica problems. as a result, most analyses examining either formal or informal relationships have the obvious limitation of producing an oversimplified view of how collaborative processes really work. 3. solving coordination and cooperation problems in multiplex networks coordination and cooperation problems pose different challenges for policy actors in fragmented governance systems. in the presence of coordination problems, actors are interested in achieving a goal that they collectively deem valuable. this means that solving coordination problem requires that the actors exchange information in a way that allows them to converge on preferred courses of actions. cooperation problems, on the other hand, are characterized by conflicting interests that can lead to widespread defection. in this type of scenario, actors require information that can help them detect and punish defection, which contributes to sustaining cooperation. previous research ties the solution to these types of problems to the existence of particular network configurations that can be formed in networks. berardo & scholz (2010), for instance, contend that actors facing coordination problems in networks seek to engage in structural relationships that give them quick access to relevant information on how article_14-7.indd 51 03/07/14 2:20 am 52 m. shrestha et al. / solving institutional collective action problems others are likely to act. using data collected in self-organizing policy networks in 10 us estuaries, they show that bridging structures that are likely to facilitate quick convergence of policy views emerge when coordination problems are dominant. on the other hand, when actors face cooperation problems, they are more prone to form bonding structures where redundant information flows among the members of the structure. this, in turn, helps promote and sustain cooperation because defection becomes costly as actors in the network can detect “cheaters” and inform others about their behavior (coleman, 1988; putnam, 2000). studying bridging and bonding structures is relatively straightforward in uniplex networks, where only one type of relationship is examined. in multiplex networks, however, it is somewhat difficult to determine how actors form bridging and bonding structures conducing for solving coordination and cooperation problems. table 1 contains structural configurations that we believe can be used to examine bridging and bonding in multiplex networks involving formal (black lines) and informal (red lines) relationships. configuration 1a represents bridging through the establishment of multiplex in-stars. configurations 1b to 1f represent bonding through the establishment of dyadic multiplex reciprocity (figure 1b) and four types of closed structures that indicate bonding where one of the links is informal and the other is formal (or vice versa). table 1 structural configurations for solving coordination and cooperation problems in multiplex networks type of ica problem (and configurations that solve them) description illustrative structural configurations coordination problems (solved through bridging configurations) multiplex in-stars (1a) multiplex reciprocity (1b) multiplex transitive closure (1c) cooperation problems (solved through bonding configurations) multiplex cyclic closure (1d) multiplex closure for shared out-ties (1e) multiplex closure for shared in-ties (1f) note: black lines represent formal relations and red lines represent informal relations among actors. article_14-7.indd 52 03/07/14 2:20 am m. shrestha et al. / solving institutional collective action problems 53 we borrow these configurations from the research on exponential random graph models applied to multiplex networks (e.g. koehly & pattison, 2005; lomi & pattison, 2006; pattison & wasserman, 1999; wang, 2013; wang, robins & pattison, 2009). this set of configurations should not be thought of as an exhaustive group of structures that indicates bonding or bridging; it is, rather, a straightforward way of showcasing the concepts in a simple multiplex setting (simple in the sense that there are only two types of relationships).1 next, we discuss how actors form bridging and bonding configurations that are linked to solving coordination and cooperation problems in multiplex networks involving formal and informal relationships. 4. solving coordination problems through bridging in multiplex networks we contend that actors facing coordination problems may benefit from creating multiplex in-stars. an in-star represents bridging because there is a central actor in the network that indirectly connects other actors (nodes) that would be completely disconnected in its absence. the central (most popular) actor in the network functions as a depository of information and experience from both its formal and informal connections to others and is well positioned to control how those resources flow between other actors linked to it. for example, in metropolitan service delivery, counties and cities that are parties to multiple service contracts are able to gain economies of scale and economies of scope in production by entering into service contracts (or exchanging information informally) with a central city in the network that is heavily involved in exchanging information with other jurisdictions (or in entering service contracts with them). attracting many service contracts means that the popular provider can consolidate the demand and coordinate the use of production and managerial inputs in service production leading to gains in economies of scale and economies of scope for all cities involved. in addition, informal exchange of information can lead to finding ways of enhancing economies of scale and scope benefits for all parties. because of the central position in the multiplex in-star configuration, the popular provider is also likely to be perceived as credible and competent, which is a critical piece of information for other nodes when they are looking for potential partners (gulati & gargiulo, 1999). 5. solving cooperation problems through bonding in multiplex networks actors facing cooperation problems can create bonding network structures that facilitate the flow of overlapping information among the participants. this, in theory, may enable them to detect and punish defective behavior thereby ensuring credibility of commitment necessary to achieve cooperation. this reasoning is directly applied to uniplex 1 the prevalence of these structures in multiplex networks can be modeled with the xpnet software available at http://sna.unimelb.edu.au/pnet. article_14-7.indd 53 03/07/14 2:20 am 54 m. shrestha et al. / solving institutional collective action problems networks where the links between the actors contain information about each other’s behavior, but the argument can be expanded to the study of multiplex networks composed of both formal and informal types of links. we first describe a simple version of bonding—multiplex reciprocity—followed by more complex versions that are defined by the existence of closure structures involving a larger number of actors. 6. multiplex reciprocity creating reciprocity across ties of different types, or multiplex reciprocity, is one way to mitigate problems of credibility of commitment and decrease the likelihood of defection. multiplex reciprocity occurs when two actors create a reciprocal relationship formed by two types of ties, as shown in figure 1b. an example of multiplex reciprocity could be city a supplying police services to city b (a formal tie), while city b supplies city a with information on how to improve its fire department capabilities (an informal tie). in multiplex networks where exchanges involve multiple resources, there are more ways in which organizations can create multiplex reciprocity than in networks where organizations are linked by only one type of relation. since these exchanges are maintained through mutually supportive relationships, organizations tend to balance their relations by entertaining exchange relations across different networks (lomi & pattison, 2006). multiplex reciprocity creates mutual commitment by both parties and thereby reduces the risk of defection. reciprocity also helps partners to know each other better which develops into mutual trust and improves collaboration by reducing behavioral uncertainties (uzzi, 1997). it can develop into expectations of taking turns, mutual obligation, and trust in exchange (coleman, 1988). this notion is consistent with the social capital argument that actors involved in a variety of cross-cutting venues are more likely to create mutually trusting environment for improved cooperation (putnam, 2000). reciprocity can also incentivize parties to fulfill their commitment as they can punish each other by declining to cooperate in the next round. the presence of “mutual deterrence” (williamson, 1981), in reciprocal relations also forces both parties to build strong, credible ties. in the case of service contractual ties susceptible to opportunism, shrestha and feiock (2009) found that local jurisdictions in florida adopted cross-service reciprocity because it provided assurance to both the recipient (buyer) and the supplier (seller) jurisdictions of the credibility of commitment in service transaction. this suggests that policy actors facing cooperation problems in institutional collective action situations are more likely to develop multiplex reciprocity to address the problems. 7. multiplex triadic closure while reciprocity is dyadic, closure in social networks is a group feature. the main idea underlying social closure is that actors have a higher tendency to connect with each other when they happen to share links with other actors in the network (snijders, pattison, article_14-7.indd 54 03/07/14 2:20 am m. shrestha et al. / solving institutional collective action problems 55 robins, & handcock, 2006; robins, pattison, & wang, 2009). formed by path-shortening strategies or a direct link between unconnected actors sharing a common partner, such a tendency for closure prevents defection more effectively because the parties can better monitor each other’s behavior, resolve conflicts through consultations, and impose more effective sanctions on wrong doers (coleman, 1988). thus, multiplex closure reduces uncertainty and sustains cooperation in exchange. multiplex closure also promotes credible behavior by contributing to building trust and group norms to ensure credible commitments (putnam, 2000). in the context of project implementation networks, for example, projects that are supported by a more cohesive group of organizations possessing various resources are found to be more successful than projects that are supported by a less cohesive group of organizations because the projects supported by a cohesive group of organizations are able to secure the credible commitment of all the partners (shrestha, 2013). it is plausible to expect greater prevalence of multiplex closure when two actors engage into multiple relationships with multiple shared partners. multiplex closure can take different forms. as shown in figure 1c, multiplex transitivity in social networks occurs where actors have a tendency to close open twopath configuration because of potential defection by the intermediaries. for example, in intergovernmental fiscal relations, local governments generally maintain direct lobbying or information sharing ties with federal agencies as transfer of federal dollars to local governments through state agencies can be subject to manipulations by the states. on the other hand, when every actor in a group is interlocked with its partner via different ties, a multiplex cyclic closure of generalized exchange type is formed (figure 1d). multiplex closure for shared out-ties (figure 1e) emerges when two structurally equivalent, yet unconnected, cities form closure networks because both share common partners to whom they provide resources. finally, multiplex closure for shared in-ties (figure 1f) is formed when two unconnected, structurally equivalent cities create closure networks because both share common partners from whom they receive resources. in both cases, the closure can be achieved by sharing information between the previously unconnected, structurally equivalent actors. each of these closure mechanisms promotes cooperation by creating social trust and thereby preventing defection. hence, in general, actors motivated to ensuring credibility of commitment of partners in ica situations should be more likely to develop multiplex closed, bonding structures. 8. challenges multiplex network analysis poses challenges for scholars at both theoretical and methodological fronts. theoretically, we need a richer conceptualization of coordination and cooperation mechanisms in ica settings that can guide our thinking of what structural configurations actors should seek in each type of situation. further effort is needed to understand how networks are shaped when these problems coexist, as it is rare that a governance system faces one, but not the other, type of problem. do actors attend to all article_14-7.indd 55 03/07/14 2:20 am 56 m. shrestha et al. / solving institutional collective action problems problems simultaneously, thus creating both bridging and bonding structures that coexist with each other? or do they attack problems sequentially, prioritizing one type of relationship over another? if the former is true, are there specific “mixed” configurations that are likely to help achieve both coordination and cooperation? if problems are addressed sequentially, however, what affects the decisions of the actors to prioritize their problems, and how does this affect the sustainability of collaborative practices? producing answers to these questions will require the concerted efforts of public policy and management scholars. there are also important methodological challenges to be faced. even with just two types (formal and informal) of ties, it becomes complex to sort out the various combinations of ties as dependency considerations move from dyadic to triadic and higher levels. the choice of a particular effect will require well-developed theory explaining structural mechanisms for addressing coordination and cooperation problems. while there has been considerable progress in modeling dependencies in multiplex networks (for examples of multivariate exponential random graph model specifications, see pattison & wasserman, 1999; wang, 2013), such modeling is mostly restricted to simple configurations involving only two types of ties. the approach delineated here can also be applied to multiplex networks of multiple formal networks or multiple informal networks. for example, a city can enter into more than one formal service contracts with another city. likewise, informal ties between two cities could consist of exchange of information and attending the same professional meetings. in other words, ties have weights too, and this introduces another layer of complexity which needs to be dealt with. finally, for reasons of space we have only covered in slight detail how the different configurations indicate bridging and bonding, but more work is needed to clarify exactly in what circumstances actors in a network would prefer to form, say, transitive triadic relationships instead of cyclic triads. the drivers of behavior in these networks will probably be highly contextual, depending not only on the problems that the nodes face, but also on the nature of the links the researcher is interested in studying. scholars interested in pursuing any of these ideas must be fully aware of these limitations. 9. conclusion in this essay, we made a quick case for simultaneous analysis of multiple networks for better understanding of how actors embedded in multiple networks are able to address coordination and cooperation problems in ica situations. in this regard, we highlighted the origin and challenges of coordination and cooperation problems in ica settings, and discussed how those problems can be addressed by utilizing bridging and/or bonding capital in multiplex networks. notwithstanding the theoretical and methodological challenges, noted above, we believe that the multiplex bridging and bonding network structures advanced in this essay article_14-7.indd 56 03/07/14 2:20 am m. shrestha et al. / solving institutional collective action problems 57 for solving coordination and cooperation problems serve as a step forward towards more general treatment of embedded relationships in situations where ica dilemmas take place. because of our focus on multiplex networks, we did not highlight the potential effect of single network structures on the multiplex network structures. we also limited ourselves to the discussion of the emergence of multiplex network structures, rather than how these networks impact the outcomes. investigating the impacts of multiplex networks opens a whole new avenue for research with implications on the management of networks. the current advancement on both theoretical and methodological fronts offers opportunities for scholars to develop and test theory explaining the emergence of multiplex network structures and their impacts. this process should pave the way forward in advancing the complexity science for better understanding of how actors self-organize to achieve coordination and cooperation in real social networks that are multiplex in nature. we urge scholars to join in this collaborative endeavor. references andrew, s. a. (2009). regional integration through contracting networks: an empirical analysis of institutional collection action framework. urban affairs review, 44(3), 378–402. andrew, s. a., & carr, j. b. (2012). mitigating uncertainty and risk in planning for regional preparedness: the role of bonding and bridging relationships. urban studies, 40(4), 709–724. bae, j., & feiock, r. c. (2012). managing multiplexity: coordinating multiple services at a regional level. state and local government review, 44(2), 162–168. berardo, r., & scholz, j. t. (2010). self-organizing policy networks: risk, partner selection and cooperation in estuaries. american journal of political science, 54(3), 632–649. coleman, j. s. (1988). social capital in the creation of human capital. american journal of sociology, 94, 95–120. edelenbos, j., & klijn, e. h. (2006). managing stakeholder involvement in decision making: a comparative analysis of six interactive processes in the netherlands. journal of public administration research and theory, 16(3), 417–446. feiock, r. c. (2009). metropolitan governance and institutional collective action. urban affairs review, 44(3), 356–377. feiock, r. c. (2013). the institutional collective action framework. policy studies journal, 41(3), 397–425. feiock, r. c., chen, s., & hseih, j. y. (2016). regional partnerships and metropolitan economic development. journal of urban affairs february. feiock, r. c., lee, i. w., & park, h. j. (2012). administrators’ and elected officials’ collaboration networks: selecting partners to reduce risk in economic development. public administration review, 72(51), 558–568. feiock, r. c., & scholz, j. t. (eds.) (2010). self-organizing federalism: collaborative mechanisms to mitigate institutional collective action. cambridge, ca: cambridge university press. feiock, r. c., steinacker, a., & park, h. j. (2009). institutional collective action and economic development joint ventures. public administration review, 69(2), 256–270. gulati, r., & gargiulo, m. (1999). where do interorganizational networks come from? american journal of sociology, 104(5), 1439–1493. kettle, d. f. (2000).the transformation of governance: globalization, devolution, and the role of government. public administration review, 60(6), 488–497. klijn, e. h., & koppenjan, j. f. m. (2000). public management and policy networks: foundation of a network approach to governance. public management review, 2(2), 135–158. article_14-7.indd 57 03/07/14 2:20 am 58 m. shrestha et al. / solving institutional collective action problems koehly, l. m., & pattison, p. (2005). random graph models for social networks: multiple relations or multiple raters. in p. j. carrington, j. scott & s. wasserman (eds.), models and methods in social network analysis (pp. 162–191). cambridge, ca: cambridge university press. lazer, d. (2011). networks in political science: back to future. ps: political science and politics, 44(1), 61–68. lee, i. w., feiock, r. c., & lee, y. (2012). competitors and cooperators: a micro-level analysis of regional economic development collaboration networks. public administration review, 72(2), 253–262. lomi, a., & pattison, p. e. (2006). manufacturing relations: an empirical study of the organization of production across multiple networks. organization science, 17(3), 313–332. lubell, m., schneider, m., scholz, j., & mete, m. (2002). watershed partnerships and the emergence of collective action institutions. american journal of political science, 46, 148–163. mcguire, m. (2002). managing networks: propositions on what managers do and why they do it. public administration review, 62(5), 426–433. meier, k. j., & o’toole, l. j., jr. (2003). public management and educational performance: the impact of managerial networking. public administration review, 63(6), 689–699. milward, h. b., & provan, k. g. (2006). a manager’s guide to choosing and using collaborative networks. washington, dc: ibm center for the business of government. minkoff, s. l. (2012). the proximate polity: spatial context and political risk in local developmental goods provision. urban affairs review, 48(3), 354–388. minkoff, s. l. (2013). from competition to cooperation: a dyadic approach to interlocal agreements. american politics research, 41(2), 261–297. pattison, p., & wasserman, s. (1999). logit models and logistic regressions for social networks: ii. multivariate relations. british journal of mathematical and statistical psychology, 52, 169–193. post, s. (2004). metropolitan area governance and institutional collective action. in r. c. feiock (ed.), metropolitan governance: conflict, competition, and cooperation (pp. 67–92). washington, dc: georgetown university press. putnam, r. d. (2000). bowling alone: the collapse and revival of american community. new york, ny: simon and schuster. robins, g., & pattison, p. (2006). multiple networks in organisations. retrieved from http://www.sna.unimelb .edu.au/publications/publications.html. robins, g., pattison, p. e., & wang, p. (2009). closure, connectivity and degree distributions: exponential random graph (p*) models for directed social networks. social networks, 31(2), 105–117. schneider, m., scholz, j., lubell, m., mindruta, d., & edwardsen, m. (2003). building consensual institutions: networks and the national estuary program. american journal of political science, 47, 143–158. scholz, j., berardo, r., & kile, b. (2008). do networks enhance cooperation? credibility, search, and collaboration. journal of politics, 70(2), 393–406. shrestha, m. k. (2008). decentralized governments, networks and interlocal cooperation in public goods supply. phd dissertation, florida state university. shrestha, m. k. (2010). do risk profiles of services alter contractual behavior? a comparison across multiple metropolitan services. in r. c. feiock & j. scholz (eds.), self-organizing federalism: collaborative mechanisms to mitigate institutional collective action (pp. 114–141). new york, ny: cambridge university press. shrestha, m. k. (2013). self-organizing network capital and the success of collaborative public programs. journal of public administration research and theory, 23(2), 307–329. shrestha, m. k., & feiock, r. c. (2009). governing us metropolitan areas: self-organizing and multiplex service networks. american politics research, 37(5), 801–823. shrestha, m. k., & feiock, r. c. (2013). institutional collective action dilemmas in service delivery: exchange risks and contracting networks among local governments. paper presented at pmrc conference, madison, wi. article_14-7.indd 58 03/07/14 2:20 am m. shrestha et al. / solving institutional collective action problems 59 snijders, t. a. b., pattison, p., robins, g. l., & handcock, m. (2006). new specifications for exponential random graph models. sociological methodology, 36(1), 99–153. uzzi, b. (1997). social structure and competition in interfirm networks: the paradox of embeddedness. administrative science quarterly, 42, 35–67. wang, p. (2013). exponential random graph model extensions: models for multiple networks and bipartite networks. in d. lusher, j. koskinen & g. robins (eds.), exponential random graph models for social networks: theory, methods, and applications (pp. 115–129). cambridge, ca: cambridge university press. wang, p., robins, g., & pattison, p. (2009). pnet: program for the simulation and estimation of exponential random graph (p*) models (user manual). melbourne, australia: university of melbourne. williamson, o. e. (1981). the economics of organization: the transaction cost approach. the american journal of sociology, 87(3), 548–577. article_14-7.indd 59 03/07/14 2:20 am article_14-7.indd 60 03/07/14 2:20 am the emergence of complexity in the art and science of governance complexity, governance & networks (2014) 17–28 17 doi: 10.7564/14-cgn2 the emergence of complexity in the art and science of governance1 geert teisman*, lasse gerrits erasmus university rotterdam e-mail: teisman@fsw.eur.nl; gerrits@fsw.eur.nl we argue that the complexity of an interconnected society and its governance require a complexity-informed approach to our domain. concepts and theories from the complexity sciences can help with this. there is a notable increase of the use of such concepts and theories but the theory transfer isn’t as straightforward as it may seem. “emergence” is helpful in understanding the particular differences between the various realms of science. within the social sciences in general and public administration in particular, emergence highlights the nondecomposable, contingent, non-compressible and time-asymmetric nature of reality. subsequently, we propose three methods that take these aspects into account when putting concepts from the complexity sciences to the test: qualitative comparative analysis, dynamic network analysis, and group model building. keywords: complexity, governance, emergence, research methods, public administration, theory transfer. 1. complexity in network societies and governance networks in this paper, we elaborate how the notions of complexity are penetrating the practice and the science of public administration. we see a growing attention for complexity science and postulate that this attention is powered by the potential of the complexity sciences on the one hand, and by an increase of complexity in society and government on the other hand. this urges scholars in public administration to develop theories and concepts that fit with this complexity and that help to understand government actions in complex systems. as with many other new paradigm shifts in public administration in the past – the emergence of network theory, just to mention a relative recent one – the application and adoption of ‘the new kid in town’ is not without questions and doubt. pollitt, by invitation, formulated his critique in teisman, van buuren, & gerrits (2009), who presented 1 we are grateful to göktuğ morçöl for his thorough review of an earlier version of this article. * corresponding author. article_14-2.indd 17 03/07/14 1:31 am 18 g. teisman et al. / the emergence of complexity their insights about government systems from a complexity theoretical paradigm. these critiques are important because they address the things that a framework under development still lacks. it inspires the authors of this contribution to continue with their attempts to make complexity theory useful for public administration and to present evidence for the added value of this paradigm. we believe that the emergence of network societies, in which global networks of important local hubs are interrelated and in interaction, contribute to societal development and, consequently, to the emergence of governance theories that deliver the pertinent insight that effective steering comes from joint actions – whether in competition, cooperation of both – than from one single organization that is presumably in charge. this is an important driver in the field of public administration to apply complexity concepts and theories. the complexity sciences have the potential to obtain a more sophisticated and more dynamic understanding of the pattern of interdependencies underlying the kind of situations practitioners find themselves confronted with. however, the aim of this contribution is not to outline main features of complexity sciences and concepts from complexity theory like co-evolution and initial conditions. for these discussions on complexity sciences in relation to public administration we can refer to previous studies (gerrits, 2012; morçöl, 2012, 2014 [in this issue]; teisman & klijn, 2008; teisman, et al., 2009). in this contribution we will turn towards the art of research in the domain of public administration from a complexity theoretical approach. we will elaborate the ability of public administration scholars to do complexity-informed research. complexity-informed research is able to deal with blurred issues that are not easily definable and demarcated, with large amounts of data that are not coherent and countable in a direct sense and with relations that reach beyond simple takes on causality. the challenge is to gain scientific and transparent insights from a variety of messy data, delivered by a variety of contributors and sources. theory transfer and complex causality are the two sensitizing concepts we use in our search into complexity-informed research techniques, methods and methodology. theory transfer is the art of constructing a cross-over between two different disciplines, sometimes far away from each other. in that cross-over, the theory is applied in a new context. by doing so the theory also can be reshaped and show new advantages. this is comparable to commercial cross-overs in which materials used in a technical context get a new application and life once applied in a social context. here we elaborate the theory transfer of complexity-informed research and research methods. 2. simple behavior emerges into complex systems to many academics in public administration, waldrop’s (1992) book provides a first introduction to complexity sciences, in particular to the works of scientists of the famous santa fe institute, including john holland, brian arthur, and murray gell-mann. waldrop’s accessible narrative presents the contributions of these scientists in a heroic fashion against a backdrop of scientific breakthroughs. it is commonly agreed that the scientists at santa fe institute have played a major role in the emergence of what has become article_14-2.indd 18 03/07/14 1:31 am g. teisman et al. / the emergence of complexity 19 known as complexity theory, or more precisely, the complexity sciences as a collection of theories that focus on a number of related phenomena in physical and social systems, such as emergence, non-linearity, self-organization and coevolution. the theories developed at the santa fe institute are characterized by an algorithmic approach. the core idea is that complex patterns emerge from simple (behavioral) rules. as an example consider the “boids simulation,” which was developed by reynolds (1987). boids simulate flock-like behavior of e.g. herds of cattle and schools of fish. many scientists are intrigued by the dynamic shapes of these phenomena and by the fact that these “boids” seem to have a life on their own that is rather complex and dynamic. an important aspect of these simulations is that simple behavioral rules can generate complex, seemingly “living”, systems, and that these complex systems are able to deal with challenges. when released in a simulated environment with a number of obstacles, the boids are able to maintain a coherent mutual structure whilst avoiding those obstacles when moving about. the simulation indicates that one does not need to design and superimpose complex rules for individual member of the boids to maintain a coherent – though not fixed structure under changing circumstances. the resulting patterns resemble the “flocking behavior” of birds, cattle and fish and demonstrate how complex wholes can emerge from simple individual action. the emergence of complex patterns and behaviors reynolds demonstrated were observed by e.g. axelrod (1986, 1997) and holland (1995, 2006) as well, who use these ideas to understand the complexity of human cooperation. holland investigated and conceptualized the mechanisms of emergence and pointed them out as the origins of complexity. the concept of emergence opens new avenues of inquiry. in particular, it promises social scientists that there is order behind the ostensible chaos in network societies and it promises public administration scientists that there is order behind the ostensible chaos of numerous government actions and changes of course. the message here is that the individual behavior has a non-linear relationship to the whole in which the individual operates. in other words: what happens on the micro-level is related to the macro-level, but not to the extent that individual actions are fully mirrored in the whole. as society becomes more fragmented, specialized and compartimentalized, it means that the ostensible disconnect of the relationship between the two levels becomes more prominent. it may further obscure that relationship but it is still there and deserves a concerted research effort. emergence, then, can help explaining the complex causal patterns social scientists are faced with. it draws attention to the fact that social reality is a living, organic and dynamic systemic whole comprised of individual behaviors. this whole cannot substantially be understood if scientists only study its discrete parts without understanding the systemic whole in which they are embedded. when taken apart in discrete elements, the integrity as a system of ongoing interaction will be violated (gerrits & verweij, 2013), attempts to understand this multilevel phenomena are being made, but not very successful because the connections are made conceptually but are notoriously hard to trace empirically. simulations, for instance, demonstrate diachronic emergence where there is a clear starting point article_14-2.indd 19 03/07/14 1:31 am 20 g. teisman et al. / the emergence of complexity and clear final result. in other words: emergence here is seen as something that has an unambiguous t 5 0. in social reality, however, there is obviously no such thing as a clear starting point where everything is fixed until some activates ‘time’. and as any scholar in public administration knows: an outcome – whether from policy or any other type of decision – is a far from clear. in other words: our object of research is characterized by synchronic emergence where the temporal division between parts and whole is brought back into one single but continuous instant (elder-vass, 2005). simplistic methods will simply not deliver here. the similarities and differences between algorithmic emergence and emergence in social reality are a marker of the similarities and differences between the natural sciences, from which complexity theories originate, and the social sciences. adaptations to the social sciences, in particular to our domain, initially provided a vocabulary and set of concepts that opened up new avenues of thought. for example, consider how teisman (1992) analyzed the decision-making processes of infrastructure projects and found that such decision-making, whilst moving forward in time, does not adhere to the assumptions of synoptic, diachronic processes and that none of the actors involved were ever able to play a constantly decisive role. yet, decision structures emerged over time and projects were built. a thoroughly empirical-based understanding of decision-making has gained ground and the complexity sciences provided new ways in which such dynamics could be understood. in other words, the complexity sciences provided a boost to theory development in our domain. early adopters of complexity theory in public administration are haynes (2001, 2003), kickert (1991), kiel (1989), and morçöl and dennard (1997). broadly speaking, we can distinguish between two types of contributions. kiel and kickert attempted to replicate the scientific approach in the source domain of complexity sciences in the target domain of public administration. they tried to find the same mechanisms of emergence and self-organization in biological systems in the world of public administration, public policy, or public management. the other group of authors used the general ideas and concepts from the source domain as points of departure and translated them for the target domain. they did not merely copy the concepts, but transformed them to make them applicable in the target domain. they applied the original concepts as metaphors to make sense of the complexity of the target domain. for example, they may draw parallels between lorenz’ butterfly effect and the actions of a public manager or they may modify kauffman’s fitness landscape to suit the daily reality of said manager (see e.g. geyer & rihani, 2010; stacey & griffin, 2006). these differences are testimony of the dynamics of theory transfer between the physical sciences from which the complexity sciences originate, and the social sciences. it raises questions such as whether it is possible to replicate methods from the natural sciences in the social sciences and whether a causal patterns established in e.g. physics or biology still holds true in social reality. upon closer examination, it seems that a considerable number of applications in our domain use concepts from the complexity sciences as a metaphor. metaphors can provide genuine insight in the target domain but may lead to article_14-2.indd 20 03/07/14 1:31 am g. teisman et al. / the emergence of complexity 21 disappointment if not applied properly (see e.g. chettiparamb, 2006; lakoff & johnson, 2003; rosenhead, 1998, for extended arguments). despite initial interest, there have therefore also been scientists who have questioned the use of the endeavor, especially given the difficulties of theory transfer. for example, pollitt (2009) states that the complexity sciences promise more than that they deliver, in particular that authors wrap fancy jargon around trite issues such as policy failures and span of control. the difficulties in applying the complexity sciences to the social realm, the liberal use of metaphors and analogies and the well-worded criticism could make the endeavor appear a passing fad. more recent works (e.g. gerrits, 2012; morçöl, 2012; room, 2011a, 2011b; teisman, et al., 2009) take into account the issues of theory transfer, operationalization of main concepts, and the use of complexity-informed methods and demonstrate how complexity theory can lead to a better understanding of the messy day-to-day reality of policymakers and administrators. over time, it has become clearer how the complexity sciences can inform public administration. for example, researchers demonstrated how complex adaptive systems can lease a second life to easton’s cybernetics (flood, 1999), how original idea of punctuated equilibrium (eldredge & gould, 1972) can be operationalized and analyzed to understand the dynamics of the policy agenda (baumgartner & jones, 1993), and how evolutionary theories can inform the analysis of institutions (nelson & winter, 1982). it is justified to say that conceptual leaps have been made and that we are now better at making sense of public administration and public management. it means that we have become better at theory transfer. but concepts need thorough testing and for that we need complexity-informed methodological tools. as argued above, not every method is well-suited for dealing with the synchronic emergence that drives social reality. in the following section, we want to highlight the importance of the complexity sciences in understanding the nature of causality in the social sciences. 3. uncovering complexity in social reality one of the most important recent developments in the realm of social complexity is the understanding that theoretical concepts, such as emergence and self-organization, are basically expressions of the nature of causality. we follow byrne (1998, 2005, 2011) and take his concept of “situated complexity” as our point of departure in understanding social reality. byrne suggests that social reality is a compound of generic or recurring patterns in conjunction with idiosyncratic events, i.e. events that are local in place and temporal in time. as such, social reality is not exclusively governed by universal laws, but each event is not unique either. causal patterns are comparable across cases, yet at the same time, they differ with unique elements generated by unique contexts or circumstances (buijs, eshuis, &, byrne, 2009). accepting this ontological point of departure means that we embrace the classic scientific idea of science that lessons can learned from cases in the (recent) past into cases of the (near) future (the art of generalization). at the same time, we accept the idea that certain events from the past recur only occasionally. they are adjusted by, and combined article_14-2.indd 21 03/07/14 1:31 am 22 g. teisman et al. / the emergence of complexity with, unique contextual variables. science can still be based on the evidence collected in the past, but one cannot learn from this evidence simply by copying it; the knowledge of the past must be adjusted to fit new circumstances. this is the creative and transformative element of using ‘best practices’. more specific, a public administration theory rooted in synchronic emergence should be based on the postulate that social reality is non-decomposable, contingent, non compressible, and time-asymmetric (gerrits & verweij, 2013). the non-decomposable nature of reality has in fact been discussed above and implies that breaking reality into discrete elements for the sake of analysis violates its interrelatedness. important external forces would be at the risk of being overlooked. reality functions as a whole and this means that science is challenged to treat it as such. contingency refers to the fact that context is explanatory for which mechanisms are actually triggered and which are not (e.g. mjøset, 2009). it highlights that complex systems are neither governed by general rules nor by pure idiosyncracies, as pointed out by byrne above. in other words: context determines equifinality and multifinality. taking these two points together, it is clear that reality is nondecompressible because that would make it loose some of its decisive aspects. time asymmetry, the fourth point, holds that complex systems are developmentally open because of the occurrence of chance events leading to non-linear change (prigogine, 1997). in other words, trajectories of the past will not be mirrored in the present and the future. this calls for a longitudinal and detailed approach to research, as the emergence of structures and processes cannot be deduced in a linear fashion (gerrits, 2011, 2012). inevitably, any scientific method or model is reductionist (cilliers, 2002, 2005a, 2005b). as researchers struggle to understand the system as a whole, respecting its integrity, they also try to keep research manageable. although a truly holistic research project is not possible, there still is an important difference between accepting a method’s limitations and accepting ockham’s razor (gerrits, 2012). 4. examples of complexity-informed methods the four ontological points above set the coordinates to guide the selection of suitable, complexity-informed methodologies for doing empirical research in our domain. here, we want to focus on three interesting methods: qualitative comparative analysis (qca), dynamic network analysis (dna), and group model building (gmb). qualitative comparative analysis (ragin, 1987, 2000, 2008) was proposed by byrne (2005, 2011) and byrne and ragin (2009) as a particularly effective method in uncovering situated complexity because it preserves the contextual details of cases through systematic and transparent comparative procedures. as such, it negates the trade-off between rich single in-depth case studies and the less-detailed case comparisons that can reveal recurring patterns across those cases. in other words, it helps gaining a better understanding of the issue of contingency mentioned above. for example, verweij and gerrits (2014) compared 18 cases of infrastructure project implementation in the netherlands to understand which management strategies led to article_14-2.indd 22 03/07/14 1:31 am g. teisman et al. / the emergence of complexity 23 satisfaction about the projects among all actors involved in them. using qca, we found that there were different pathways to similar outcomes. we found that a general pattern that showed that under diverse circumstances, an externally-oriented project management approach in which public and private actors cooperate, leads to high satisfaction with the projects’ outcome. but under certain specific conditions and depending on the nature of cooperation between principal and contractor, an internally-oriented management approach was also associated with high satisfaction. in addition we found that such an internally-oriented approach could also lead to dissatisfaction under specific different circumstances (verweij & gerrits, 2014). in short, we found that there are several contingent pathways towards the outcomes “high satisfaction” and “low satisfaction” with the projects’ results. as such, the authors have demonstrated equifinality in the management of infrastructure projects by focusing on how outcomes – note the plural here – are produced in conjunction with the specific conditions of a certain phenomenon. dynamic network analysis (dna; carley, diesner, reminga, & tsvetovat, 2007) is an extension of social network analysis. it builds on the assumption that actors operate in networks, and adds that such networks are dynamic in composition and ties and that these dynamics can be captured by juxtaposing actor-networks with others, such as knowledge networks and resource networks. this juxtaposition of different types of networks in dna is based on the “affiliation networks” concepts and methods developed by social network analysis researchers. dna can also capture the state of the overall network at several points in time (carley et al., 2007). although not mentioned explicitly in the literature on dna, it embodies the nature of complex adaptive system by introducing time and feedback loops in the analysis in terms of actions and responses. as such, it enables to researchers to investigate structure and process simultaneously in a rigorous way. schipper and gerrits (2014) deployed dna in an investigation of the daily operations of the dutch railway system (schipper & gerrits, 2012, 2014). while the network of train operators, train dispatcher, signalers, etc. is quite capable in dealing with most of the common disruptions in the railway system, it fully fails when trying to cope with major events such as heavy snowfall or technical failures in rolling stock. in such cases, decisions made by those actors cascade through the network, where each decision reinforces the effect of the previous decision to the extent that the railway system as a whole collapses (leveson, dulac, marais, & caroll, 2009). using dna, the researchers were able to reconstruct the information flows in the system, i.e. which decisions were taken and how others responded to that, the tasks that were executed and how these changed over time, and the cognitive workload associated with this. as such, the researchers obtained a systemic overview of the railway system during a disruption and were able to trace the systemic properties back to the individual actions, i.e. emergence. among the methods discussed here, group model building (gmb) has been around for the longest time. gmb originated from the studies in system dynamics modeling ( majone & quade, 1980; quade, 1975). while early iterations of systems theories appeared too mechanistic, they have been improved continuously, particularly business administration and organizational sciences. checkland (1981) and flood (1999) developed article_14-2.indd 23 03/07/14 1:31 am 24 g. teisman et al. / the emergence of complexity increasingly sophisticated conceptualizations and methods for understanding system dynamics that take into account semiotics, learning, and sense-making. gmb represents one of such iterations (vennix, 1996). it combines the modeling of a particular policy problem with participatory group learning and a sense-making process among stakeholders. as such, it provides researchers with a tool to discover the causal linkages and feedback loops of a given issue, whilst at the same time giving stakeholders the opportunity to learn about the real causes and consequences of that particular issue. in an application of gmb in two projects with the rotterdam municipal authorities (vaandrager, gerrits, & bressers, 2014), it was found that the respondents sought the causes of the policy deadlocks outside their realm of control. they pointed towards politicians or to other organizations and felt themselves as victims of the circumstances. by jointly building a system’s model of that particular policy deadlock, they started to see that they were dealing with a systemic issue that was badly understood because their own views and needs where firmly bounded by the organizational interests, which made them feel helpless while they were in fact able to change the way things were being done. they gained this insight through joint model-building and visual representations of the models made by the researchers. these visualizations helped to capture the complex systemic dynamics, consequently leading to a change in the mental modes of the participants and offering sweet spots in the system where policy intervention was possible. the methods acknowledging complexitymentioned here are a selection from a broader suite of methodologies that range from narrative analysis (see e.g. uprichard & byrne, 2006; wagenaar, 2007 for examples) to computation modeling (see e.g. catlaw & kim, 2012; koliba & zia, 2012 for examples, and morçöl, 2012, for more extended overview). increasingly, accessible software programs allow researchers to unlock the potential of such methods and to deal with the high number of observations necessary to gain a deeper understanding of the complexity of public policy issues. in addition, such programs can visualize this complexity, which is very helpful in communications. for example, vensim provides a comprehensive yet accessible toolbox to create systemic visualization of the system of stocks and flows. ora does the same for the combined networks in dna, and tosmana for qca visualizes the solution formula with maximum five independent conditions. we want to stress the importance of visualization because it really helps to convey the intricate details found in the analysis. it is our experience that many people, especially practitioners, understand complexity in an intuitive way but a concerted research effort is necessary to uncover the real complex causality underlying social phenomenon and to communicate about them with scientists and practitioners. 5. conclusions the networked society is characterized by self-generating mutual interdependencies between actors. networks challenges the 21st century’s organizing principle of hierarchical bureaucracies. this is reflected in public administration in the shift from government article_14-2.indd 24 03/07/14 1:31 am g. teisman et al. / the emergence of complexity 25 to governance. similarly, the increasing complexity of society needs to be reflected in contemporary research in our domain. the complexity sciences offer a promising conceptual and methodological suite to approach complexity. in this contribution, we looked at the aspects of theory transfer and complex causality. we found that theories in the source domain can be very helpful in the target domain, i.e. public administration, but that such a transfer needs to be tailored to the specifics of social reality. the concept of emergence makes this point most prominently. subsequently, we postulated four aspects of emergence to which methodologies should adhere, and we discussed three suitable research methods. qca focuses on the nested nature of social reality, dna focuses on the dynamics of connections in networks, and gmb combines a participatory approach with identification of the leverage points and feedback loops that build the system. naturally, the proof of the pudding is in the eating and it is therefore necessary to put these methods and the concepts to the (empirical) test. this journal will provide an excellent outlet for that. references axelrod, r. (1986). an evolutionary approach to norms. the american political science review, 80(4), 1095–1111. axelrod, r. (1997). the complexity of cooperation: agent-based models of competition and collaboration. princeton, nj: princeton university press. baumgartner, f. r., & jones, b. d. (1993). agendas and instability in american politics. chicago, il: the university of chicago press. buijs, m. j., eshuis, j., & byrne, d. s. (2009). approaches to researching complexity in public management. in g. r. teisman, m. w. van buuren, & l. m. gerrits (eds.), managing complex governance systems:  dynamics, self-organization and coevolution in public investments (pp. 37–55). new york, ny: routledge. byrne, d. s., & ragin, c. c. (2009). the sage handbook of case-based methods. london, england: sage. byrne, d. s. (1998). complexity theory and the social sciences: an introduction. london, england: routledge. byrne, d. s. (2005). complexity, configurations and cases. theory, culture & society, 22(5), 95–111. byrne, d. s. (2011). applying social science: the role of social research in politics, policy and practice. bristol, england: the policy press. carley, k. m., diesner, j., reminga, j., & tsvetovat, m. (2007). toward an interoperable dynamic network analysis toolkit. decision support systems, 43(4), 1324–1347. catlaw, t.j., & kim, y. (2012). philosophies of multiplicity and simulation: possibilities for public affairs inquiry. in l. m. gerrits, & p. marks (eds.), compact i: public administration in complexity (pp. 26-40). litchfield park, az: emergent. checkland, p. (1981). systems thinking, systems practice. chichester, england: john wiley & sons. chettiparamb, a. (2006). metaphors in complexity theory and planning. planning theory, 5(1), 71–91. cilliers, p. (2002). why we cannot know complex things completely. emergence, 4(1/2), 77–84. cilliers, p. (2005a). complexity, deconstruction and relativism. theory, culture and society, 22(5), 255–267. cilliers, p. (2005b). knowledge, limits and boundaries. futures, 37(7), 605–613. elder-vass, d. (2005). emergence and the realist account of cause. journal of critical realism, 4(2), 315–338. eldredge, n., & gould, s. j. (1972). punctuated equilibria: an alternative to phyletic gradualism. in t. j. m. schopf (ed.), models in paleobiology (pp. 82–115). san fransisco, ca: cooper & co. flood, r. l. (1999). rethinking the fifth discipline: learning within the unknowable. london, england: routledge. article_14-2.indd 25 03/07/14 1:31 am 26 g. teisman et al. / the emergence of complexity gerrits, l. m. (2011). a coevolutionary revision of decision making processes: an analysis of port extensions in germany, belgium and the netherlands. public administration quarterly, 35(3), 309–339. gerrits, l. m., & verweij, s. (2013). critical realism as a meta-framework for understanding the relationships between complexity and qualitative comparative analysis. journal of critical realism, 12(2), 166–182. gerrits, l. m. (2012). punching clouds: an introduction to the complexity of public decision-making. litchfield park, az: emergent. geyer, r., & rihani, s. (2010). complexity and public policy: a new approach 21st century politics, policy and society. new york, ny: routledge. haynes, p. (2001). complexity, quantification and the management of policy. social issues, 1(2), 1–12. haynes, p. (2003). managing complexity in the public services. maidenhead, england: open university press. holland, j. h. (1995). hidden order: how adaptation builds complexity. jackson, ms: perseus books. holland, j. h. (2006). studying complex adaptive systems. journal of systems science and complexity, 19(1), 1–8. kickert, w. j. m. (1991). complexiteit, zelfsturing en dynamiek. rotterdam, nl: erasmus universiteit rotterdam. kiel, l. d. (1989). nonequilibrium theory and its implications for public administration. public administration review, 49(6), 544–551. koliba, c., & zia, a. (2012). theory testing using complex systems modeling in public administration and policy studies: challenges and opportunities for a meta-theoretical research program. in l. m. gerrits, p. m. marks & f. a. a. boons (eds.), public administration in complexity. litchfield park, az: emergent. lakoff, g., & johnson, m. (2003). metaphors we live by. chicago, il: chicago university press. leveson, n., dulac, n., marais, k., & carroll, j. (2009). moving beyond normal accidents and high reliability organizations: a systems approach to safety in complex systems. organization studies, 30(2–3), 227–249. majone, g., & quade, e. s. (eds.). (1980). pitfalls of analysis. chichester, england: wiley. mjøset, l. (2009). the contextualist approach to social science methodology. in d. s. byrne & c. c. ragin (eds.), the sage handbook of case-based methods (pp. 39–68). london, england: sage. morçöl, g. (2012). a complexity theory for public policy. new york, ny: routledge. morçöl, g (2014). complex governance networks: an assessment of the advances and prospects. complexity, governance and networks, 1, 5–16. morçöl, g. & dennard, l. f. (1997). learning from the natural sciences: an introduction. administrative theory & praxis, 19(3), 299–304. nelson, r. r., & winter, s. g. (1982). an evolutionary theory of economic change. cambridge, england: harvard university press. pollitt, c. (2009). complexity theory and evolutionary public administration: a skeptical afterword. in g. r. teisman, m. w. van buuren, & l. m. gerrits (eds.), managing complex governance systems: dynamics, self-organization and coevolution in public investments (pp. 37–55). new york, ny: routledge. prigogine, i. (1997). the end of certainty: time, chaos, and the new laws of nature. new york, ny: free press. quade, e. s. (1975). analysis for public decisions. new york, ny: american elsevier. ragin, c. c. (1987). the comparative method: moving beyond qualitative and quantitative strategies. los  angeles, ca: university of california press. ragin, c. c. (2000). fuzzy-set social science. chicago, il: university of chicago press. ragin, c. c. (2008). redesigning social inquiry: fuzzy sets and beyond. chicago, il: university of chicago press. reynolds, c. w. (1987). flocks, herds, and schools: a distributed behavioral model. computer graphics, 21(4), 25–34. room, g. (2011a). complexity, institutions and public policy: agile decision-making in a turbulent world. cheltenham, england: edgar elgar. room, g. (2011b). social mobility and complexity theory: towards a critique of the sociological mainstream. policy studies, 32(2), 109–126. rosenhead, j. (1998). complexity theory and management practice. (scientific paper no. lseor 98.25). london, england: department of operational research, london school of economics. article_14-2.indd 26 03/07/14 1:31 am g. teisman et al. / the emergence of complexity 27 schipper, d. & gerrits, l. (2012). on the origin of the mra. a coevolutionary analysis of the life-and-death cycles of urban governance in the amsterdam metropolitan region. aspa annual work conference: las vegas (march 2 march 6, 2012). schipper, d., & gerrits, l. m. (2014). an exploration of the possibilities of using dna in analyzing complex system disruptions. paper presented at egos 2014, rotterdam. stacey, r. d., & griffin, d. (eds.). (2006). complexity and the experience of managing in public sector organizations. new york, ny: routledge. teisman, g. r. (1992). complexe besluitvorming, een pluricentrisch perspectief op ruimtelijke investering ((netwerken, complexiteit en dynamiek)). den haag, nl: elsevier. teisman, g. r., & klijn, e. h. (2008). complexity theory and public administration; an introduction. public management review, 10(3), 287–297. teisman, g. r., van buuren, m. w., & gerrits, l. m. (eds.). (2009). managing complex governance systems: dynamics, self-organization and coevolution in public investments. new york, ny: routledge. uprichard, e., & byrne, d. (2006). representing complex places: a narrative approach. environment and planning a, 38(4), 665. vaandrager, d., gerrits, l. m., & bressers, n. (2014). reconstructing the blue connection: a systemic inquiry into the improbable project of building a new river through a (sub)urban area. systems research and behavioral science. [accepted] vennix j. a. m. (1996). group model building, facilitating team learning using system dynamics. chichester, england: john wiley & sons. verweij, s., & gerrits, l. m. (2014). how satisfaction is achieved in the implementation phase of large transportation infrastructure projects: a qualitative comparative analysis into the a2 tunnel project. public works management and policy, doi:10.1177/1087724x13518743. wagenaar, h. (2007). governance, complexity and democratic participation: how citizens and public officials harness the complexities of neighborhood decline. the american review of public administration, 37(1), 17–50. waldrop, m. m. (1992). complexity: the emerging science at the edge of order and chaos. new york, ny: touchstone. article_14-2.indd 27 03/07/14 1:31 am article_14-2.indd 28 03/07/14 1:31 am complex governance networks: an assessment of the advances and prospects han for meticulously reviewing the manuscript for possible stylistic errors. complexity, governance & networks (2014) 5–16 5 doi: 10.7564/14-cgn5 complex governance networks: an assessment of the advances and prospects* göktuğ morçöl pennsylvania state university at harrisburg e-mail: gxm27@psu.edu the case is made for a complex governance networks conceptualization. the contributions of the literatures on governance, networks, and complexity studies are discussed. governance researchers challenged the traditional ways of thinking in public policy and administration and highlighted the multi-centered nature of policy and administrative processes. governance networks researchers applied the refined network concepts and methods to our understanding of governance processes. complexity researchers provide ontological and epistemological grounding to governance networks studies and conceptual and methodological tools to study the selforganizational, emergent, and coevolutionary processes within and among complex governance networks. it is argued that the concepts and methods of the three streams should be synthesized and that complex governance network researchers should incorporate the insights of wellestablished theories of policy processes, such as the institutional analysis and development framework and the advocacy coalition framework. keywords: complexity, governance, networks, policy processes, public administration. 1. introduction it has long been recognized that the study and practice of public policy and administration are complex. in recent decades three streams of academic literature emerged that recognize this complexity: the literatures on governance, network studies, and complexity theory. although each of these streams has evolved somewhat autonomously, there have been considerable overlaps among them and some attempts to synthesize selected concepts from them. my argument is that these streams can and should be joined together under a complex governance networks theoretical framework to make better sense of policy and administrative processes. as i demonstrate in this paper, the theoretical developments and supporting empirical work conducted in the three streams can constitute the bases of this framework, but there are also areas where more theoretical refinement is needed and more empirical studies would be required. * i want to thank lasse gerrits for his most valuable comments on an earlier version of this paper and sehee article_14-5.indd 5 03/07/14 1:33 am 6 g. morçöl / complex governance networks as i briefly discuss in the following paragraphs, the three literatures have developed at different paces and each reached a different level of maturity. for instance, network conceptualizations and network analysis methods are more advanced than the conceptualizations and methods in the governance and complexity literatures. the fast-growing literature on governance studies has opened up new venues of research and practice, but compared to the networks studies they are less articulate. the applications of complexity theory in public policy and administration have been relatively more recent. a coherent and unified conceptual framework has not emerged from these applications yet, but, as i will argue, complexity theory offers a meta-theoretical framework that can help advance our understanding of complex governance networks. in the following sections, i summarize the evolutions of the concepts and methods used in the governance, networks, and complexity literatures and highlight the strengths and shortcomings in each. these summaries are not meant to be exhaustive or definitive reviews of the literatures, but to highlight their potentials to jointly contribute to our understanding of complex governance networks. 2. governance the increased interest in governance in the academic literature in recent decades is the result of a series of related observations social theorists and public policy and administration researchers make: that societies have become multi-centered (castells, 1996; jessop, 1990), that in today’s world no governmental or private actor has the capacity to solve the increasingly complex and dynamic problems of societies (kooiman, 1993), and that consequently the roles of governments in policy processes have diminished, or at least changed (kettl, 2002; koliba, meek, & zia, 2011; rhodes, 1997; torfing, peters, pierre, & sørensen, 2012). according to these authors, the policymaking capabilities of nation states have diminished and the political and policymaking processes have become multi-centered. this multi-centeredness creates institutional collective action dilemmas, according to feiock (2013), and this fragmentation makes governance inherently complex (feiock & scholz, 2010). all these observations underscore the multi-centered and complex nature of governance processes. the concept of governance has not been defined well. in the vague definition of the world bank, which popularized the concept, governance refers to the relationships between governing bodies and the governed. governance is the process “by which authority is conferred on rulers, by which they make the rules, and by which those rules are enforced and modified” (world bank, undated, para. 1). in the public management literature, governance refers more narrowly to the funding and oversight roles of government agencies, especially regarding the activities of private organizations that have been contracted to provide public services (provan & kenis, 2007, p. 230). lynn, heinrich and hill (2000) offer a definition that stresses the multiplicity of autonomous individuals and organizations that are involved in policymaking and the need to direct, control, and coordinate their actions: governance is the “means for achieving direction, control, and article_14-5.indd 6 03/07/14 1:33 am g. morçöl / complex governance networks 7 coordination of wholly and partially autonomous individuals or organizations on behalf of interests to which they jointly contribute” (p. 235). then the question is, who or what will direct, control, and coordinate these actors? can governments do that? what is a “government,” or “the state,” anyway? according to rhodes (1997, p. 57), in the “new era of governance” what we call “the state” is no longer a unified entity that is sovereign over a territory; it is actually composed of networks of what is traditionally considered “governmental” actors and other societal actors. then, can these governmental actors direct, control, or coordinate other actors? according to buijs, van der bol, teisman, and byrne (2009), the guiding ability of governmental agencies has diminished in recent decades because their actions are “highly influenced by the spontaneous actions of many other agents” (p. 97). kettl (2002, p 161.) and agranoff (2007, p. 192) argue that governmental actors still play special roles in governance processes: they can guide, steer, control, and manage non-governmental actors. what really are the roles governmental entities in governance processes? as i discuss in the next section, some governance networks researchers addressed this question, but more conceptual and empirical work needs to be done to answer the question. 3. networks the theory, methods, and empirical studies on networks have deep roots and they are at fairly advanced stages of development. the sociometric studies of moreno and others in the 1930s were the precursors of today’s social network analyses (sna) (wasserman & faust, 1994, pp. 9–17). the sna methods have come a long way since these initial sociometric studies, as a result of the advances in graph theory, statistical and probability theory, and algebraic models, according to wasserman and faust. these methods are described in detail in wasserman and faust’s book and others (e.g. freeman, 2008; knoke & yang, 2008; scott, 2013). there have also been advances in conceptualizing and empirically investigating the relations in social networks. one of these key advances in the concept of “governance networks,” which is used usually in conjunction with, or interchangeably with, the terms “policy networks” or “public management networks.” each of these three concepts has its conceptual history and there are differences, as well as similarities, among them (for different usages and meanings of the terms, see compston, 2009 and agranoff, 2007). governance networks is the most meaningful of the three because it captures the basic insights of the governance conceptualizations discussed in the previous section. in governance network conceptualizations, some of the key insights of network and complexity theories, and others, blended together. the history of this concept and the current state of its applications are discussed in depth by koliba et al. (2011). koliba, meek, and zia’s characterizations of governance networks include the self-organizing, and interdependent organizational and individual actors (e.g. resource dependencies) in them. in their comprehensive analysis of the complexity of governance networks, klijn and koppenjan (2004, 2014) also draw on the insights of network and complexity theories. article_14-5.indd 7 03/07/14 1:33 am 8 g. morçöl / complex governance networks in their comprehensive review of the governance/policy literature, provan, fish, and sydow (2007) note that although there is no overarching conceptualization that would guide empirical studies of governance networks yet, the studies on them have advanced at two levels: the micro level (analyses of individual and organizational actors) and the macro level (analyses of “whole networks”). provan and his colleagues point out that the micro-level studies have helped answer questions like, what are the impacts of network ties on organizational performance, which positions in networks are most influential, and how the positions within networks in response to changes within and outside the network? they note that although many measures have been developed to analyze whole networks (e.g. measures of centralization, density, fragmentation), there are very few whole network studies and they are mostly conceptual, anecdotal, or descriptive case studies. shrestra, berardo, and feiock’s (2014), article is a good example of how far advanced and refined micro-level network conceptualizations have become. they propose a model to be used in studying how actors shape their formal and informal relationships in dealing with institutional collective action problems. in the sna literature models of these multiple, forms of relationships are called “multiplex networks.” shrestra and his colleagues contribute to our understanding of multiplex networks by offering specific mechanisms of coordination and cooperation in them. provan & kenis’s (2007) typology of governance networks is one of the very few macro-level conceptualizations. they observe that there are participant-governed networks, lead-organization governed networks, and networks with a network administrative organization. agranoff (2007) also developed a typology of public management networks: informational networks, developmental networks, outreach networks, and action networks. according to provan, et al. (2007), governance network studies should go beyond these typologies and answer some key questions about the macro structures of networks. for example, what are the conditions of the emergence of different network structural forms? particularly, what is the role of governmental entities in shaping and constraining network structures? these questions have not been answered in the literature. another problematic area is how networks evolve. for example, does network evolution occur in predictable ways? do networks change continually or reach points of stability? there are some studies on how the relations among network actors evolve over time, but they focused mainly in the network dynamics at micro levels (e.g. doz, 1996; human & provan, 2000; isett & provan 2005; morçöl, vasavada, & kim, 2013; ring & van de ven, 1994). another important area of study is network effectiveness. more specifically, what kinds of network structures are more effective? forrest’s (2003) review of the international literature on policy networks indicates that there are conflicting findings about whether open (more inclusive) networks or closed (less inclusive) networks are more effective. the studies also have conflicting findings about the implications of policy networks for democracy: some suggest that networks can increase participation in policymaking. others find the informal nature of networks give power elites the opportunity to enhance their powers and decrease transparency and accountability in policymaking. article_14-5.indd 8 03/07/14 1:33 am g. morçöl / complex governance networks 9 4. complexity governance processes and the structures of governance/policy/public management networks are complex, as many theorists who represent various perspectives acknowledge (e.g. koliba et al., 2011; peters, 1993; sharkansky, 2002). then the question is, what do we mean by complexity? complexity theorists make two key contributions in answering this question. first, they show that complexity theory concepts challenge the deep ontological and epistemological assumptions of the newtonian/positivist science, which dominates most of the social science research (morçöl, 2002). second, complexity theorists define the concept of complexity and offer the conceptual and methodological tools to be used in identifying and describing the complex patterns and mechanisms of natural and social systems. thus they help us better understand the structural properties and dynamics of governance networks/systems, as well as the relationships among the elements (actors, agents) involved in these systems. complexity theory has the potential to make significant contributions to our understanding of governance networks, but it is not a well-developed theory—not yet. this is why many researchers would not even use the term “complexity theory” because, they argue, there is no coherent body of concepts that this name refers to (e.g. mitchell, 2009). kiel (2014), on the other hand, argues that the conceptual development of complexity theory in the public administration and policy studies is complete. in my view, no complete and comprehensive theory of the complex governance networks has been formulated, but the term “complexity theory” can still serve a useful purpose as a meta-theoretical framework (the body of related concepts—such as, nonlinearity, self-organization, emergence, and coevolution—and analytical tools—such as, agent-based simulations) from which concepts and analytical tools can be drawn to be applied in the studies of governance processes (morçöl, 2012a). in this paper i use the terms complexity theory and complexity theorist to refer to this framework, although it is only loosely articulated as of now, and those who contributed to it; i will leave aside the debates on the appropriateness of the uses of these terms. a few authors applied complexity concepts and methods in their studies of policy systems/governance networks (e.g. gerrits, 2012; geyer & rihani, 2010; kiel, 1994; morçöl, 2012a; rhodes, murphy, muir, & murray, 2011; teisman, van buuren, & gerrits, 2009), but there is no commonly applied framework in these studies. there are two theoretical frameworks that are emerging, however: the socio-ecological systems framework and the micro–macro framework. in this section, i will summarize the general implications of complexity theory and then discuss these two emerging frameworks and the methodological advances made in the studies of governance networks within these frameworks. complexity theorists challenge some of the core assumptions of the newtonian/ positivist sciences and propose a post-newtonian understanding of science, as i discussed elsewhere (morçöl, 2001, 2002, 2005, 2012a). complexity researchers are not in full agreement on the epistemological implications of the theory. mitchell (2009) and richardson (2007, 2010) argue that the theory suggests a pluralistic epistemology, whereas prigogine article_14-5.indd 9 03/07/14 1:33 am 10 g. morçöl / complex governance networks and stengers (1984) argue that it suggests a phenomenological view of science and in cilliers’s (1998) view it suggests a post-structuralist epistemology. despite these different characterizations, most complexity researchers, if not all, agree that the theory offer a different view of scientific knowledge. a significant implication is that complexity is not a residual category, a category that can be used to dump into what we do not understand about natural or social phenomena. therefore, researchers should aim to understand the complexities of the phenomena, not reduce them into simplified and linear explanations. complexity theorists offer a set of concepts and methodological tools to help us understand complexity (see morçöl, 2012a). they stress, for example, that what makes a system complex is not only in the large number of elements it is composed of, but also in the differences among its components and particularly the nonlinearity in the relations among them. complexity theorists also offer conceptualizations of the self-organizational mechanisms in systems, emergence of structural properties in systems, and co-evolution of systems. these conceptualizations are compatible with, and provide the theoretical bases of, some of the key propositions in the governance network conceptualizations: that the actions of multiple self-organizing actors constitute governance processes and that no governmental or private actor alone has the capacity to solve the increasingly complex problems of societies (kooiman, 1993). complexity theorists suggest that the macro (structural) properties of complex systems are emergent: these systemic properties emerge from the interactions of the systems elements; they are not designed. because complex systems are self-organizational and composed of self-organizing elements (actors, agents), they cannot be controlled or directed centrally, nor can their “problems be solved” centrally. a key observation complexity theorists make about policy/governance systems is that the relations between the policy actions of governments and “their outcomes” are nonlinear and therefore the outcomes are not easily predictable or controllable (salzano, 2008). there is no direct causal connection between policy actions and social outcomes; social processes and structures, such as “policy outcomes” emerge from the nonlinear interactions of self-conscious and self-organizing actors. axelrod (1997) and holland (1995, 1998) made significant contributions to our understanding of the emergence of social (macro) structures and processes in general, which can be applied in studies of governance networks. a key implication of the emergent nature policy outcomes is that policy interventions can at best be used to “nudge” systems toward socially desirable states (zia, et al., 2014). as stewart and ayres (2001) put it, public policies should be viewed as interventions into self-organizing social systems and the aim of a policy intervention should not be to reach a pre-determined goal, but to enable the “target system to enhance . . . its capacity for self-steering” (p. 87). therefore, the success of a policy intervention should be based on how much the self-steering capacity of the target system has been enhanced. complexity researchers demonstrated that governance/policy systems/networks do not self-organize in isolation; they coevolve with natural systems. the term coevolution was coined by ehrlich and raven (1964) and the theory of the coevolution of biological article_14-5.indd 10 03/07/14 1:33 am g. morçöl / complex governance networks 11 systems was refined by kauffman (1993, 1995). kauffman posits that as organisms, which are self-organizing systems, adapt to their environments to survive, their environments (other species) adapt to their adaptive behaviors, i.e. all these systems coevolve. gerrits’s (2008, 2010) case studies of the estuaries in northern europe demonstrated that policy systems coevolve with natural systems and ignoring these coevolutionary processes can lead to unexpected adverse outcomes. the notion that governance/policy systems coevolve with other systems is closely related to one of the two emerging frameworks in complexity theory applications in the studies on governance networks: the socio-ecological systems framework. this framework is illustrated in the works of zia (2013) and zia et al., (2014), as well as in gerrits’s (2008, 2010). this framework is rooted in the earlier socio-technical systems conceptualizations in organization theory (perrow, 1986, pp. 119–156). in the socio-ecological (or socio technical) systems framework, social systems are conceptualized as closely intertwined with natural systems. the researchers who adopted this framework studied the mechanisms of the interactions of human systems with specific technologies that were developed to harness/control/manipulate natural phenomena (e.g. perrow’s study of nuclear power plants) or those interactions with specific or general ecologies (e.g. gerrits’s study on estuaries and zia’s (2013) conceptualization of global warming). the socio-ecological systems framework is primarily concerned about system (macro-) level interactions and processes. the second framework that is emerging in the complexity studies of governance systems is the micro–macro framework, whose primary focus is the interactions among network actors (i.e. the micro level) and the macro structures and patterns that emerge from these interactions. more specifically, the “micro– macro problem” is the problem of how the actions of individual actors generate collective outcomes and how, in turn, the collective structures (social institutions, systems of rules) affect individual actions. this theoretical problem is also known as the “agency–structure problem” (simmel’s sociology), the “transformation problem” (european sociology), or the “collective action problem” (rational choice theories; for a specific articulation, see feiock, 2013) (coleman, 1986). as i discussed extensively elsewhere (morçöl, 2012a, 2012b), emergence is the central concept of complexity theorists’ conceptualizations of micro–macro processes (e.g. axelrod, 1997; holland, 1995, 1998; sawyer, 2005). complexity researchers have made significant advances in understanding the mechanisms of emergence, but there are conceptual and empirical problems to be addressed (see morçöl, 2012a, pp. 89–92). for example, once emerged, are macro-level (system-, or network-level) properties “irreducible” (do they constitute a separate realm that should be studied completely separately from the actions of individual actors? also, how do the macro properties of governance systems affect the actions and beliefs/perceptions of individual actors? within both the socio-ecological and micro–macro frameworks, many questions remain unanswered, but complexity researchers have developed, or adopted from others, innovative methods and applied them to address these questions empirically. for example, to study system-(macro-) level interactions and processes, they used phase diagrams to article_14-5.indd 11 03/07/14 1:33 am 12 g. morçöl / complex governance networks study “phase spaces” (e.g. zia, et al., 2014) and systems dynamics modeling to study the complex interactions of system-level variables (e.g. fiddaman, 2007). social network analyses (sna) and agent-based simulations have been used to study micro–macro interactions and transformations. sna methods are used to investigate “how large-scale systemic transformations emerge out of the combined preferences and purposive actions of individuals. . . . [they provide] conceptual and methodological tools for linking changes in microlevel choices to macrolevel structural alterations” (knoke & yang, 2008, p. 6). these methods have been applied at increasing rates in the studies on the complex governance networks that deal with natural disasters (e.g. kapucu, 2006; kapucu, arslan, & collins, 2010) and health and human service delivery networks ( provan, huang, & milward, 2009). abs are used to study the dynamics of micro–macro relationships and transformations in complex systems (e.g. johnston, nan, zhong, & hicks, 2008; zia, norton, noonan, rodgers, & dehart-david, 2006). i discussed the strengths and shortcomings of sna and abs elsewhere (morçöl, 2012a, pp. 210–243). one of the major shortcomings of both methodologies is that researchers must remove from their analyses the contexts of the cases and situations they study. in qualitative case studies (qcs), which are more commonly used by european complexity researchers (e.g. teisman et al., 2009; rhodes et al., 2011), this de-contextualization problem is lessened. qca have been applied successfully to describe the structural properties of complex policy/governance systems and the properties of policy actors. qca has the shortcoming of being too particular and not allowing researchers make generalizations, however. therefore, in studies of governance networks, ideally, sna, abs, and qca) should be combined, as i proposed and discussed elsewhere (morçöl, 2012a, p. 248). 5. conclusions governance researchers challenged the traditional ways of thinking in public policy and administration and raised questions about a series of important issues, from whether governments can control multi-centered governance processes, to how autonomous policy/ administrative actors interact in these processes. governance theorists have offered only few answers to the questions they raised, however. the students of “governance networks” began to answer some of these questions. complexity theorists and researchers have contributed to our understanding of governance networks by describing how networks structures emerge from the interactions of self-organizing actors (individuals and organizations) (see morçöl, 2012a, pp. 11–119) and how governance networks (or policy systems) coevolve with natural systems (e.g. gerrits, 2008, 2010). two frameworks have emerged from these studies: the socioecological systems framework, whose main focus is the system(macro-) level, and the micro–macro framework, as i discussed in the previous section. it is yet to be seen which directions these two emerging frameworks will take. ideally, these frameworks should be combined to gain more a comprehensive understanding of complex governance networks. more comprehensive insights can be gained also by combining the multiple analytical tools used article_14-5.indd 12 03/07/14 1:33 am g. morçöl / complex governance networks 13 by complexity researchers: systems dynamics modeling, agent-based simulations, social network analyses, and qualitative case studies. can the complex governance networks conceptualization and the methods associated with help us understand how the governance processes work in the emerging multicentered societies (castells, 2000; jessop, 1990)? can these conceptual and methodological tools help the current fragmented political and administrative systems deal with their institutional collective action dilemmas (feiock, 2013)? to be able to meet these challenges, complex governance networks researchers need to engage theoretically the insights and systematic conceptualizations made by other theorists and could learn from them. the institutional analysis and development framework of ostrom (1990, 2005) and the advocacy coalition framework of sabatier (sabatier & jenkins-smith, 1993) are the primary candidates for such engagements. ostrom expressed her interest in complexity theory and its concepts in multiple occasions (e.g. ostrom, 2005, pp. 242–243, 256; 2007). sabatier and his colleagues adopted some of the key concepts of systems theorists (e.g. policy subsystems) and made frequent references to network concepts in the later versions of their framework (sabatier & weible, 2007). complex governance network researchers would benefit from learning from the insights of ostrom and sabatier. ostrom made significant contributions to our understanding of the conditions of self-governance (self-organization). she and her colleagues demonstrated with their studies over a few decades that the self-organizational processes in the management of “common pool resources” (i.e. natural or human-made resources that are shared by large numbers of economic actors) are conditioned by various factors and mechanisms and they codified these factors and mechanisms. her and their works are exemplary for complexity researchers. they demonstrate that for a theoretical framework to be influential (ostrom’s works were cited in over 70,000 other works, as of april 2014, according to google scholar, and she was awarded the nobel prize in economics in 2009), it should be carefully and meticulously constructed and its concepts and propositions should be empirically verified. ostrom’s impressive framework is based on the simplifying assumptions of rational choice theory about human behaviors, such as that they are (boundedly) rational actors (ostrom, 2005, chap. 4). sabatier’s framework, which is also impressive and based on decades of meticulous conceptualization and empirical verification, is based on a different set of assumptions. one of his key propositions is that the belief systems of policy actors play roles in how they come together to form advocacy coalitions (networks) in policy processes. these beliefs systems are not necessarily “rational” in the sense that rational choice theory assumes them to be. because complex governance researchers study social systems (networks), it is crucial for them to understand the beliefs (or more broadly the social constructions) of the actors these networks as they are, and not simply assume that they are (boundedly) rational. ostrom’s and sabatier’s frameworks, as well as the meticulous conceptual and empirical works that contributed to their developments, can be inspirational for complex governance networks researchers and theorists. article_14-5.indd 13 03/07/14 1:33 am 14 g. morçöl / complex governance networks references agranoff, r. (2007). managing within networks: adding value to public organizations. washington, dc: georgetown university press. axelrod, r. (1997). the complexity of cooperation: agent-based models of competition and collaboration. princeton, nj: princeton university press. buijs, j.-m., van der bol, n., teisman, g. r., & byrne, d. (2009). metropolitan regions as self-organizing systems. in g. teisman, a. van buuren, & l. gerrits (eds.), managing complex governance systems: dynamics, self-organization and coevolution in public investments (pp. 97–115). london, england: routledge. castells, m. (1996). the rise of the network society. oxford, uk: blackwell. cilliers, p. (1998), complexity and postmodernism: understanding complex systems. london, england: routledge. coleman, j. s. (1986). social theory, social research, and a theory of action.the american journal of sociology, 91(6), 1309–1335. compston, h. (2009). policy networks and policy change. new york, ny: palgrave macmillan. doz, y. l. (1996). the evolution of cooperation in strategic alliances: initial conditions or learning processes? strategic management journal, 17, 55–83. ehrlich, p. r., & raven, p. h. (1964). butterflies and plants: a study in coevolution. evolution, 18, 586–608. feiock, r. c. (2013). the institutional collective action framework. policy studies journal, 41(3), 397–425. feiock, r. c., & scholz, j. t. (eds.). (2010). self-organizing federalism: collaborative mechanisms to mitigate institutional collective action. cambridge, england: cambridge university press. fiddaman, t. (2007). dynamics of climate policy. system dynamics review, 23 (1), 21–34. forrest, j. b. (2003). networks in the policy process: an international perspective. international journal of public administration, 26, 591–607. freeman, l. c. (ed.). (2008). social network analysis (vols. 1–4). los angeles, ca: sage. gerrits, l. (2008). the gentle art of coevolution: a complexity theory perspective on decision making over estuaries in germany, belgium and the netherlands (phd dissertation, erasmus university rotterdam). retrieved from http://publishing.eur.nl/ir/repub/asset/11152. gerrits, l. (2010). public decision making as coevolution. emergence: complexity & organization, 12(1), 19–28. gerrits, l. (2012). punching clouds: an introduction to the compelxity of public decision-making. litchfield park, az: emergent publications. geyer, r., & rihani, s. (2010). complexity and public policy: a new approach to 21st century politics, policy and society. london, england: routledge. holland, j. h. (1995). hidden order: how adaptation builds complexity. new york, ny: basic books. holland, j. h. (1998). emergence: from chaos to order. new york, ny: basic books. human, s. e., & provan, k. g. (2000). legitimacy building in the evolution of small-firm networks: a comparative study of success and demise. administrative science quarterly, 45,327–365. isett, k. r., & provan, k. g. (2005). the evolution of interorganizational network relationships over time: does sector matter? journal of public administration research and theory, 15, 149–165. jessop, b. (1990). state theory: putting the capitalist state in its place. state college, pa: the pennsylvania state university press. johnston, e. w., nan, n., zhong, w., & hicks, d. (2008). between implementation and outcomes, growth matters: validating an agent-based modeling approach for understanding collaboration process management. the innovation journal: the public sector innovation journal, 13(3), 1–31. kapucu, n. (2006). interagency communication networks during emergencies: boundary spanners in multiagency coordination. american review of public administration, 26(2), 207–225. kapucu, n., arslan, t., & collins, m. l. (2010). examining intergovernmental and interorganizational response to catastrophic disasters: toward a network-centered approach. administration & society, 42(2), 222–247. kauffman, s. (1993). the origins of order: self-organization and selection in evolution. new york, ny: oxford university press. article_14-5.indd 14 03/07/14 1:33 am g. morçöl / complex governance networks 15 kauffman, s. (1995). at home in the universe: the search for laws of self-organization and complexity. new york, ny: oxford university press. kettl, d. f. (2002). the transformation of governance: public administration for twenty-first century america. baltimore, md: the johns hopkins university press. kiel, l. d. (1994). managing chaos and complexity in government: a new paradigm for managing change, innovation, and organizational renewal. san francisco, ca: jossey-bass. kiel, l. d. (2014). complexity theory and its evolution in public administration and policy studies. complexity, governance & networks, 1, 71–78. klijn. eh., & koppenjan, j. f. m. (2014 this issue). complexity in governance network theory. complexity, governance & networks, 1, 61–70. knoke, d., & yang, s. (2008). social network analysis (2nd ed.). los angeles, ca: sage.. koliba, c., meek, j. w., & zia, a. (2011). governance networks in public administration and policy. boca raton, fl: crc press. kooiman, j. (1993). social-political governance: introduction. in j. kooiman (ed.), modern governance: new government–society interactions (pp. 1–6). london, england: sage. koppenjan, j. f. m., & klijn, e. -h. (2004). managing uncertainties in networks. london, england: routledge. lynn, l. e., heinrich, c. j., & hill, c. j. (2000). studying governance and public management: challenges and prospects. journal of public administration research and theory, 10, 233–261. mitchell, m. (2009). complexity: a guided tour. oxford, uk: oxford university press. morçöl, g. (2001). what is complexity science: postmodernist or postpositivist. emergence: a journal of complexity issues in organizations and management, 3(1), 104–119. morçöl, g. (2002). a new mind for policy analysis: toward a post-newtonian and postpositivist epistemology and methodology. westport, ct: praeger. morçöl, g. (2005). phenomenology of complexity theory and cognitive science: implications for development of an embodied knowledge of public administration and policy. administrative theory & praxis, 27(1), 1–22. morçöl, g. (2012a). a complexity theory for public policy. london, england: routledge. morçöl, g. (2012b). the micro–macro problem as the core problem of complexity theory of policy processes. in l. gerrits, p. marks, & f. boons (eds.), compact 1: public administration in complexity (pp. 307–322). litchfield park, az: emergent publications. morçöl, g., vasavada, t., & kim, s. (2013). business improvement districts in urban governance: a longitudinal case study. administration & society (published online before print january 31, 2013, doi: 10.1177/0095399712473985). ostrom, e. (1990). governing the commons: the evolution of institutions for collective action. cambridge, uk: cambridge university press. ostrom, e. (2005). understanding institutional diversity. princeton, nj: princeton university press. ostrom., e. (2007). biography of robert axelrod. ps, political science and politics, 40(1), 171–174. perrow, c. (1986). complex organizations: a critical essay (3rd ed.). new york, ny: mcgraw hill. peters, b. g. (1993). american public policy: promise and performance (3rd ed.). chatham, nj: chatham house. prigogine, i., & stengers, i. (1984). order out of chaos: man’s new dialogue with nature. new york, ny: bantam books. provan, k. g., fish, a., & sydow, j. (2007). interorganizational networks at the network level: a review of the empirical literature on whole networks. journal of management, 33(3), 479–516. provan, k. g., huang, k., & milward, b. (2009). the evolution of structural embeddedness and organizational social outcomes in a centrally governed health and human services network. journal of public administration research and theory, 19(4), 873–893. provan, k. g., & kenis, p. (2007). modes of network governance: structure, management, and effectiveness. journal of public administration research and theory, 18, 229–252. rhodes, m. l., murphy, j., muir, j., & murray, j. a. (2011). public management and complexity theory: richer decision-making in public services. london, england: routledge. rhodes, r. a. w. (1997). understanding governance: policy networks, governance, reflexivity and accountability. maidenhead, uk: open university press. article_14-5.indd 15 03/07/14 1:33 am 16 g. morçöl / complex governance networks richardson, k. a. (2007). complex systems thinking and its implications for policy analysis. in g. morçöl, (ed.), handbook of decision making (pp. 189–222). boca raton, fl: taylor and francis. richardson, k. a. (2010). thinking about complexity: grasping the continuum through criticism and pluralism. litchfield park, az: emergent. ring, p. s., & van de ven, a. h. (1994). developmental processes of cooperative interorganizational relationships. academy of management review, 1, 90–118. sabatier, p. a., & jenkins-smith, h. (1993). policy change and learning: an advocacy coalition approach. boulder, co: westview press. sabatier, p. a., & weible, m. (2007). the advocacy coalition framework: innovations and clarifications. in p. sabatier (ed.), theories of the policy process (2nd ed., pp. 189–220). cambridge, ma: westview press. salzano, m. (2008). economic policy hints from heterogeneous agent-based simulations. in k. richardson, l. dennard, & g. morçöl (eds.), complexity and policy analysis: tools and methods for designing robust policies in a complex world (pp. 167–196). goodyear, az: isce. sawyer, r. k. (2005). social emergence: societies as complex systems. cambridge, uk: cambridge university press. scott, j. (2103). social network analysis (3rd ed.). los angeles, ca: sage. sharkansky, i. (2002). politics and policymaking: in search of simplicity. london, england: lynne rienner. shrestra, m., berardo, r. & feiock, r. c. (2014). solving institutional collective action problems in multiplex networks. complexity, governance & networks, 1, 49–60. stewart, j., & ayres, r. (2001). systems theory and policy practice: an exploration. policy sciences, 34(1), 79–94. teisman, g. r., van buuren, a., & gerrits, l. (eds.). (2009). managing complex governance systems: dynamics, self-organization and coevolution in public investments. london, england: routledge. torfing, j., peters, g. b., pierre, j., & sørensen, e. (2012). interactive governance: advancing the paradigm. oxford, uk: oxford university press. wasserman, s., & faust, k. (1994). social network analysis: methods and applications. cambridge, uk: cambridge university press. world bank (undated). what is governance. retrieved from http://web.worldbank.org/wbsite/external/ countries/menaext/extmnaregtopgovernance/0,contentmdk:20513159~pagepk:340 04173~pipk:34003707~thesitepk:497024,00.html. accessed on february 28, 2014. zia, a. (2013). post-kyoto climate governance:confronting the politics of scale, ideology and knowledge. london, england: rutledge. zia, a., kauffman, s., koliba, c., beckage, b., vattay, g., bomblies, a. (2014). from the habit of control to institutional enablement: re-envisioning the governance of social-ecological systems from the perspective of complexity sciences. complexity, governance & networks, 1, 79–88. zia, a., norton, b. g., noonan, d. s., rodgers, m. o., & dehart-david, l. (2006). a quasi-experimental evaluation of high-emitter non-compliance and its impact on vehicular tailpipe emissions in atlanta, 1997–2001. transportation research part d, 11, 77–96. article_14-5.indd 16 03/07/14 1:33 am defining the entropy of hierarchical organizations complexity, governance & networks (2014) 41–56 41 doi: 10.7564/14-cgn17 defining the entropy of hierarchical organizations david chappella, t. gregory deweyb,* adepartment of math, physics and computer science, university of la verne, la verne, ca 91750 e-mail: dchappell@laverne.edu boffice of the provost university of la verne la verne, ca 91750 e-mail: gdewey@laverne.edu a measure of the order within a hierarchical organizational structure is proposed based on an analogy with the thermodynamic entropy of physical sciences. organizational entropy is defined in a manner that readily allows for practical calculations. this calculation could be used to relate the order or entropy of an organization to its functional role. additionally, it could be used to monitor the change in entropy over time and provide an impetus for periodic restructuring. this theory gives two general contributions to the entropy of an organizational structure: horizontal entropy due to changes within a level of the hierarchy and vertical entropy due to changes between levels along reporting lines within the organization. in addition to employing the thermodynamic entropy analog, identical theoretical results are obtained when calculating the kolmogorov entropy or algorithmic complexity of an organizational hierarchy and establishes the generality of the approach. computer simulations on model hierarchical structures show the boundaries of the vertical and horizontal contribution to entropy. it is postulated that each organization will have a specific entropy that optimizes organizational functions. keywords: organizational structure, entropy. 1. why define organizational entropy and how it can be useful many fields will claim that “form follows function”. this is true in architecture, engineering design, biology, and also in organizational management. the structure of an organization is often tied to the function [carley, 1995]. organizations that require flexibility and rapid decision making will have “flat” structures with considerable authority at lower levels of the organization. strongly hierarchical organizations will exist where high risk, strategic decision making is required. one can argue that each organization should customize its structure to suit its mission and goals. organizations are not static and their structure and very mission can change over time. the functions of the organization may change in response to changes in the industry. alternatively, the organizational structure may “degrade” over time due to loss * corresponding author. article_14-17.indd 41 03/03/15 3:06 pm 42 d. chappell et al. / defining the entropy of hierarchical organizations of personnel, reassignments or merging of functional units. so there may be a tendency for an optimal organizational structure to evolve over time to a less effective structure. in these cases, a restructuring of the institution is called for to bring the organization back in alignment with its optimal structure. the development of computational and mathematical organization theory over the last several decades has led to a wide array of quantitative metrics that may be used to characterize the structure, dynamics and function of organizations. these tools have benefited theorists interested in evaluating formal organizational models and managers who seek practical assessments to improve efficiency and innovation of their organizational units. often, quantitative metrics used in organizational management are borrowed from other fields. graph theory has been used to develop a range of measures to quantify organizational structures and social network connections (see, for example, hummon [1995], krackhardt et al. [1994], mcgrath et al. [2003]). metrics and computational organizational models have also been inspired by the natural sciences, particularly physics (see fabac [2008], stepanic et al. [2000]), and by information theory [schlick et al., 2013]. the physics-based studies implicitly ask to what extent social systems of interacting “agents” mimic physical systems of interacting particles. thermodynamics offers one of the most promising avenues for relating the two fields. attempts to find correlates between physical thermodynamic quantities and social or organizational quantities have been made by kasac [2004] and stepanic et al. [2000]. these studies are largely theoretical in nature and model broad, generic social characteristics. our paper has a narrower focus. we present a simple, quantitative metric based on thermodynamics that is tailored to measure a specific property (the entropy) of hierarchical organizations. this paper proposes a measure, the organizational entropy, to provide a quantitative measure of the disorder of the reporting structure of an organization. a highly ordered, low entropy organization would have a very hierarchical structure with strict lines of reporting. a federation or consortium would be a low order, high entropy structure with comparable authority in each member, a comparatively weaker administrative unit and very loose lines of communication. high order may be good in one instance and not desirable in another. a quantitative measure of order or organizational entropy makes no judgment of the benefit or disincentive to the order of an organization. it just provides a rigorous way of measuring that order. the term “entropy” appears in a number of different contexts with different meanings. in this work, organizational entropy is defined by analogy to thermodynamic entropy as used in physics and chemistry. entropy is a measure of disorder of a system. a perfect, flawless crystal at the zero of temperature has zero entropy. there is perfect order and, at the zero of temperature, there are no internal motions to disturb that order. a gas, on the other hand, has much higher entropy as its particles are free to move around in any direction, colliding with each other and moving off into new directions. for an organization, low entropy would correspond to highly uniform reporting structures. figure 1 and figure  2 shows organizational charts where the nodes represent individuals within the organization and the lines represent oversight or reporting relationships. comparatively article_14-17.indd 42 03/03/15 3:06 pm d. chappell et al. / defining the entropy of hierarchical organizations 43 speaking, figure 1 would be a low entropy organization because the structure is very orderly with a perfect ordering of reports. different areas of expertise, represented as red, orange and blue, are each contained within the three main organizational units. figure 2 on the other hand shows an organization of higher entropy because the regularity of the structure is disrupted by variability of the reporting structure and the areas of expertise are mixed across units. both organizations have the same number of individuals but clearly have different structures. the breaking of order in the second graph gives rise to a high entropy. in this paper, we propose how to measure the organizational entropy. this begs the question: why would one want to measure this entropy? if “form follows function”, there may be a value to measuring the entropy of each form. a given organizational structure can be assigned an organizational entropy value. we anticipate for a specific function there will be common structural elements that may have a common range of entropy values. it is then possible to compare or rank organizations according to their entropy and compare these rankings to some measure of the effectiveness of the organization. such a ranking may lead to insights about the factors that lead to an effective organization. for instance, a rigid hierarchical organization may not be very innovative and this could affect overall performance. similarly, a randomly connected line of reporting would also lead to a dysfunctional organization. so highly ordered (low entropy) or highly disordered (high figure 1. a highly structured organization with 4 hierarchical levels and 40 employees. every manager directly oversees three subordinates. the numeric string encodes the reporting structure of the organization. to generate the coded string, the tree is scanned level-by-level to record the number of subordinates under each node. figure 2. a disordered organization with 6 hierarchical levels and 40 employees. the numeric string encodes the reporting structure of the organization. article_14-17.indd 43 03/03/15 3:06 pm 44 d. chappell et al. / defining the entropy of hierarchical organizations entropy) organizations are both unlikely to be effective performers. there may well be an optimal organizational entropy in between complete order and complete disorder that affords the best structure for a given performance criterion (see figure 3). a second application would be to measure the entropy of an organization over time and see how it “evolves”. the second law of thermodynamics says that physical systems will naturally evolve from high order to low order. that is the entropy of a system will increase over time. while we are not positing that organizational structures necessarily follow the second law in this manner, it is a common observation that organizations need to restructure from time to time. such restructuring can be a result of an accumulation of internal personnel and reporting line decisions that result in organizational dysfunction or inefficiencies. alternatively, the external environment of the organization may change and require that a new set of functions and expertise be brought into the organization. both effects could move the organizational structure and its associated entropy away from the optimum in figure 3. so a potential use of organizational entropy is to be a bell weather to observe how organizations drift away from their optimal structure. 2. how to measure organizational entropy there is an arsenal of techniques for defining organizational entropy. thermodynamic entropy has the longest history and relates to our everyday physical world. a drinking glass dropped on the floor is more likely to break into hundreds of shards than for hundreds of shards to spontaneously assembly into a drinking glass. nature spontaneously moves from low entropy (order) to high entropy (disorder). over time, a number of definitions of entropy have arisen from various fieldsshannon information entropy from communication theory, kolmogorov entropy or computational complexity from theoretical computer science and maximum entropy techniques from bayesian statistics. while all of these definitions are motivated by problems in their specific fields, they have deep connections with each other and demonstrating the relationship between them is an active figure 3. hypothesized functional dependence of an organizations performance and its entropy. article_14-17.indd 44 03/03/15 3:06 pm d. chappell et al. / defining the entropy of hierarchical organizations 45 field of research. to demonstrate the generality of our definition of organizational entropy, we derive identical sets of equations for calculating organizational entropy from both a thermodynamic entropy and from a kolmogorov entropy approach. these two approaches are briefly described below: thermodynamic entropy this model is based on the thermodynamic concept of entropy used in numerous examples in physical chemistry. the two primary calculations used in this paper are the entropy of mixing (to apply to mixing of expertise within a functional unit) and the entropy of expansion of an ideal gas (to apply to expansion or contraction of staffing in an organizational unit). kolmogorov entropy this model arises from kolmogorov’s concept of computational complexity used in theoretical computer science. the organizational complexity is given by the shortest description in binary code required to describe the organizational structure. complicated structures require long descriptions while simple structures can be easily described. thus, encoding the description of a complicated structure will require more bits than that of a simple structure. 3. the calculation of organizational entropy this section develops the mathematics behind the calculation of organizational entropy. the first subsection focuses on analogies with simple examples from physical chemistry and shows how these can be used to describe entropy changes due to “personnel shuffling” and also due to the expansion or contraction of organizational units. the intent of this section is to be heuristic and to provide a number of practical examples of calculations. the following subsection is more mathematical in nature and describes the effects of the alteration of reporting lines and consequently the shuffling of entire units within the organization. a formal mathematical development of personnel shuffling would follow along the same lines as that of unit shuffling, but is excluded here for brevity. so all the effects of both personnel and unit shuffling can be obtained using the approach presented in this section and should be inclusive of all entropic effects of the model. before proceeding, a definition of terms and the specifics of the model are first required. we define a hierarchical model as one in which a single individual (referred to as the ceo) is the convergent point for all reporting lines of the whole organization. parenthetically, this requirement means that all the graphical organizational structures generated by this model will have the property of a “rooted tree” in graph theoretical terms. the direct reports to the ceo represent the first tier in the hierarchy. a unit in the hierarchy is a functional department that reports to a unit above it in the hierarchy. for instance, the vp for operations may have budget, marketing and human resource (among others) reporting to him or her. these are then the units in the next tier. each tier of the hierarchy has units that only report to an individual in the tier above it. note that this representation may not be strictly accurate. for instance, a ceo will typically receive reports from a number of vps as well as administrative assistants. under this model, all these individuals are identified on the same tier even though their authority within the organization might be quite article_14-17.indd 45 03/03/15 3:06 pm 46 d. chappell et al. / defining the entropy of hierarchical organizations different. all direct reports to the ceo are in the first tier. all units that report to a first-tier unit reside in the second tier, etc. borrowing a term from genealogy, a “lineage” of a given unit within the hierarchy is the collection of subordinate units whose reporting lines trace back to the unit in question. note that not all lineages will necessarily have the same depth. just like family trees that have lineages that terminate at different generations, organizational trees may have lineages that terminate at different tiers. in a general derivation of organizational entropy as in section 3.2, mathematical terms are identified that relate to the entropy contribution from within tiers and a separate term for the entropy contribution between tiers. the sum of the entropy contribution within tiers is referred to as the “horizontal entropy”. the entropy contribution produced by variation across tiers is referred to as the “vertical entropy” and may be thought of as being a “generationally-dependent” measure of the lineages within the hierarchy. the total entropy is equal to the sum of the vertical entropy and horizontal entropy. these two different contributions mean that hierarchies can look quite different and still have the same total organizational entropy. in general, the horizontal entropy is maximal when units within each tier are populated with the same number of individuals. also, the vertical entropy is maximal when the population at each generation or tier along a linage remains the same. 3.1. thermodynamic analogies for the entropy of personnel shuffling invariably over the history of an organization, people get moved from one unit to the next. these moves are often motivated by personnel or human resource issues and are not necessarily made to make the organization stronger, more efficient or more innovative. this phenomenon is referred to here as personnel shuffling. personnel shuffling is like shuffling a new deck of cards where everything is ordered by suit. after shuffling the suits are mixed and the deck is overall in greater disorder or higher entropy. in a similar fashion, the organizational entropy increases with personnel shuffling. however, as we will see personnel shuffling can be more complicated than just shuffling a deck of cards. in addition to changing personnel, there can be an expansion or contraction of a unit and also there can be a shuffling of supervisors of a unit. two equations are needed to calculate the change in organizational entropy resulting from these personnel changes. the first equation is the entropy of mixing, a well-known equation in physical chemistry that can also be derived from information theory (see below). when two pure components are mixed the change in entropy that results from this mixing is given by (cf. atkins [1997]): ∑∆ 5 2 5 s f flogmixing i i i l 2 1 (1) where the sum is over the l components that are mixed and fi is the fraction of the ith component. in thermodynamics, eq. 1 would have a physical constant multiplier on the right hand side to give entropy units. in information theory, there is no need for this article_14-17.indd 46 03/03/15 3:06 pm d. chappell et al. / defining the entropy of hierarchical organizations 47 constant because the units are in “bits”, e.g. binary digits. in information theory, the logarithm is taken as base 2 to give units in bits. in our approach, the information theory notation is used. the total entropy change of a hierarchical unit is given by the sum of the entropy changes across each level: ∑∑∆ 5 2 55 s f flogmixing j i i l j i j n , 1 2 , 1 (2) where n is the total number of organizational units, and fj,i is the fraction of the i th component in the jth unit. the second equation needed is the equation that accounts for expansion or contraction of units. this equation also has an analog in physical chemistry with the expansion of an ideal gas (cf. atkins [1997]). the equation for unit expansion or contraction is: ∆     5s n n logexp cont after before / 2 (3) where nafter and nbefore represent the number of staff in the unit after and before the change. note that when nafter > nbefore, the entropy change is positive, and when nafter < nbefore, the entropy change is negative indicating a decrease in entropy. the total entropy change of the organization is given by the sum of the entropy changes of each unit: ∑∆     5s n n logexp cont i n i after i before / 2 , , (4) when the organization as a whole loses positions, the entropy is reduced and when it gains positions the entropy increases. 3.1.1. sample calculation of expansion and contraction when an organization adds positions, the entropy of expansion increases (δsexp > 0) and when it reduces positions δsexp < 0. for example, if a unit expands from 3 to 5 personnel, we find δsexp 5 log 2(5/3) 5 0.737 (see figure 4a). if the unit loses the two new employees later on, then δsexp 5 log 2(3/5) 5 0.737 (see figure 4b). the entropy change depends on how units are defined. suppose a large firm that has 60 employees in the marketing division hires a new employee. if one considers this division as a single unit then the entropy change is modest: δsexp 5 log 2(61/60) 5 0.0238. however, if the marketing division is composed of twelve sub-units, each with 5 employees then the entropy increase of the division will be more than ten times larger: δsexp 5 ∑ 12 i51 log2(ni,after/5) 5 11 log2  figure 4. example entropy change due to (a) expansion or (b) contraction of a single unit. the squares represent personnel in each unit. article_14-17.indd 47 03/03/15 3:06 pm 48 d. chappell et al. / defining the entropy of hierarchical organizations 5/5 1 log2(6/5) 5 0.263. in other words, the entropy of expansion is sensitive to coarse graining effects: it depends on how finely units are defined within the organization. when applying this method, a standardized method must be used when defining the units across the organization. if personnel are transferred from one unit (unit b) to another (unit a), then unit b will experience a reduction in sexp while sexp will increase for unit a (see figure 5). the net entropy of expansion sexp will increase if the reorganization acts to even out the number of staff in each unit and will decrease if it leads to a larger disparity in staffing between the units. for a fixed number of staff, the entropy of expansion is maximized when each unit in the organization has equal numbers of employees. for example if units a and b each have 5 staff initially, and two staff from b are transferred to a (resulting in 7 and 3 staff for a and b respectively), then sexp 5 log2(3/5) 1 log2(7/5) 5 20.252. in this example, the net entropy of expansion is reduced as the distribution of staff across units becomes more heterogeneous. 3.1.2. sample calculation of mixing with expansion and contraction the entropy of mixing smixing is nonzero when an organizational unit houses staff with expertise from different areas. it is maximized when a unit has equal numbers of staff from multiple areas. suppose, for example, an hr department replaces one of its 10 staff members with someone from another department (e.g. sales). the change in mixing entropy would be δsmixing 5 (1/10) log2(1/10) 1 (9/10) log2(9/10) 5 0.469. the positive value indicates the mixture of skills within the unit increased. figure 6 shows an example where a staff member from one unit (say sales) is transferred to another unit (say hr). if the sales team originally had 5 staff members and the hr department had 3 staff, the net entropy change would be δsmixing 1 δsexp 5 2.415 1 0.093 5 2.51. in this scenario the transfer of a staff member to a unit outside his/her expertise leads to a net increase in the figure 5. example entropy change when two personnel are transferred from one unit (unit b) to another (unit a). in this example, the initial equal distribution of personnel across the units corresponds to the maximum entropy state (for a fixed number of employees). the net entropy is reduced when personnel shuffling leads to a more unequal distribution of staff across units. article_14-17.indd 48 03/03/15 3:06 pm d. chappell et al. / defining the entropy of hierarchical organizations 49 entropy of the organization. however, net negative entropy changes are also possible if the entropy of expansion is sufficiently negative. these examples suggest that both measures δsmixing and δsexp could be important statistics to consider when evaluating the reason for a net increase or decrease in the overall entropy of an organization. 3.1.3. a more complicated sample calculation a specific example is now considered that illustrates the various types of personnel shufflingshuffling between units, contraction/expansion of a unit, and replacement of a unit head. these are shown in figure 7 where the organization before and after shuffling is shown. there are three units in the organizational structure consisting of 7 employees in marketing, 6 in sales and 6 in human resources. the color represents the expertise within the unit and of individuals. historically, the director has always come from sales and so that person is coded blue as well. now consider the following changes to the organization (illustrated in figure 5): • the director retires and is replaced by a promising employee in marketing. • an employee in sales is let go. one of the human resource staff, although not technically qualified for the job, applies and gets the job. • human resources has lost an important employee and is viewed as need an immediate replacement. marketing is generally viewed as being over staffed, so in addition to losing an employee to the directors position, a second employee is shifted over to human resources. as a result of these personnel changes, the organizational entropy has changed in a number of different ways. first, sales and human resources are no longer “pure” units because they have staff coming in from the outside that do not necessarily have direct expertise in this area. this shuffling results in an increase in entropy. although the director has changed, it is figure 6. units a and b are initially staffed by personnel with different expertise (color coded red and blue). the mixing of expertise in unit b leads to an increase in the mixing entropy. because the net entropy of expansion is also positive, the net entropy change is positive. article_14-17.indd 49 03/03/15 3:06 pm 50 d. chappell et al. / defining the entropy of hierarchical organizations still a pure unit and, therefore, there is no entropy associated with this. parenthetically, there may be a slight information entropy associated with this change depending on the descriptors of that position and this will be discussed in the next section. finally, there was a contraction in marketing and this results in a decrease of entropy or increase in order. to conceptualize this a unit with a hundred individuals, even if they are all “pure” in their expertise, is more complicated than a unit with one individual. so the smaller the unit, the more order. we now calculate the total entropy change in the example introduced in figure 7. the number of personnel in marketing, sales and human resources changes from (nm,  ns,  nhr) 5 (7, 6, 5) to (5, 6, 5) respectively. the resulting entropy of expansion is δsexp 5 20.48, which reflects the overall reduction in staffing of the organization. the mixing ratios for the three units are: marketing fm,i 5 (1, 0, 0), sales fs,i 5 (0, 5/6, 1/6), and human resources fhr , i 5 (1/5, 0, 4/5) giving mixing entropies for each of the units to be δsm 5 0, δss 5 0.65, δshr 5 0.72. the total mixing entropy of the organization is δsmix 5 0 1 0.65 1 0.72 5 1.37. the total entropy change for the organization is the sum of the expansion and mixing entropies: δs 5 20.48 1 1.37 5 0.89. we see that the overall entropy of the organization increased due to personnel shuffling despite the overall downsizing. 3.2. the entropy of organizational unit shuffling organizational entropy can arise from mechanisms other than personnel shuffling. personnel shuffling represent the movement of nodes within a hierarchical structure. however, we can also consider what happens when the connections or lines are moved. this corresponds to changing the reporting structure while keeping the units intact. a uniform figure 7. hypothetical organization before personnel shuffling. figure 8. hypothetical organization after personnel shuffling. article_14-17.indd 50 03/03/15 3:06 pm d. chappell et al. / defining the entropy of hierarchical organizations 51 unit is defined as a cohesive unit in which all members have expertise in the functionality of the unit. for instance, a marketing department consisting entirely of marketers is uniform. a marketing department with marketers plus a financial analysis and an accountant has higher disorder because it contains individuals normally not assigned to it and that are outside of its function. organizational disorder can also arise from a reporting relationship that is not aligned with the function. for instance, if the head of marketing is reassigned to report to the vice president for human resources, this represents a disorder in the hierarchy because of a functionally improper line of reporting. such alterations are treated as disorders in the hierarchy of an organization. in this section, the organizational entropy for a hierarchical structure is calculated. we consider a hierarchy that has l different levels and amongst those l levels are m departments or organizational units. the number of units at the lth level of the hierarchy is given by m l, so that ∑5 5 m m l i l 1 (5) the hierarchy is organized in such a manner that each unit at the lth level reports to a specific unit at the l 1 1 level. the number of units reporting to the uth office at the l 1 1 level is given by: mlu, and assigning all units to a unit above them gives: ∑5m ml ul u (6) we now ask the question, how many possible structures would fit this condition and this number is designated as ωtree and the entropy of the organization by analogy to the statistical mechanics of physical systems is defined as: ω5s lntree tree (7) to calculate this entropy, we first calculate how many ways that m units can be distributed over l levels. this is a combinatorial problem and is given by: m!∏li51 m l!. we then ask the question on how the m l units on a level can be distributed across the different reporting structures in the units above them. this is given by: ml!∏i51 mlu!. the total number of possible organization structures under these conditions is: m m m m ! ! ! ! tree i l l l u l i l l 1 1 1 5 5 5 5∏ ∏ ∏ω       (8) following some manipulations and the use of stirling’s approximation (note: stirling’s approximation will probably not hold), the entropy of the hierarchical tree is: ∑ ∑∑ ∑ ∑∑         5 2 2 5 2 5 5 5 5 s m m m m m m m p m p ln ln ln ln tree l l i l u l u l l ui l l l l l u l u l ul l 1 1 1 1 (9) article_14-17.indd 51 03/03/15 3:06 pm 52 d. chappell et al. / defining the entropy of hierarchical organizations where pl 5 (ml/m) and plu 5 (m l u /m l). the first term on the right had side represents the entropy of mixing of all the units across the levels of the hierarchical tree. we define this term as the horizontal mixing entropy: ∑∑5 2 5 s m plnh u l u l ul l 1 (10) the second sum is the sum over all levels of the entropy of mixing of the units within a level. we define it as the vertical mixing entropy: ∑5 2 5 s m plnv l l l l 1 (11) 3.2.1. sample calculations illustrating vertical and horizontal entropy effects as an example, consider an organization with 15 employees. we assume that the organization is headed by a single individual (producing what graph theorists refer to as a rooted tree). numerical simulations reveal that there are 2,674,440 possible reporting structures, five examples of which are shown in figure 9. figure 10 shows the vertical and horizontal mixing entropies for the n 5 15 tree family. the diagonal lines indicate lines of constant total entropy sv 1 sh. the flat reporting structure (tree a) has the minimum total entropy and the purely vertical structure (tree c) has the maximum total entropy. the hierarchical binary tree (b) has a net entropy that is intermediate between these two bounds. perhaps surprising is the fact that it has a maximum horizontal entropy for the class of trees with the same level populations. the horizontal mixing entropy measures the degree to which the units are spread evenly across the organization. consider the physics analogy of a gas: the entropy is maximized if the figure 9. example reporting structures with 15 nodes. each structure is labeled by the horizontal, vertical and total mixing entropies and a lexicographic string that encodes the reporting structure. article_14-17.indd 52 03/03/15 3:06 pm d. chappell et al. / defining the entropy of hierarchical organizations 53 gas uniformly fills its container. the entropy would be reduced if the gas molecules were confined to a smaller region within the container. thus, the horizontal entropy is maximized when subordinates are distributed evenly across a given level of the organization. as an organization evolves its entropy will evolve as well. one can imagine following the trajectory of an organization in the sv vs. sh “phase space” shown in figure 10. we constructed a simple simulation to follow the organizational entropy over time assuming two simple rules: (1) the total number of nodes in the organization remained constant, (2) single nodes were chosen at random and the reporting line was randomized. the aim of this simulation is not to model the dynamics of real-world organizations, but rather to gain insight into what characteristics the entropy measures are sensitive to. figure 8 shows the evolution of the vertical, horizontal and net entropies as a function of the iteration number. figure 9 shows the organization at three times during the simulation. the horizontal and vertical entropies appear to be roughly anti-correlated: as the vertical entropy rises, the horizontal entropy declines. this behavior is due to the conservation of nodes. removing a node from a highly populated level and adding it to a less populated level will reduce the horizontal entropy while increasing the vertical entropy. this anticorrelated behavior means that the variance of the total entropy is significantly less that the individual entropies. as a result of this effect, the phase-space trajectory will tend to fluctuate along the diagonals of constant total entropy. preliminary fourier analyses of the time series indicate that the statistical fluctuations of the entropies appear to roughly follow a | f (k)| ∝ k23/2 power law, which we speculate arises from the branching pattern of the trees. it is interesting to note that this power figure 10. structural entropy measures sv vs. sh for all possible trees with n 5 15 nodes. the diagonal lines indicate lines of constant total entropy sv 1 sh . the locations of organizational trees a, b and c are marked. article_14-17.indd 53 03/03/15 3:06 pm 54 d. chappell et al. / defining the entropy of hierarchical organizations law is intermediate between 1/f noise associated with critical phenomena and random walk processes on euclidean spaces. 4. connection with computational or kolmogorov entropy the kolmogorov entropy or computational complexity of an object is defined as the shortest possible description in bits of that object (cf. cover and thomas [2006]). while there are a number of theoretical challenges in defining this quantity, as a practical matter there are simple estimates for computational complexity that can be quite useful. we take the simple case of descripting the individuals within a level of a hierarchical organization. one way to estimate the computational complexity is to use the “lexicographic trick” [cover and thomas, 2006]. if we are to describe the assignments of individual workers within a level of the hierarchy, a very short description of this would be the following instructions: “make a lexicon of all the possible working assignments according to a prescribed rule. the desired description can be found on line n* of the lexicon”. the figure 11. evolution of the vertical, horizontal and total mixing entropies for a ternary hierarchical tree with 4 levels and 40 nodes. the iteration number counts the number of personnel changes over time. the three arrows correspond to the three reporting structures shown in figure 12. figure 12. evolution of a hierarchical tree given the dynamical rules described in the text. the trees correspond to iteration numbers 0, 500 and 3000 in figure 11. article_14-17.indd 54 03/03/15 3:06 pm d. chappell et al. / defining the entropy of hierarchical organizations 55 encoding of the instructions will actually be quite small and not make a large contribution to the complexity. in general, the largest contribution will come from the number n*. the encoding of a number will take log2 n* bits. in general, we will have λ so the encoding of this number, log2 λ represents an upper limit of the complexity. if there are n individuals in the level of the hierarchy and each unit had ni employees, then the number of possible combinations is: ∏ λ n n ! ii 5 (12) the computational complexity is then bounded by: ∑∑λ n n n n n flog log log logi i i i ii 2 2 2 25 2 5 2 (13) where stirling’s approximation to the logarithm of a factorial has been used and fi is defined as in eq.1. the complexity per number of employees, log2 λ/n then gives eq. 1. thus, the estimate of the computational complexity of the level in the hierarchy is identical to the entropy of mixing as in eq. 1. this shows that both the physical entropy approach and the computational complexity approach yield the same result for the organizational entropy. this argument is readily extended to the other calculations described by the physical (i.e. thermodynamic) entropy approach. 5. conclusion in this work, a mathematical definition of organizational entropy motivated by analogies with both thermodynamic and kolmogorov entropy is presented. as demonstrated in the text, this definition allows for the entropy to be readily calculated for any organization given the personnel distribution along reporting lines. the organizational entropy has two contributions: the horizontal entropy that measures the disorder within tiers and the vertical entropy that measure the disorder along the lineages or between tiers. as described earlier, there are a number of potential applications of organizational entropy. it can be us be used as an empirical parameter to monitor an organization over time and indicate when a restructuring might be appropriate. similarly, it might be used to classify structures designed for specific functions and indicate when an optimal structure has been achieved. the mathematical formalism presented here is a starting point for exploration of optimal organizational structures. using maximum entropy approaches (cf. cover and thomas [2006]), the statistical properties of organizational structures that maximize the entropy under different constraints can be determined. for instance, a “cost function” may be introduced that distinguishes the cost or value of personnel between tiers or between lineages. maximizing entropy under fixed costs would lead to a mathematical description of the optimal structure given these costs. again, exploiting the analogy with thermodynamics, the total cost could be related to the analogue of the energy the system and relationships between entropy and costs could be derived that are analogous to thermodynamic relationships between energy and entropy. such extensions of the present theory are under investigation. article_14-17.indd 55 03/03/15 3:06 pm 56 d. chappell et al. / defining the entropy of hierarchical organizations references atkins, p.w. (1997). physical chemistry. macmillan higher ed. carley, k.m. (1995). computational and mathematical organization theory: perspective and directions. computational and mathematical organization theory, 1, pp. 39–56. cover, t.m. and thomas, j.a. (2006). elements of information theory (2nd ed.). john wiley and sons. fabac, r. and stepanic, j. (2008). modeling organizational design applying a formalism model from theoretical physics. jios, 32, pp. 25–32. hummon, p.n. and fararo, t.j. (1995). actors and networks as objects. social networks, 17, pp. 1–26. kasac, j., stefancic, h., and stepanic, j. (2004). comparison of social and physical free energies on a toy model. phys. rev. e, 70, 016117. krackhardt, d. (1994). graph theoretical dimensions of informal organizations. in computational organizational theory, carley, k. and prietula, m. (eds.), (lawrence erlbaum associates, inc.), pp 89–111. mcgrath, c., krackhardt, d. and blythe, j. (2003). visualizing complexity in networks. comput. math organ. theory, 19, pp. 480–515. schlick, c.m., duckwitz, s., and schneider, s. (2013). project dynamics and emergent complexity. comput. math organ. theory, 19, pp. 480–515. stepanic jr., j., stefancic, h., zabec, m.s., perackovic, k. (2000). approach to a quantitative description of social systems based on thermodynamic formalism. entropy, 2, pp. 98–105. article_14-17.indd 56 03/03/15 3:06 pm process complexity 5 university of bamberg press process complexity author: jean boulton cranfield school of management, cranfield university, united kingdom; department of social and policy sciences, university of bath, united kingdom email: jboulton@claremont-mc.co.uk we develop the ontology of “process complexity” and describe how the dynamics of “becoming” can be framed as the emerging, stabilising, and ultimate dissolving of “patterns of relationships.” by extending traditional complexity thinking through introducing a “field theory” view, we develop a more nuanced and inclusive perspective of the processual complex world. we show how this leads to the idea of ontological uncertainty. we demonstrate how process complexity resonates with the ontologies of many different schools of thought including quantum gravity, the process philosophers of the axial age, and the early modern process philosophers impacted by darwin’s theory of evolution, such as bergson, whitehead, and james. the remarkable alignment of these diverse perspectives from science and philosophy adds conviction and depth to the development of process complexity. we conclude by indicating how process complexity influences our approach to policy and management practice. keywords: complexity; irreversibility; indeterminism; emergence; theory; ontology; process; uncertainty; processual; patterns; quantum gravity; daoism; process philosophy; ontological uncertainty introduction the focus of this paper is on the ontology of the complex world. we take as its foundation prigogine’s concern with the thermodynamics of open systems and link this to the deepening of a more processual turn in complexity thinking. the foundations of our work in complexity theory derives from the legacy of ilya prigogine and peter allen (boulton et al, 2015). prigogine was a physical chemist, and his abiding interest was in the irreversible nature of time and in the seeming contradiction between the physics of thermodynamics and theories of evolution of the living world. equilibrium thermodynamics suggests that the universe winds down and becomes less structured over time, but evolutionary theory shows how living systems tend to become more ordered, more sophisticated. prigogine’s nobel prize was awarded for exploring the thermodynamics of open systems, and showing that, locally, order can emerge out of chaos (prigogine & stengers, 1984). thus, a seeming contradiction between the behaviour of natural and physical systems was resolved. prigogine was one of the founding fathers of complexity theory and showed how the behaviour of complex situations is in part derived from the necessity of openness to their wider context. this paper is primarily about ontology. the central aim is to show that the ontological stance of processual complexity, following the tradition of prigogine, is almost indistinguishable from the ontology embedded in daoist process philosophy, from the 5th century bce, and, equally, resonates with the much more recent development of quantum gravity (rovelli, 2018). we are keen to illustrate how ideas that were developed within the physical and natural sciences closely match ideas abstracted from the lived experience of social beings. this aim to demonstrate ontological parity between ideas emerging from very different traditions may seem rather remote from the practical world of management and policy. jackson (2019) and morgan (1986) would argue, for example, that one can be pragmatic about selecting an ontological stance. given that no description of “reality” is ever incontrovertibly “true” or “right” and can never describe things in their entirety, then, they would argue, it is sufficient to pick an ontological lens or image that seems helpful and “works” for the problem at hand. whilst agreeing that there is no one unique all-encompassing description of anything (and indeed that is part of what this processual complex ontology is suggesting), the congruence between these radically different bodies of thought, from daoism to modern physics, adds strength and confidence in embracing this processual complex ontological stance, which we shall refer to as process complexity. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 5-14 doi: http://dx.doi.org/10.20377/cgn109 mailto:jboulton@claremont-mc.co.uk 6 university of bamberg press this still begs the question as to why we are of the view that surfacing and exploring issues of ontology is of such fundamental and practical importance. it is founded in the conviction that how we view the nature of things, how we view the way things stabilise or change, has a strong influence on what knowledge we deem important, and what methodologies we adopt in the world of strategy and policy. seeing the world as complex, uncertain, emergent rather than linear, predictable, and controllable has a significant impact on theories of policy development (room, 2011; haynes, 2015; boulton, 2010; geyer and rihani, 2010). complexity thinking is largely seen as an epistemological project, driven by computational modelling built on various levels of simplification (allen and boulton, 2011). a smaller number of authors focus on its ontological implications, e.g. martin and sunley (2007), holland (2015), bradbury (2006), byrne and callaghan (2014), and boulton et al (2015). yet our worldviews, our ontological positions, and our beliefs and biases, whether we are conscious of them or not, have huge power to shape what we see, how we interpret what we see, and what action we take. our ontological stance is deeply interwoven with our epistemological perspective and some authors speak of onto-epistemology (voros and riegler, 2017). some, indeed, speak of ethico-onto-epistemology (barad, 2007), to emphasise, as did paul cilliers (woermann & cilliers, 2016), that in a world that cannot entirely be determined, every choice – of perspective, of action is inevitably an issue of ethics. it is also important to point out that not all traditions in complexity theory place so much emphasis, as did prigogine, on “becoming,” on the role of variation, and the shaping of emergent patterns through history and context. uncertainty is both necessary and inevitable. the use of modelling in the world of complex systems can create implicit ontologies which implicitly suggest that complex systems are real things, are reified “entities” whose behaviour over time can be relatively well mapped out. even when there is recognition that the map is not the territory, this implicit reification can imbue and shape choices of policy, action, and interpretation. concern over this tendency to reify was the drive behind ralph stacey’s shift from complex adaptive systems to the more processual, contextual, and granular complex responsive processes (stacey, 2001). complex systems can sound like a mix of artefacts: strange attractors, power laws, simple rules, fractals, sensitivity to initial conditions, and edges of chaos (boulton, 2015:71). these artefacts are qualities of various mathematical models based on differing levels of simplification but are not necessarily or generally aspects of the “real world.” this pull towards reification can lead to preferencing classification over research, design over adaptation and outcome-focus over experimentation. prigogine’s work lends itself much more to framing complexity as a situated relational quality, seeing organisations or communities or economies as processes that are co-constituted by the people within them, impacted by their environment, and in relationship with other organisations and forms. the emphasis is more on “embracing” complexity as an inevitable and largely positive quality of the world, as a source of novelty, emergence, resilience, and change, rather than on viewing it as something to reduce, or manage, or categorise. this is not to suggest that there is complete polarisation of positions in the field of complexity, but there are without doubt differences of emphasis, and these are perhaps associated with differing implicit ontologies, not all of which have the processual flavour that is embedded within prigogine’s tradition. with this preamble in mind, there are three aims for this paper. first, we develop the ontology of “process complexity” and show how it derives from prigogine’s original approach to complexity theory and connects to process philosophies more generally. process ontologies emphasise the primacy of “becoming” over “being,” of flow over stasis. but this focus on flow does not necessarily emphasise what is the nature of that flow. process complexity adds a particular flavour to this and suggests that the flow of change can be envisaged as the emerging, stabilising and eventual dissolving of “patterns of relationships.” we show how this processual understanding of the complex world has irreversibility at its root and leads to an understanding of situations as constantly adapting, shaped but not determined by history, context-sensitive, and emergent. traditionally, complexity thinking works implicitly with the idea that the world can be modelled as “things” connected by “forces.” by introducing a processual view, which understands “things” as interpenetrating patterns which may be fleeting or persistent, we develop a more nuanced complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 5-14 doi: http://dx.doi.org/10.20377/cgn109 7 university of bamberg press and inclusive perspective of the complex world and find that the distinction between patterns and events itself dissolves. we then explain how this extension of prigogine-inspired complexity theory leads to the idea of ontological uncertainty. the second aim of the paper, interwoven throughout it, is to demonstrate how process complexity resonates with the ontological stance of very different perspectives: quantum gravity, process philosophies developed during the upheaval of the axial age, and those of early modern process philosophers such as whitehead, bergson, and james. the remarkable alignment of these understandings of the way things are in the world lends depth, richness, and salience to process complexity. the third aim of this paper is to comment, briefly, on the implications of adopting a process complexity ontology with its presumption of ontological uncertainty, for management practice and policy. prigogine and irreversibility: the roots of complexity thinking when ilya prigogine gave his nobel prize acceptance speech, he said he was entranced, as a young man, by reading henri bergson’s ‘creative evolution’ (1964) and by bergson’s question: “why does life mount the incline that matter descends?” that is to say, why does physical matter, according to the second law of thermodynamics, decay into featureless dust, whereas evolution leads, in many cases, to “transformations to higher levels of complexity” (prigogine, 1980:xii). prigogine’s answer was to recognise that for open systems that can exchange energy and information with the wider world, entropy can decrease locally, and “structure”—patterns of increased order—can emerge. as he said in his biography “living things provided us with striking examples of systems which were highly organized and where irreversible phenomena played an essential role” (prigogine 1977). order emerges out of chaos (prigogine & stengers, 1984). prigogine recognised that for this to happen, there were three conditions at play. first, situations need to be open in order to exchange energy, matter, and information with their surroundings; secondly, there needs to be fluctuations, diversity, messiness; and thirdly, the diverse elements need to be interconnected reflectively, that is with the possibility of feedback (prigogine et al, 1977:2). in this way, physics, in the form of the thermodynamics of open systems, and evolution meet. as biologist and fellow nobel prize winner jacques monod said (1972:118): “evolution in the biosphere is therefore a necessarily irreversible process, …indeed it is legitimate to view the irreversibility of evolution as the expression of the second law [of thermodynamics] in the biosphere.” situations that are irreversible follow a path that cannot be undone or reversed. things happen which then create the conditions for subsequent things to happen; there is no going back. this is what physicists call path dependence1. if we were to film a pond, or a period in the life of a family, or a mountain river, we would know if the film were played backwards. irreversibility can be understood as a loss of information. it happens in situations where more than one path could have occurred. in effect, the path that did emerge then closed down the potential for all other possible paths, and the information associated with those unrealised paths is lost. william james, writing in 1884 (bird, 1995:274-5) puts it like this: “possibilities may be in excess of actualities…of two alternative futures…. one becomes impossible only at the very moment when the other excludes it by becoming real itself.” stuart kauffman (2000) names this shaping of the future, as paths are selected and others are no longer available, as non-ergodicity. non-ergodic systems do not visit all of their possible states. he explains: “the evolution of life in our biosphere is profoundly “non-ergodic” and historical. the universe will not create all possible life forms. non-ergodicity gives us history.” irreversibility is at the root of complexity thinking. it is the source of path-dependence—the idea that history shapes but does not determine the future. it creates the “arrow of time.” it underpins 1 path dependence is used within social science in various ways. we are here adopting the approach of physicists, who are referring to the way the particularity of what happens shapes and limits (but does not determine) what happens next. they speak of the system having memory. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 5-14 doi: http://dx.doi.org/10.20377/cgn109 8 university of bamberg press the way the particularities of context, of sequence and of choice are fundamental to understanding how situations in the social and natural world evolve. irreversibility emphasises the need for a dynamic, processual understanding of situations and draws the importance of time and duration into the heart of complexity thinking. the core of prigogine’s contribution: how the interplay of patterns and events creates a focus on “becoming” rather than “being” as prigogine (1980) married the implications of evolutionary biology with the physics of open systems, he came to the recognition that we should think of the world not as stable—as “being”—but as irreversibly “becoming.” change is the norm and stability the exception. in general, there is an ongoing irreversible shift in conditions and qualities, an underlying potential for change, for evolution, for creativity and adaptability. we need to pay attention to the dynamic of situations. we need to follow this dynamic and trace what happens over time. it is not enough to interrogate the present; we need to explore how the past shaped the present and how the present is co-responsible for the future (cilliers, 2001:6). that is to say, the present shapes the future, in that it constrains possible future paths, but it does not fully determine which future path will ensue. the future is a complex weave of present conditions, chance, choice, impacts from the wider context, together with a sprinkling of messiness and variation. this process of “becoming” is not steady and incremental, but what boulton et al (2015) have called “episodic.” to call change episodic, or punctuated (tushman & romanelli, 1994; gould, 1982) is to signify that, in general, development over time consists of periods that are self-regulating, wobbly but stable, interspersed with periods of change during which new factors are likely to emerge. in the quest for ontological parity, it is noteworthy that this notion of episodic change, the idea that “the door is not always open” is captured in the following extract from a commentary on the dao de jing, a chinese philosophical text from the 5th century bc (ames and hall, 2003:83): “within the rhythms of life, the swinging gateway opens, and novelty emerges spontaneously to revitalise the world. whatever is most enduring is ultimately overtaken in the ceaseless transformation of things.” during periods of relative stability, the conditions in a given situation tend to self-organise, to settle down into identifiable patterns of relationships. for example, organisations or communities or families tend to develop relatively stable cultural norms of behaviour and routines. these are not fixed, but they can persist, and the persistence is due to the interdependent, reflective nature of relationships. as previously discussed (boulton et al, 2015), one key insight that came out of prigogine’s work (prigogine, allen and herman, 1977) is the way forms, patterns, and institutions emerge and become established. these patterns are then constantly challenged, and potentially invaded by the particular events, variations, decisions, shocks, and so on that take place in particular places at particular times, initiated both from within the situation of interest or from without. thus, we have an image of a world that is patterned, that is to say it has structure or form. and yet those patterns or forms are not fixed forever, but may collapse, dissolve, or shift into new patterns. boulton et al (2015:29) describe this as “a dance between patterns and events.” the “dance between patterns and events” focuses attention on the mechanisms by which complex situations change over time, and how new qualities emerge which could not have been predicted. patterns emerge and stabilise through the interdependent reflective relationships between the constitutive elements, coupled with their openness to exchange matter and information with the wider world. ralph stacey (2001), in his work on complex responsive processes, is focused on this perspective, on the importance of the particularity of the details of interactions and responses in understanding how patterns of meaning or routines of behaviour are established in the social world and how change emerges. to take an ecological example: in and around a pond, certain inter-relationships between species become established. however, over time, other species may gain traction, may “invade” these stable patterns of relationships and eventually replace them, aided perhaps through shifts of weather or changes to farming habits in the surrounding area. equally, new species or new conditions may not make much impact and may be transitory. so, we have an interplay between “events,” which might complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 5-14 doi: http://dx.doi.org/10.20377/cgn109 9 university of bamberg press include new seeds dropped by birds, or a particularly wet summer which preferences some species over others, and (in this example) ecological “patterns” which are constantly flexing and shifting and may indeed shift entirely into something new and unforeseen. this insight into the interplay between relatively stable patterns and how these are potentially invaded or destabilised by events is central to complexity theory following prigogine. as has already been noted (boulton et al, 2015), studying (relatively) stable patterns constitutes the primary fare of economists, marketeers, ecologists, and many systems thinkers. in contrast, “history” tends to deal with the particularity of events, conditions, and individuals. but complexity thinking, in the tradition of prigogine, marries the two and provides us with a sophisticated and distinct theory of change (allen, 1997:17). in summary, the ontology underpinning complexity theory in the tradition of prigogine is a version of process ontology, an ontology which emphasises becoming over being, which sees flow and change as the norm. prigogine-informed complexity theory adds a particular flavour to this process ontology, seeing this flow as textured, as consisting of the emerging and stabilising and eventual dissolving of patterns of relationships. how does this ontological image, emerging from the physical and natural sciences, relate to other ontologies? the focus on the way things change over time is, of course, of central concern in the field of process philosophy, both ancient and modern. daoist, buddhist, and pre-socratic thinkers from india, china, and the west during the axial age (puett and gross-loh, 2016) were thinking in this way; the substance for their deliberations was empirical in that it derived from lived experience. daoist thought in particular captures many of the insights from complexity theory, for example, the episodic nature of change, the emergence of novelty2, and in notions of pattern formation3. early modern process philosophers, such as william james, henri bergson and alfred north whitehead were substantially motivated by the processual implications of evolution, as of course was prigogine (seibt, 2020). are processes and events really different in kind? from prigogine to process complexity. the “restricted complexity” (morin, 2008) consistent with the computational modelling of complex systems can convey an idea that the world is made up of concrete “things” or elements connected by forces, albeit nonlinear forces. but things are made up of molecules and molecules are made up of atoms which share electrons. these electrons can be considered as waves or processes as much as particles. in the end, the “thingness” of things vanishes into interpenetrating fields of energy and probability. the notion of what is a “thing” and what is a “force” becomes blurred. we can instead envisage interactions akin to the way ripples interact on the ocean. ripples are identifiable entities, albeit temporary and with ill-defined boundaries. they impact each other, may merge, may separate again, and may disappear. we would not describe their interactions and impacts as caused by forces; they interact due to their own characteristics (of speed, amplitude, temperature), due to their unique nature, at a particular place and time. this is a so-called “field-theoretic”4 approach. thus, the description of the shaping of the future as a “dance between patterns and events” can be extended and reframed as a dance between long patterns and short patterns, or indeed between long events and short events. in this appeal to field theory, we move beyond positing “patterns” and “events” as distinguishable entities, as different in kind. we then arrive at a description of a processual ontology understood as patterns or processes, longer or shorter, larger or smaller, which are interpenetrating, intersecting, emerging, and dissolving. to illustrate with an example, think of a new 2 for example, “it is the under-determined nature of way-making that makes it inexhaustibly capacious; and gives the quite appropriate impression that nothing is perfect or complete.” (ames & hall, 2003:83) 3for example, “integrity is something becoming whole in its co-creative relationships with other things. integrity is consummatory (is the inevitable consequence of) relatedness.” (ames & hall, 2003:16) 4 field theory is a fundamental approach within physics, underpinning electromagnetism, sound, quantum theory, relativity, and so on. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 5-14 doi: http://dx.doi.org/10.20377/cgn109 10 university of bamberg press leader. we can envisage this leader as an event which happens to the organisation. but in actuality, that leader is part of other social patternings—family, community, peer group, and so on. should the leader be construed an event acting on the organisation or is the leader better thought of as part of preexisting, intersecting social patterns that engage with the organisational patterning in a way that changes and shifts and reshifts all of them? this processual view of complexity, this image of ripples intersecting on the ocean, has remarkable similarities with sciences and philosophies from many different traditions. for instance, from the field of quantum gravity, rovelli explains that “the world is made up of events not things. even the things that are most “thing-like” are nothing more than long events. the world is not so much made of stones as of fleeting sounds or waves moving through the sea” (rovelli, 2018:8588). in similar vein, ames and hall (2003:13), in reflecting on daoist ontology, argue that “particular things are in fact processual events and are thus intrinsically related to other things that provide them context… these processual events are porous, flowing into each other in the ongoing transformation we call experience.” this allowance of the interpenetration of “things” within complexity theory, an approach that physicists would recognise from field theories, provides a richer more subtle ontology than the implicit notion of “things connected by forces.” this paves the way for a richer understanding of the characteristics and qualities of the ever-emerging complex world. we can call this process complexity. what gives it ontological strength is that process complexity resonates strongly with such different disciplines as quantum gravity and process philosophy, both ancient and modern. more recently, those working with process theories of organisation have reached not dissimilar understandings (e.g. tsoukas, 2016 and 2006; hernes, 2014; chia, 1998). ontological uncertainty this paper is focused on the ontology of process complexity, an extended version of complexity theory building on the work of prigogine and its basis in irreversibility. one aspect of this ontology is the notion of ontological uncertainty: that the future is not knowable in its entirety because there is no one future. complexity theory emphasises that when new qualities emerge, there is no way of knowing for sure when this will happen and what these new features will be. new variables may appear which have not before been considered or seen. it is not that we know nothing; there may be signs of increasing instability or emerging newness, there may be indications of likely future states. but there remains a level of uncertainty that cannot be overcome. the question is, at times of emergence, are we dealing with the issue that the situation is too complicated for the modeller to both know everything pertinent and be able to include in a model? this is epistemological uncertainty. or are we dealing with something more fundamental, that even la place’s daemon sitting out there in the heavens could not know (morin, 2008), because uncertainty is woven into the very fabric of the universe. this is ontological uncertainty; what will happen at points of emergence cannot be known in its entirety, although neither is it random. william james (bird, 1997:274-5) understood this point. he says there is no “one unbending unit of fact.” things “may really in themselves be ambiguous,” not just to our understanding but because reality can have in and of itself an ambiguous quality. the notion of ontological uncertainty sits in contrast to the assertion that the universe is platonic, i.e., that it could in principle be described entirely by mathematics. if that were the case, then in effect everything, including evolution, would be pre-determined and unfold towards a pre-existent future. it is clear that the deterministic laws of physics take us a very long way towards understanding both the very big (for example the gravitational redshift following einstein’s relativity was measured decades after it was predicted) and the very small (as, for example, the prediction of the higgs boson). but, as leibniz pointed out to newton (alexander, 1956), newton’s laws have nothing to say about the particular structure of, say, our solar system or our galaxy. if the universe is entirely to be understood through mathematics, it would, in principle be entirely predictable if the initial conditions were known. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 5-14 doi: http://dx.doi.org/10.20377/cgn109 11 university of bamberg press to quote william james, what had been “laid down” would “absolutely appoint and decree” the future, which would have “no ambiguous possibilities hidden within its womb” and would operate like an “iron block” (bird, 1995:274-5). thus emergence, evolution, surprise, adaptation or indeed, for living beings, choice or freewill would either not exist or be accidental. lee smolin (smolin 1997:17) points out “the properties of things are not fixed absolutely, they arise from the interactions and relationships among the things in the world.” the universe evolves to have particularity—of galaxies and their solar systems, and their actual unique planets. the important point that we are bringing out through this foray into cosmology is that we cannot understand the particularity of where planets and galaxies are, and what forms they have taken, just through recourse to fixed natural laws. uncertainty and indeterminism happen at this level too. as prigogine (1980:vii) and lee smolin and brian swimme (2011) all suggest, the evolutionary principle seems to operate in the universe as a whole not just in the living world. “the fact that we live in a universe with structures at so many scales is so commonplace, so manifest, that it is easy to miss its significance” (smolin 1997:204). this worldview, congruent with complexity thinking, emphasises that uncertainty and messiness are necessary conditions for emergence, irreversibility, surprise, novelty, and change. and this ontological uncertainty, this indeterminism, also is what gives space for mystery, ambiguity, and paradox. and even at the level of the universe, there seems to be an intrinsic ontological paradox or ambiguity in its very fabric; it is an inescapable and necessary part of how things are. ontological uncertainty has huge implications for how we consider “reality,” what we regard as knowledge and how we seek it. kauffman (2008:231) said: “in this [complexity-framed] ...scientific worldview… we live in an emergent universe of ceaseless creativity in which life, agency, meaning, consciousness and ethics have emerged. our entire historical development as a species, our diverse cultures, and our embedded historicity, have been self-consistent, co-constructing, evolving, emergent and unpredictable. our histories, inventions, ideas and actions are also parts of the creative universe.” discussion we can of course, reach processual ontologies and uncertainty via routes that start in sociology or in philosophy. why, then, is it useful to start with a theory from the natural sciences and reach a similar point from that direction? in part, this focus on ontological uncertainty counters the not uncommon view that, if we have detailed enough experiments or models or methodologies, we will overcome and “manage” complexity and find the simple and deterministic mechanisms hidden in its depths. to be able to counter this argument from the same stable from which it emerged, that is from the physical sciences, confers a particular type of power in discourse. this is not to ignore the dangers of taking physics theories and applying them to the social world unquestioningly, as happened to newton’s theories during the enlightenment and to equilibrium thermodynamics and its assimilation into classical economics5. we are still reaping the outcomes of the assumptions that organisations behave like machines, or economic systems tend toward a natural equilibrium. but if we can show that the science of open systems is remarkably congruent with process philosophies both ancient and modern, then does that congruence not add weight to both? can we not feel more confident to explore concepts from complexity thinking and “try them on for size,” see if they are any help in guiding practices in management and policy? seeing how the science of complexity, particularly in its processual form, reaches a similar place from a starting point emerging from the natural sciences seems a valuable contribution. it can strengthen our confidence to adopt these approaches to working in the complex world. it moves the ontological quest from a choice driven by personal preference to one triangulated between science, philosophy, and empiricism. 5 walras (1874) famously said: “this pure theory of economics is a science which resembles the physico-mathematical sciences in every respect.” complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 5-14 doi: http://dx.doi.org/10.20377/cgn109 12 university of bamberg press implications we have written previously about the implications of complexity theory for policy (boulton, 2010), emphasising the need for systemic adaptive policy making which explicitly weaves together the economic, the environmental, the social, and the political, and allows for the unanticipated, for learning and for regrouping in light of unexpected outcomes or unforeseen events. policies need to be both more agile to respond to the unexpected and allow for a degree of customisation to meet the particularities of context and history. whilst the broad purpose and key principles of, say, education policy may be overarching, the way best to enact it could well be to allow local schools or regions to respond more adeptly to their own particular demographies, resources, and histories. the need for agility, adaptability, and contextuality in policy making is well-recognised (e.g. room, 2011). further to this understanding, the processual view of complexity emphasises how the complex world is interwoven through to the smallest scale, the tiniest of ripples, the most fleeting of interactions. thus, in thinking systemically about policy, it is not sufficient to develop separate policies and then connect them or negotiate between them. rather, policy and management processes must emerge from the detailed interweaving of these factors from the ground up. for example, when buying a car, the person does not separate the economic (what does it cost) from the social (this car suits the image i want to convey) from the political (should i even be buying a car in this age of global warming). these factors interweave in a way that is hard to deconstruct and together result in a decision. would we not do better, for example, to develop socio-environmental, rather than primarily environmental understandings and policies to tackle climate change (biggs et al, 2021)? if we put into battle economic arguments against social or environmental ones (for example in decisions to build a new airport) is it not often the case that the economic argument wins? there are many local examples as to where housing policy, infrastructure development, attention to the environment, and local commerce are not developed systemically and integratively and where each case for, say, new housing, is argued separately. a process complexity perspective supports the position that it is not sufficient to try and couple policies or engage in debates between policies; rather, policy-making should be systemic from its inception. the embracing of ontological uncertainty emphasises that the search for the right method arrived at through objective and rational analysis is largely futile. ontological uncertainty requires us to use a multiplicity of methods and involve a multiplicity of diverse participants to home in on what is happening and what might be useful in response. heron and reason (1997) call this “critical subjectivity.” our epistemology is necessarily subjective, imperfect, and contextual when there is no possibility of a defined objective reality. it emphasises the inevitability of ambiguity and paradox, the need to wrestle with, for example, balancing short term goals with long term aims, or working with the tensions between freedom and responsibility or between innovation and efficiency (mowles, 2015). it underlines that there is no perfect political ideology, or universally appropriate methods of management or change. we are not of course arguing that the recognition of paradox is unique to complexity, but its derivation from the world of science may be helpful in discouraging managers from searching for the perfect answer and help to shift the balance between analysing and experimenting. the implication of embracing ontological uncertainty is also to place centrally the issue of ethics. what we chose to focus on, what we regard as important information, or desired outcomes or appropriate methods is inevitably an ethical choice. if the world is not objectively real, then our choices, even when working “scientifically,” will always be influenced by what we and others value, what we see as relevant and appropriate (woermann & cilliers, 2016). ontological uncertainty underlines the importance of surfacing biases, beliefs and predilections in a world that is inevitably incomplete, subjective, and subject to the emergence of the radically new. there are further implications for practice, and these relate to emphasis on the processual, on the way the past shapes the present and the present shapes the future. for example, by making the effort to investigate the history of an organisation or a region, we can judge the strength and longevity complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 5-14 doi: http://dx.doi.org/10.20377/cgn109 13 university of bamberg press of cultural patterns to inform our strategies6. or, through positing that we can understand situations as emerging, stabilising, and dissolving patterns, we can consider which of those patterns we would aim to strengthen, e.g., through building relationships, sharing resources, and weaving common intentions, and which we would aim to disrupt, e.g., through “nudging,” or activism, or supporting outsider views, switching resources, or piloting alternatives. theories of systemic change can sometimes place more attention on what needs to be strengthened than on what needs to be disrupted, unfrozen, or indeed encouraged to collapse. one important practice in seeking to engender change is to be alert to emerging “weak signals” of change, even when these are small and subjective (scharmer & senge, 2016). daoism expresses a similar interest in the way change starts with the small and seemingly insignificant. jullien (2004:69) notes that, for daoists, “the skill [of the leader] lies in the earliest possible detection of the slightest tendencies that may develop. by spotting these almost before they have begun, before they have had time to emerge and manifest their effects, the leader will be in a position to foresee where they will lead.” recent discussions with shaw (2021) and melo (2020) demonstrate that there are those working with ideas akin to processual complexity who see this skill in spotting and responding to early signs of emerging change as one of the most important aspects of engendering change. finally, even contemplating a complexity mindset may engender a sense of humility, a greater willingness to experiment. it may erode unfounded confidence and certainty and suggest a different balance between learning through doing and planning through analysis. embracing a complexity mindset has the potential to open up new questions and new perspectives at the same time as constraining the desire for definitive answers. conclusion first, we have developed process complexity as an extension to the “things connected by forces” version of complexity theory that emerges from mathematical modelling and scientific rationalism. we have built on the work of prigogine, whose starting point was irreversibility and the thermodynamics of open systems. we then explored how this extension of complexity theory leads to the idea of ontological uncertainty and its connection to emergence. we have explored how process complexity resonates with the ontological stances of such very different perspectives as those from quantum gravity, daoism, cosmologists like smolin and swimme, and those of early modern process philosophers such as whitehead, bergson, and james. this congruence seems remarkable and lends conviction and richness to this ontological quest for understanding the world as complexity in process. this thinking leads on to questions of practice for shaping strategies and policies, for understanding change and for considering the implications for leadership in a complex, deeply interconnected and interdependent world. references alexander, h.g. (1956) the leibniz-clarke correspondence. manchester university press allen, p. (1997) cities and regions as self-organising systems. gordon and breach science publishers. allen, p & boulton, j. (2011) complexity and limits to knowledge: the importance of uncertainty. in allen, p., maquire, s., & mckelvey, b. (eds), the sage handbook of complexity and management. sage ames, r.t. and hall, d.l. (2003) dao de jing: a philosophical translation. ballantine books barad, k. (2007) meeting the universe halfway. quantum physics and the entanglement of matter and meaning. duke university press bergson, h. (1964) creative evolution. macmillan and co ltd (original work published in 1911) bird, g.h. (1995) william james: selected writings. everyman biggs, r, et al (2021) the routledge handbook of research methods for social-ecological systems. routledge. boulton, j. (2010) complexity theory and implications for policy development. emergence: complexity and organisation 12(2): 31-41. 6 see boulton et al (2015: 196-8) for further exposition. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 5-14 doi: http://dx.doi.org/10.20377/cgn109 14 university of bamberg press boulton, j., allen, p.m., & bowman, c. (2015) embracing complexity. oxford university press bradbury, r. (2006) towards a new ontology of complexity science. in perez, p. & batten, d. complex science for a complex world: exploring human ecosystems with agents. anu press, pp. 21–26. jstor, www.jstor.org/stable/j.ctt2jbhz2.8. byrne, d, & callaghan, g. (2014) complexity theory and the social sciences. routledge chia r. (1998) from complexity science to complex thinking: organization as simple location. organization 5(3): 341– 370. geyer, r. & rihani, s. (2012) complexity and public policy: a new approach to 21st century politics, policy and society. taylor and francis gould, s. j. (1982) punctuated equilibrium—a different way of seeing. new scientist 94 (apr. 15): 137-139 haynes, p. (2015) managing complexity in the public services. routledge hernes, t. (2014) a process theory of organisations. oxford university press heron, j. & reason, p. (1997) a participative inquiry paradigm. qualitative inquiry, 3 (3) holland, d. (2015) complex realism, applied social science and post-disciplinarity: a critical assessment of the work of david byrne. journal of critical realism 13 (5) pp 534-554. https://doi.org/10.1179/1476743014z.00000000042 jackson, m. (2019) critical systems thinking and the management of complexity. wiley jullien, f. (2004) a treatise on efficacy. university of hawaii press kauffman, s. (2008) reinventing the sacred. basic books kauffman, s. (2000) investigations. oxford university press martin, r. & sunley, p. (2007) complexity thinking and evolutionary economic geography. journal of economic geography vol 7(5), pp 573-601. oxford university press melo, a. (2020) performing complexity: building foundations for the practice of complex thinking. springer. monod, j. (1972) chance and necessity. william collins sons and co ltd morgan, g. (1986) images of organisation. sage morin, e. (2008) on complexity. hampton press. mowles, c. (2015) managing uncertainty. routledge prigogine, i. (1977) biographical notes on gaining the nobel prize. https://www.nobelprize.org/prizes/chemistry/1977/prigogine/biographical/ prigogine, i. (1980) from being to becoming: time and complexity in the physical sciences. w h freeman and company prigogine, i. & stengers, i. (1984) order out of chaos. bantam prigogine, i., allen, p.m., & herman, r. (1977) long term trends and the evolution of complexity. in laszlo, e., & bierman, j. (eds.) goals in a global community. pergamon press puett, m., & gross-loh, c. (2016) the path. penguin books rescher, n. (2000) process philosophy: a survey of basic issues. university of pittsburgh press. romanelli, e. & tushman, m. (1994) organizational transformation as punctuated equilibrium. the academy of management journal. 37 (5), pp. 1141-1166 room, g. (2011) complexity, institutions and public policy. edward elgar rovelli, c. (2018) the order of time. allen lane scharmer, o. & senge, p. (2016) theory u. berrett-koehler publishers seibt, j. (2020) process philosophy. in zalta, e.n. (ed.) the stanford encyclopaedia of philosophy (summer 2020 edition). <https://plato.stanford.edu/archives/sum2020/entries/process-philosophy/>. shaw, p. (2021) private communication smolin, l. (1997) the life of the cosmos. oxford university press swimme, b., & tucker, m.e. (2011) journey of the universe. yale university press stacey, r. (2001) complex responsive processes in organizations: routledge tsoukas h. (2006) complex knowledge: studies in organizational epistemology. oxford university, press: oxford. tsoukas, h. (2016) don't simplify, complexify: from disjunctive to conjunctive theorizing in organization and management studies. journal of management studies 54 (2) voros, s. & riegler, a (2010) a plea for not watering down the unseemly: reconsidering francisco varela’s contribution to science. constructivist foundations 13(1): 1-10. https://constructivist.info/13/1/001.editorial woermann, m. & cilliers, p. (2016) the ethics of complexity and the complexity of ethics, in critical complexity: collected essays. de gruyter complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 5-14 doi: http://dx.doi.org/10.20377/cgn109 https://archive.org/details/processphilosoph00resc institutional complexity is complexity with an adjective complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 82-93 doi: http://dx.doi.org/10.20377/cgn-101 82 university of bamberg press institutional complexity is complexity with an adjective author: marielle papin mcgill university, canada email: marielle.papin@mail.mcgill.ca a review of the studies on institutional complexity reveals that the many definitions of institutional complexity and related concepts share similarities with the understanding of complexity and complex systems of complexity science. yet few publications on institutional complexity engage explicitly with complexity science. most observers still confuse complicated and complex systems, for instance. furthermore, the variety of definitions may create disarray regarding what institutional complexity and its related concepts are and what they imply. highlighting the similarities between institutional complexity and complexity science in global governance, this think piece offers a conceptual and operational definition of institutional complexity using a complexity science lens. it highlights the attributes and properties of institutional complexity. it also presents the benefits of such an approach. besides offering advantages in terms of concept clarification, this approach aims to engage theoretically, epistemologically, and methodologically with the complexity of global governance, as well as propose a way to answer remaining questions on this crucial topic. keywords: institutional complexity; international relations; complexity science; conceptualisation many definitions of institutional complexity without much complexity an increasingly acknowledged property of global governance is its institutional complexity, visible in the increasing number and diversity of international institutions overlapping in addressing a given issue area of global governance (abbott, green & keohane, 2016; zelli, möller, & van asselt, 2017). studying institutional complexity is key to both our understanding of the functioning of global governance and our managing of the issues addressed through global governance. accordingly, scholars have looked at this phenomenon directly and indirectly through diverse perspectives and concepts. they have offered a plethora of definitions of institutional complexity and related concepts. these generous efforts have been fruitful, but they have also generated disarray as to what institutional complexity actually is. seeking to make sense of them, i perceive three analytical dimensions widely used in the study of institutional complexity, i.e. fragmentation, polycentricity, and the emergence of novelty.1 first, fragmentation has been a prominent issue in the literature on institutional complexity. it generally refers to the coexistence of different clusters of institutions working towards cooperation in global governance (kim, 2019). the image of global governance architecture, which describes the allencompassing system of institutions (whether public or private) working on a specific issue area in world politics (biermann, pattberg, van asselt, & zelli, 2009), is deeply linked to the concept of fragmentation. it emphasises the presence of distinct substructures, which may appear at distinct levels, comprising the wider system. similarly, albeit related specifically to the climate governance subfield, hybrid multilateralism describes the increasing interactions of state and non-state actors in global cooperation on climate issues (bäckstrand, kuyper, linnér, & lövbrand, 2017). doing so, this concept draws attention to actors and relationships the international relations literature has traditionally ignored. these 1 other scholars have paid attention to slightly different analytical dimensions of institutional complexity. zürn and faude (2013), for instance, add to the analysis of fragmentation the concepts of differentiation and coordination. more recently, kim (2019) identifies three main concepts in the discussion on institutional complexity, i.e. fragmentation, polycentricity, and complexity. building on this effort, i choose to look at the emergence of novelty rather than complexity, considering that the former captures a more specific process than the latter, which, as i show later on, is involved in all aspects of institutional complexity. mailto:marielle.papin@mail.mcgill.ca complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 82-93 doi: http://dx.doi.org/10.20377/cgn-101 83 university of bamberg press characterisations of global governance, or global climate governance, seek to show how entities interact and overlap constantly, increasing the complexity of the system. an important concern in this line of work lies in the conflicts to which institutional complexity may lead, generally understood as institutional overlaps (hofmann, 2011). institutional complexity publications specifically emphasise the diversity of international institutions focusing on an issue area and overlapping in their work (zelli et al., 2017). fragmentation often leads to conflicts and overlaps, yet it might also lead to synergies, interplay, cooperation, and connectivity (gómez-mera, morin, & van de graaf, 2020; kim, 2019). several scholars have studied these synergies through the analytical dimension of polycentricity. often used in environmental studies, polycentricity can be understood as a situation in which there is no central authority, but multiple interacting entities at distinct levels which adjust and self-organise (galaz, crona, österblom, olsson & folke, 2012; morrison, adger, brown, lemos, huitema, & hughes, 2017). using a polycentricity perspective helps scholars insist on the benefits of a decentralised approach to global governance (kim, 2019). building more on complexity-related concepts, some witness the emergence of a variety of regimes complexes, defined as “nested, partially overlapping, and parallel international regimes that are not hierarchically ordered” (alter & meunier, 2009, p. 13; see also gehring & faude, 2013; keohane & victor, 2011; orsini, morin, & young, 2013). as interacting systems regulating distinct issue-areas (gómez-mera et al., 2020), regime complexes underline both the fragmentation and polycentricity of global governance. similarly, studies on institutional complexes describe these as “complex interaction situations in which two or more international institutions interact to cogovern issue areas in international relations (raustiala & victor, 2004) and form interlocking structures (underdal & young, 2004, p. 374375) of global governance” (oberthür & stokke, 2011). also looking at substructures of global governance, bernstein and hoffmann (2019) identify a global carbon lock-in based on the existence of interdependent systems locked into fossil energy use. all these studies generally pay attention to the sturdiness of institutional structures in global governance. the third analytical dimension i identify in the literature on institutional complexity is the emergence of novelty. the interactions of multiple institutions, which imply synergy or conflict and may cause blockage, sometimes lead to the rise of new institutions or regimes (gómez-mera et al., 2020). regime complexes or institutional complexes generate order in global governance, which appears as a new property thereof (gehring & faude, 2013, 2014). interactions might also facilitate the emergence of new norms which might enable a governance breakthrough (klijn & koppenjan, 2014).2 other evolutions of norms include processes of norm cascade or institutionalisation (kelley, 2008; margulis, 2013). norm changes are often slow, but periods of punctuated equilibrium sometimes occur which enable rapid change (galaz, tallberg, boin, ituarte-lima, hey, olsson, & westley, 2017). one reason why novelties emerge is the diversity of parties involved in a regime complex treaty (hollway, morin, & pauwelyn, 2020). another reason might be the emergence of exploration processes through which negotiators go in order to develop new norms (morin, pauwelyn, & hollway, 2017). altogether, the three dimensions presented uncover prominent aspects of the institutional complexity of global governance. paying attention to interactions and their effects gives a clearer picture of the complex functioning of institutions in global governance. nevertheless, the multiplicity of concepts related to institutional complexity and definitions thereof also reveals a lack of consensus and possible confusion as to what institutional complexity actually is. numerous and diverse perspectives help give a wider understanding of a phenomenon and consider more aspects thereof. yet, they also make it harder for scholars to understand each other, discuss and answer specific questions collectively. 2 while klijn and koppenjan's publication focuses on public policy, its general observations apply to the institutional complexity of global governance as well. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 82-93 doi: http://dx.doi.org/10.20377/cgn-101 84 university of bamberg press resorting to complexity science might prove useful to deal with this problem. indeed, while the literature on institutional complexity highlights important aspects of complexity, it does so mostly without explicitly making use of the complexity science literature. complexity science represents a diverse body of literature seeking “to understand the coexistence of order and disorder and the constant dynamic transformations from one to the other.” (morçöl, 2012, p. 7) it studies many diverse natural and social complex systems, such as human brains, ant colonies, national or global economies or politics. because of its interest in both natural and social systems, it is relevant to analyse the institutional complexity of global governance. using a complexity science lens might help clarify the definition and implications of institutional complexity in the global governance context. it might mitigate issues of lack of consensus or coherence noted by several scholars (gómez-mera et al., 2020; kim, 2019). it might also help answer some remaining questions in this line of work related to the adaptation of systems of institutions (kim, 2019). in certain fields, such as the governance of climate change, where uncertainty forces us to adjust governance at the same time as knowledge evolves, understanding how systems of institutions adapt is crucial. taking stock of the past studies of institutional complexity, this think piece seeks to offer a conceptual and operational definition of institutional complexity building on complexity science and recent publications in international relations. the remainder of the piece is organised as follows. after underlining the similar arguments yet distinct vocabularies of global governance understandings of institutional complexity and complexity science definitions of complexity, i introduce one way to clarify the concept of institutional complexity using a complexity science lens. i then offer a conceptual and operational definition of institutional complexity through the lens of complexity science. building on former studies of complex systems in international relations, this definition includes a description of the attributes and properties of institutional complexity. to conclude, i stress the benefits of using a complexity science approach to analyse institutional complexity. between institutional complexity and complexity in complexity science the characteristics and properties of institutional complexity mostly parallel those of complex systems presented by the wide complexity science literature, including the few international relations studies that have used complexity science to analyse international phenomena. yet, the institutional complexity literature has not truly engaged with complexity science. in addition to often using a different vocabulary, it has tended to ignore some core tenets of complexity highlighted by complexity science, such as nonlinearity or the emergence of new properties. ‘precising’ (collier & levitsky, 1997; sartori, 1984) the concept of institutional complexity through a complexity science lens appears relevant. it could lead to several benefits, including differentiating institutional complexity from other notions and clarifying the meaning of related concepts, avoiding concept-stretching, and helping answer remaining questions in the literature. how complexity is understood in complexity science in order to highlight the similarities between institutional complexity and complexity in complexity science, it is necessary to provide more insights on complexity and complex systems from complexity science. the variety of concepts and definitions thereof in the institutional complexity literature echoes that of complexity and complex systems in the complexity science literature. defining complexity is tricky, if not impossible (cilliers, 1998). several authors indeed see complexity as deeep (i.e. difficult to explain, evolve, engineer or predict) (page, 2015, quoted pattberg & widerberg, in orsini, le prestre, haas, brosig, pattberg, widerberg, gómez-mera, morin, harrison, geyer, & chandler, 2019). it complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 82-93 doi: http://dx.doi.org/10.20377/cgn-101 85 university of bamberg press encompasses many ideas that scholars include or not in their arguments (orsini et al. , 2019). as bousquet and curtis put it, “complexity is best thought of as the array of concepts, methods and intuitions that emerged piecemeal from an engagement with specific non-linear, adaptive, emergent and/or selforganizing phenomena/problems that revealed the limitations of existing scientific approaches.” (2011, pp. 1-2) several authors have managed to provide descriptions of the defining characteristics of complex systems. orsini and le prestre highlight three features of complex systems, i.e. the interconnectedness of multiple diverse actors, operating at different levels, in an open system (in orsini et al., 2019). others stress the non-linearity, emergence, self-organisation, and openness of complex systems (bousquet & curtis, 2011; pattberg & widerberg, in orsini et al., 2019). young also insists on telecoupling, or the idea that the driver of a change is far removed from its effects in space or functions (2017, p. 93). telecoupling, nonlinearity, and the rise of emergent properties all lead to increased uncertainty. mitchell more broadly defines complex systems as “system[s] in which large networks of components with no central control and simple rules of operation give rise to complex collective behavior[s], sophisticated information processing, and adaptation via learning or evolution” (2009, p. 13). for le prestre (2017), it is crucial to distinguish between the attributes of complex systems and their properties, what constitutes them and what they produce, their causes and their consequences. doing so makes it easier to “[ask] questions regarding which characteristics of the system lead to what kind of properties” (2017, p. 135). orsini and le prestre (in orsini et al., 2019) identify three properties of complex systems, i.e. self-organisation, emergence, and adaptation. self-organisation reveals the lack of central authority in complex systems (mitchell, 2009). emergence is a crucial property, which some scholars of complexity science often consider to be as the core of complex systems (orsini & le prestre, in orsini et al., 2019; teisman & gerrits, 2014). holland (1992) sees parallelism, competition, recombination as the three mechanisms enabling the adaptation of systems. others see variety, diversity, and variations as factors of adaptability (boulton, allen, & bowman, 2015). similarities between the institutional complexity literature and complexity science drawing from the literature on institutional complexity and complexity science, table 1 identifies the most important elements of institutional complexity and complexity and complex systems.3 it shows that, in many cases, the two sets of literature refer to similar ideas, although they use distinct concepts.4 they usually note that the global governance system has become more crowded with institutions (alter & meunier, 2009; raustiala & victor, 2004). they agree that institutional density, or a large number of entities in a system, does not make a system complex (cilliers, 1998), yet they do argue that global governance has become more complex (gómez-mera, 2017). they perceive several common elements as accounting for complexity. 3 the analytical dimensions regarding the literature on institutional complexity do not apply to the literature on complexity science. 4 some exceptions include morin and collaborators (2017), kim (2019), gehring and faude (2014) and alter and meunier (2009). the latter notably link international regime complexity and complex systems. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 82-93 doi: http://dx.doi.org/10.20377/cgn-101 86 university of bamberg press institutional complexity literature complexity science fragmentation numerous and diverse institutions (bäckstrand et al., 2017; biermann et al., 2009; gómezmera, 2017; zelli et al., 2017) numerous and diverse entities (cilliers, 1998; le prestre, 2017; mitchell, 2009) institutional overlap, interplay or interdependence (biermann et al., 2009; gehring & faude, 2013; hofmann, 2011; zelli et al., 2017) interconnections of entities (boulton et al., 2015; le prestre, 2017; miller & page, 2007) polycentricity porous boundaries of regime complexes, regime shifting (gómez-mera et al., 2020; orsini et al., 2013) openness (boulton et al., 2015; le prestre, 2017; cilliers, 1998) involvement of distinct levels, multilevel governance (bäckstrand et al., 2017; biermann et al., 2009; kim, 2019; morrison et al., 2017; raustiala & victor, 2004) hierarchy (le prestre, 2017) adjustments of entities in a polycentric system (galaz et al., 2012; kim, 2019; morrison et al., 2017) adaptation (holland, 1992; le prestre, 2017; mitchell, 2009) order (gehring & faude, 2013), niche selection or competition (gómez-mera et al., 2020 ; stokke, in oberthür & stokke, 2011) self-organisation (boulton et al., 2015; le prestre, 2017) novelty emergence new properties and norms in global governance, norm institutionalisation (hollway et al., 2020; kelley, 2008; margulis, 2013; morin et al., 2017; gehring & faude, 2013) emergence (boulton et al., 2015; miller & page, 2007; mitchell, 2009; morçöl, 2012; teisman & gerrits, 2014) punctuated equilibrium and rapid changes, norm cascade (galaz et al., 2017; kelley, 2008) nonlinearity (cilliers, 1998; miller & page, 2007; young, 2017) and unpredictability (le prestre, 2017) table 1 important elements identified in the institutional complexity literature and complexity science regarding attributes, both approaches first highlight a high number of diverse entities. second, they are interested in the interactions (direct or indirect) of those entities. this is visible in their focus on interplay, overlap, interlinkages, or interconnectedness among the institutions or entities studied. more specifically, the two approaches consider the high degree of interactions among institutions as a crucial element of complexity. these interactions make institutions prone to becoming interdependent, which increases the level of complexity. besides, both sets of theories witness the entanglement of distinct levels of action, although the institutional complexity literature visibly engages less with the multiplicity of hierarchical levels. furthermore, there seems to be some porosity or flexibility in the borders of institutional complexes or regimes. they exchange information with their environments, which participate in shaping them. as orsini and collaborators describe, “perceptions draw the boundaries of the complex, indicating which regimes are recognized—or not—as elements of a complex.” (2013: 31) this echoes the openness of complex systems, which interact with their environment (le prestre, 2017). regarding properties, a few scholars explicitly discuss the self-organisation and presence of emergent properties in institutional complexes, although they are often those who already use a complexity complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 82-93 doi: http://dx.doi.org/10.20377/cgn-101 87 university of bamberg press science lens (gehring & faude, 2013, 2014). others more implicitly indicate the presence of selforganisation processes by looking at more specific phenomena. for instance, processes of niche selection or competition among institutions of regime or institutional complexes highlight the need for institutions to revise their mission to survive in a crowded system (gómez-mera et al., 2020; stokke, in oberthür & stokke, 2011). similarly, some researchers indirectly point to the adaptability of these complexes by looking more into how entities of a complex might adjust their goals to international concerns or changes (galaz et al., 2012; kim, 2019). by and large, these similarities reveal that, on many occasions, scholars working on institutional complexity use concepts that are close to those of complexity science, while not referring directly to it. building on this observation, i argue that we could make the definition of institutional complexity more precise and gain in clarity by using a complexity science lens and the vocabulary thereof. “precising” the definition of institutional complexity what ‘concept precising’ is about referring explicitly to complexity science when discussing institutional complexity is relevant and could help make the concept more precise. ultimately, this could bring several benefits to the study of the phenomenon institutional complexity represents. to build a complexity science definition of institutional complexity, i draw from several political science publications that stress the importance of concept formation (collier & levitsky, 1997; gerring, 1999; goertz, 2005; sartori, 1970). according to goertz, “concepts are theories about ontology: they are theories about the fundamental constitutive elements of a phenomenon” (2005, p. 5). when specifying the concept of institutional complexity, one builds a theory about the phenomenon to which it refers and, more generally speaking, about global governance. furthermore, sartori insists on the idea that one cannot measure the extent of a phenomenon before defining the concept to which it refers: “concepts are not only elements of a theoretical system, but equally tools for fact-gathering, data-containers.” (1970, p. 1052) concepts are thus also important for empirical testing. sartori also argues that we need to accept the tradeoff between avoiding conceptual stretching (or “vague, amorphous conceptualizations,” at p. 1034) by making our concepts more abstract, and enabling differentiation among similar, yet distinct ideas. for collier and levitsky (1997) however, a trade-off is not always necessary. other strategies for concept innovation, such as determining the sub-types of a concept, “precising” the concept, or shifting to the overarching notion of which it is part, can prevent this trade-off. i envision the use of a complexity science lens as a way of “precising” the concept of institutional complexity. making the concept of institutional complexity more precise by adding defining criteria might help differentiate similar phenomena, such as institutional density or complicated institutional systems (which are still reducible, while complex ones are not). it might also facilitate the identification of mechanisms inherent to adaptation processes, such as parallelism or recombination. doing so, concept “precising” also avoids conceptual stretching by preventing scholars from using the concept of institutional complexity when looking at any institutionally crowded issue-area. the goal of this operation is also to make explicit some criteria that are implicitly understood as part of the definition of institutional complexity, yet not expressed as such (collier & levitsky, 1997). below, i specify the concept of institutional complexity by enumerating its attributes and properties and highlighting some possible indicators as a way of operationalising the definition and making institutional complexity ultimately more easily observable and measurable. i build on orsini and complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 82-93 doi: http://dx.doi.org/10.20377/cgn-101 88 university of bamberg press le prestre’s efforts to define complex systems in international relations.5 i apply their division of attributes and properties of complex systems to institutional complexity and add indicators for each element, which help operationalise the definition of institutional complexity. thus, figure 1 illustrates the conceptual and operational definition of institutional complexity highlighting its attributes and properties. on the left of the institutional complexity case appear the attributes (or causes), while on the right appear the properties (or consequences). the far left and far right cases are possible indicators of the different variables presented. the attributes and properties of institutional complexity i see institutional complexity as a dimension of global governance. as such, it bears the attributes of complex systems underlined by orsini and le prestre as open systems of multiple interconnected entities of diverse types that have nonlinear interactions and operate at various levels (in orsini et al., 2019). it also has the same properties, i.e. self-organisation, emergence, and adaptability. thus, i define institutional complexity as the multiplicity and diversity of interconnected international and transnational institutions (operating at distinct levels or in diverse sectors, interacting, or overlapping in their work on a specific issue area), in the open system of global governance; altogether, these institutions are likely to self-organise, have emerging systemic properties, and adapt to their environment. institutional complexity has several defining attributes which one can observe using diverse indicators. the multiple institutions operating in the system are present in different sectors and levels, thus mixing public, private, local, national, regional, and international entities. this is particularly visible in the global climate governance arena, in which the mix of actors of different types and levels of action has been increasingly prominent, leading some to talk about a transnational climate regime complex (abbott, 2012). figure 1 institutional complexity, its attributes, properties, and indicators 5 i consider orsini and collaborators’ forum an international relations' import from complexity science rather than a publication belonging to the institutional complexity literature. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 82-93 doi: http://dx.doi.org/10.20377/cgn-101 89 university of bamberg press the second attribute of institutional complexity lies in the interconnectedness of institutions. units of complex systems have different types of interactions. they might interact directly through collaboration, coordination, or competition, for instance (e.g. in the climate regime, the united nations framework convention on climate change, and the convention of biological diversity, see zelli, gupta, & van asselt, 2013). another type of interaction might consist of one institution creating another institution (e.g. in the trade regime, the mercosur and the framework treaty on the environment of the mercosur). institutions might also interact indirectly. for instance, institutions that overlap interact indirectly, because they focus on the same issue-areas (e.g. the forest carbon partnership facility, the forest investment programme, and the un collaborative programme on reducing emissions from deforestation and forest degradation, see zelli et al., 2017), or include many members in common (e.g. world trade organization, and the montreal protocol). the interactions of the entities of global governance understood as a complex system might be nonlinear: the interconnectedness of entities and the openness of the system make it likely that some interactions might have an unexpectedly strong influence on others, and vice-versa. nonlinearity may also be combined with telecoupling. the last attribute of institutional complexity is openness, which is the capacity of complex systems to interact with their environment (orsini & le prestre, in orsini et al., 2019). it appears slightly harder to observe than the other attributes mentioned above in the literature on institutional complexity. yet, the case of the plant genetic resources regime complex shows, over time, an increase in the number of institutions involved in the regime complex, and the boundaries of the elemental regimes appearing progressively blurred (raustiala & victor, 2004). the boundaries of regime complexes might also appear porous. frequent interactions with the environment might cause them to change. issue shifts are possible. for instance, in the global health governance complex system, the interactions of health, economic, and development agendas have led to a paradigm shift (hill, 2010). perceptions of internal and external actors can influence these boundaries. alter and meunier elaborate on this idea through the notion of regime shifting: “in regime-shifting, actors may use forum-shopping, strategic inconsistency, or other strategies with the ultimate goal of redefining the larger political context so as to ultimately reshape the system of rules itself” (2009, p. 17). institutional complexity also has several properties, which one can understand as consequences of its attributes, or causes. self-organisation lies in the absence of a clear authority and the order of the system through the effects of the system itself and its interactions (orsini & le prestre, in orsini et al., 2019). for cilliers, it is “a process whereby a system can develop a complex structure from fairly unstructured beginnings” (1998, p. 12). self-organisation stems from the interplay and overlap of institutions, which lead the system to organise by creating, for instance, division of labour among the distinct institutions. there can still be overlap, but many institutions tend to differentiate themselves from others at least in some aspects. abbott’s transnational climate change governance triangle illustrates both functional differentiation and overlap (2012, p. 575). types of transnational climate actors vary (i.e. actors belong to state, civil society organisation, or firm actor groups, and might be state-led, collaborative, or private-led). these diverse types match diverse functions (i.e. standards and commitments, operational, information and networking, and financing). a variety of similar actors have the same function, which shows functional overlap. no central authority appears to be coordinating the distribution of functions. despite some efforts to orchestrate the global climate governance subsystem, the unfccc has failed to play that part. its support to the actions of the elements of the subsystem have surely had a role to play in the development of the climate regime complex, yet many transnational initiatives have developed outside of this framework. another property of institutional complexity is the possibility of new systemic properties emerging. emergence is the notion of unexpected systemic outcomes, which arise from the interactions of the system’s entities (orsini & le prestre, in orsini et al., 2019). put differently, “emergence is a complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 82-93 doi: http://dx.doi.org/10.20377/cgn-101 90 university of bamberg press phenomenon whereby well-formulated aggregate behavior arises from localized, individual behavior” (miller & page, 2007, p. 46). for instance, envisioning the earth as a natural complex system, its climate may be seen as an emergent property (kim & mackey, 2014). here again, the literature on institutional complexity provides few explicit examples of emerging properties. a valuable indicator to observe emergence in global governance is cooperation. for instance, the global governance field of trafficking in persons has grown from almost nothing in twenty years. since the 2000 un protocol to prevent, suppress, and punish trafficking in persons, especially women and children, cooperation has evolved to make it an important transnational regime complex (gómez-mera, 2017). another indicator to which one should pay attention is balance of power in the institutions of global governance. after the cold war, the global governance system experienced the rise of a unipolar configuration as an emerging property influencing international outcomes. this unipolar configuration has led, for instance, to the reinforcement of financial institutions demanding open markets and free trade (finnemore, 2009). these elements illustrate emerging properties of institutional complexity. finally, the last property of institutional complexity is adaptability. adaptability can be understood as “the potential capacity of units to learn from and coevolve with their environment” (orsini & le prestre, in orsini et al., 2019). units adapt to changes in their environment, which helps the system adapt. adaptability is visible in the systemic responses to the environment through changes in the rules, creation, and disappearance of institutions. for instance, international environmental law has adapted to new circumstances through decisions based on the evolution of science and knowledge taken in conferences of the parties (kim & mackey, 2014). the rules of the convention on international trade in endangered species of wild fauna and flora thus regularly evolve regarding targeted endangered species in light of new information on their status. the last two properties, i.e. emergence and adaptability, as well as the openness attribute appear to be understudied in the literature on institutional complexity. gehring and faude (2013) mention the question of emergence in regime complexes but they do not elaborate. likewise, some studies of polycentricity indirectly bring up the question of adaptation through the notion of adjustment (morrison et al., 2017). kim (2019) also refers to it as an issue that needs to be further investigated. the example of international environmental law above thus draws from the international relations complexity science literature. more empirical analysis is necessary to emphasise the presence of these elements in global governance institutional settings. overall, a variety of examples show the relevance of using a complexity science perspective to study institutional complexity. yet, as pertinent as ‘precising’ the definition of institutional complexity might appear, discussing the benefits of such an enterprise is crucial. what are the benefits of this approach? the study of institutional complexity has so far helped go beyond the consideration of the increasing number of international and transnational institutions and pay attention to several aspects of complexity. in that sense, it represents a step forward in the acknowledgement of the growing complexity of global governance and its consequences. yet, it is important to engage more with a variety of challenges linked to complexity. ‘precising’ the definition of institutional complexity using complexity science could help us do so. first, there are benefits related to conceptual formation itself. as argued by gerring (1999), concept formation not only offers more clarity, but it is also an attempt to respond to the following demands: familiarity, resonance, parsimony, coherence, differentiation, depth, theoretical utility, and field utility. for sartori (1970), concept formation allows for better conscious empirical research. concepts, after all, are data containers. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 82-93 doi: http://dx.doi.org/10.20377/cgn-101 91 university of bamberg press second, using a complexity science lens to study institutional complexity is important for theoretical, epistemological, and methodological reasons. the reductionist tradition seeking to decompose systems into individual parts has been prominent in international relations. yet, this newtonian and mechanistic thinking has revealed its limits in the study of global governance. while complexity theories remain marginal in international relations (notwithstanding kavalski’s 2007 claim of the emergence of a fifth debate), the discipline could benefit from their integration. indeed, they offer theories of change it may lack. regime theory, for instance, has favoured a static view of regimes (hoffmann, 2006). however, a complexity approach might help better account for its dynamic dimension and facilitate the production of new theories explaining the vivid and unpredicted changes of global governance over the last decades. regarding epistemological implications, instead of trying to beat unpredictability, scholars should accept it and agree to step away from the recurring and related goals of explaining and predicting. a complexity science lens offers new ways to describe and understand global governance. scholars should pay particular attention to the benefits of description. instead of trying to predict the unpredictable, complexity science might help us generate plausible scenarios of what might happen in a certain context (van der ven, bernstein, & hoffmann, 2017). this could help us establish strategies and policy interventions for better anticipation, an optimal goal for social sciences. complexity approaches might not be the only way to study complex problems, yet they surely are a valuable one. regarding methodology, defining institutional complexity through the lens of complexity science may foster the use of appropriate, yet still understudied methods that focus, for instance, on causal complexity and interconnections. a wide range of methods might help further investigate the complexity of global governance, including network analysis, system dynamics modelling, or scenario building. finally, and most importantly, the use of a complexity lens in the study of institutional complexity could help us answer remaining questions such as that of the adaptation of systems of institutions in a changing environment. this piece has mentioned several issue-areas in which institutional complexity is prominent, among which climate change. considering the uncertainty regarding the future of the earth’s climate and the constant need to adapt the measures taken and the governance instruments implemented to new scientific discoveries and political demands, understanding how not only climate governance institutions but also and most importantly their system may adapt is crucial. the adaptation of complex systems has been the subject of many studies in complexity science (holland, 1992; miller & page, 2007; young, 2017). these studies have analysed mechanisms such as variation, parallelism, or recombination which the institutional complexity literature has not tackled, although doing so might provide answers to its own questionings. several elements of the complexity science toolkit are likely to help us answer this question and others with which, as scholars of institutional complexity, we still struggle. overall, this think piece has tried to specify the definition of institutional complexity using principles of complexity science. depending on the subsystem considered, global governance may display more or less institutional complexity. therefore, when scholars talk about institutional complexity, going beyond the number and diversity of institutions at play in global governance may prove useful. in other words, one may see institutional complexity as complexity with an adjective. in that context, it is crucial to examine some fundamental principles of complex systems that apply to this concept, especially those that have been overlooked by institutional research so far (e.g. adaptability). this think piece is just a preliminary attempt at agenda-setting that needs further applications and refinements. hopefully it will achieve its goal of encouraging students of institutional complexity to engage more with complexity and see its opportunities rather than be held up by its challenges. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 82-93 doi: http://dx.doi.org/10.20377/cgn-101 92 university of bamberg press references abbott, k. w. (2012). the transnational regime complex for climate change. environment and planning c-government and policy, 30(4), 571-590. abbott, k. w., green, j. f., & keohane, r. o. (2016). organizational ecology and institutional change in global governance. international organization, 70(2): 247-277. alter, k. j., & meunier, s. (2009). the politics of international regime complexity. perspectives on politics, 7(1), 13-24. bäckstrand, k., kuyper, j. w., linnér, b.-o., & lövbrand, e. (2017). non-state actors in global climate governance: from copenhagen to paris and beyond. environmental politics, 26(4), 561-579. bernstein, s., & hoffmann, m. j. (2019). climate politics, metaphors and the fractal carbon trap. nature climate change, 9, 919925. biermann, f., pattberg, p., van asselt, h., & zelli, f. (2009). the fragmentation of global governance architectures: a framework for analysis. global environmental politics, 9(4), 14-40. boulton, j. g., allen, p. m., & bowman, c. (2015). embracing complexity: strategic perspectives for an age of turbulence. oxford: oxford university press. bousquet, a., & curtis, s. (2011). beyond models and metaphors: complexity theory, systems thinking and international relations. cambridge review of international affairs, 24(1). cilliers, p. (1998). complexity & postmodernism: understanding complex systems. new york: routledge. collier, d., & levitsky, s. (1997). democracy with adjectives: conceptual innovation in comparative research. world politics, 49, 430-451. finnemore, m. (2009). legitimacy, hypocrisy, and the social structure of unipolarity: why being a unipole isn’t all it’s cracked up to be. world politics, 61(1), 58-85. galaz, v., crona, b., österblom, h., olsson, p., & folke, c. (2012). polycentric systems and interacting planetary boundaries — emerging governance of climate change—ocean acidification—marine biodiversity. ecological economics, 81 (c): 2132. galaz, v., tallberg, j., boin, a., ituarte-lima, c., hey, e., olsson, p., & westley, f. (2017). global governance dimensions of globally networked risks: the state of the art in social science research. risk, hazards & crisis in public policy, 8(1), 4-27. gehring, t., & faude, b. (2013). the dynamics of regime complexes: microfoundations and systemic effects. global governance, 19, 119-130. gehring, t., & faude, b. (2014). a theory of emerging order within institutional complexes: how competition among regulatory international institutions leads to institutional adaptation and division of labor. review of international organizations, 9, 471-498. gerring, j. (1999). what makes a concept good? a criterial framework for understanding concept formation in the social sciences. polity, 31(3). goertz, g. (2005). social science concepts: a user's guide. princeton: princeton university press. gómez-mera, l. (2017). regime complexity and global governance: the case of trafficking in persons. european journal of international relations, 22(3), 566-595. gómez-mera, l., morin, j.-f., & van de graaf, t. (2020). regime complexes. in f. biermann & r. e. kim (eds.), architectures of earth system governance: institutional complexity and structural transformation (pp. 137-157). cambridge: cambridge university press. le prestre, p. (2017). global ecopolitics revisited: towards a complex governance of global environmental problems. new york: routledge. morçöl, g. (2012). a complexity theory for public policy. new york: routledge. hoffmann, m. j. (2006). beyond regime theory. complex adaptation and the ozone depletion regime. in n. e. harrison (ed.) complexity in world politics: concepts and methods of a new paradigm (pp. 95-121). albany: state university of new york press. holland, j. h. (1992). complex adaptive systems. daedalus, 121(1), 17-30. hollway, j., morin, j.-f., & pauwelyn, j. (2020). structural conditions for novelty: the introduction of new environmental clauses to the trade regime complex. international environmental agreements: politics, law and economics, 20(1), 61-83. hill, p. s. (2010). understanding global health governance as a complex adaptive system. global public health, 6(6), 593-605. hofmann, s. c. (2011). why institutional overlap matters: csdp in the european security architecture. journal of common market studies, 49(1), 101-120. kavalski, e. (2007). the fifth debate and the emergence of complex international relations theory: notes on the application of complexity theory to the study of international life. cambridge review of international affairs, 20(3), 435-454. kelley, j. (2008). assessing the complex evolution of norms: the rise of international election monitoring. international organization, 62: 221-255. keohane, r. o., & victor, d. g. (2011). the regime complex for climate change. perspectives on politics, 9(1), 7-23. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 82-93 doi: http://dx.doi.org/10.20377/cgn-101 93 university of bamberg press kim, r. e. (2019). is global governance fragmented, polycentric, or complex? the state of the art of the network approach. international studies review, viz052, 1-19. kim, r. e., & mackey, b. (2014). international environmental law as a complex adaptive system. international environmental agreements: politics, law and economics, 14(1), 5-24. klijn, e.-h., & koppenjan, j. (2014). complexity in governance network theory. complexity, governance & networks, 1(1), 61-70. margulis, m. e. (2013). the regime complex for food security: implications for the global hunger challenge. global governance, 19: 53-67. miller, j. h., & page, s. e. (2007). complex adaptive systems: an introduction to computational models of social life. princeton: princeton university press. mitchell, m. (2009). complexity : a guided tour. new york: oxford university press. morin, j.-f., pauwelyn, j., & hollway, j. (2017). the trade regime as a complex adaptive system: exploration and exploitation of environmental norms in trade agreements. journal of international economic law, 20(2), 365-390. morrison, t. h., adger, w. n., brown, k., lemos, m. c., huitema, d., & hughes, t. p. (2017). mitigation and adaptation in polycentric systems: sources of power in the pursuit of collective goals. wires climate change, 8(5). oberthür, s., & stokke, o. s. (eds.). (2011). managing instiutional complexity: regime interplay and global environmental change. cambridge: mit press. orsini, a., le prestre, p., haas, p. m., brosig, m., pattberg, p., widerberg, o., gomez-mera, l., morin, j.-f., harrison, n. e., geyer, r., chandler, d. (2019). forum: complex systems and international governance. international studies review, 131. orsini, a., morin, j.-f., & young, o. r. (2013). regime complexes: a buzz, a boom or a boost for global governance? global governance, 19(1), 27-39. raustiala, k., & victor, d. g. (2004). the regime complex for plant genetic resources. international organization, 58(2), 277309. sartori, g. (1970). concept misformation in comparative politics. the american political science review, 64(4), 1033-1053. sartori, g. (ed.). (1984). social science concepts: a systematic analysis. beverly hills: sage publications. teisman, g., & gerrits, l. (2014). the emergence of complexity in the art and science of governance. complexity, governance & networks, 1(1), 17-28. van der ven, h., bernstein, s., & hoffmann, m. (2017). valuing the contributions of non-state and subnational actors to climate governance. global environmental politics, 17(1), 1:20. young, o. r. (2017). governing complex systems: social capital for the anthropocene. cambridge: the mit press. zelli, f., gupta, a., & van asselt, h. (2013). institutional interactions at the crossroads of trade and environment: the dominance of liberal environmentalism? global governance, 19(1), 105-118. zelli, f., möller, i., & van asselt, h. (2017). institutional complexity and private authority in global climate governance: the cases of climate engineering, redd+ and short-lived climate pollutants. environmental politics, 26(4), 669-693. zürn, m., & faude, b. (2013). on fragmentation, differentiation, and coordination. global environmental politics, 13(3), 119-130. self-organization in collective action: elinor ostrom’s contributions and complexity theory of this paper and the two anonymous reviewers’ helpful comments and guidance on the submitted version of the paper. complexity, governance & networks (2014) 9–22 9 doi: 10.7564/14-cgn14 self-organization in collective action: elinor ostrom’s contributions and complexity theory* göktuğ morçöl pennsylvania state university at harrisburg e-mail: gxm27@psu.edu elinor ostrom’s contributions to the understanding self-organization in collective action processes are discussed from a complexity theory perspective. it is argued that complexity researchers can learn from ostrom’s theory building process, as well her conceptualization of the conditions of self-organization in the management of common-pool resources. her focus on self-organization helps rectify the problems with the assumption in the mainstream policy analysis that policy processes can be explained with external causes. the conceptual problems in her utility maximizing rational actor assumption and the potential for conceptual advancements in her recognition of complexity concepts are discussed. it is argued that ostrom’s conceptual framework is sophisticated, but it lacks a dynamic understanding of the micro–macro relationships in complex governance systems, and that complexity theory offers the conceptual tools to remedy this problem. keywords: collective action, complexity theory, elinor ostrom, self-organization, rational choice complexity theory concepts like self-organization can help us better understand governance processes. self-organization is not original to complexity theory; the concept has a long history and it has been used by other theorists as well. it is important to trace the lineage of the concept and acknowledge the contributions of others, so that the contributions of complexity theory can be distinguished. in this paper, i focus on the works of the late elinor ostrom, not only to recognize her contributions, which are substantial, and show how complexity theorists can learn from them, but also to highlight the shortcomings in her conceptualizations and point to the areas in which complexity theorists can offer better conceptualizations. ostrom’s primary contribution is her demonstration that self-organization happens under specific conditions and within specific domains in collective action processes. she and her colleagues conceptualized and empirically verified the conditions of self organization particularly in the management of common-pool resources (ostrom, 1990, * i want to express my appreciation for stefan verweij’s insightful and incisive comments on an earlier version article_14-14.indd 9 03/03/15 3:05 pm 10 g. morçöl / self-organization in collective action 2005, 2007a, 2009; ostrom, gardner, & walker, 1994). ostrom and her colleagues observe that self-organization happens in structured environments. the idea that self organizational ability of a system depends on external boundary conditions is not alien to complexity researchers (e.g., moreno & ruiz-mirazo, 2007), but they do not elaborate on the implications of this idea for complex governance systems/networks. ostrom and her colleagues conceptualized and empirically verified the structural characteristics that enable, or hamper, self-organizational dynamics in collective action processes. ostrom’s conceptualizations have shortcomings, as well as strengths. her institutional analysis and development framework (ostrom, 2005, 2007a) has its roots in the core assumptions of rational choice theory. although she distanced herself from some of the assumptions of this theory, her continued commitment to its assumption that actors in collective action processes are rational utility maximizers creates conceptual problems. this assumption is the basis of her static conceptualization of the conditions of selforganization in common-pool resource systems. she recognized the contributions of complexity theory concepts and adopted some of them in her works. in this paper i will elaborate on the problems the rational choice assumptions create in ostrom’s conceptualizations, as well as the potential for conceptual advancements her recognition of complexity concepts generates. 1. a brief history of self-organization the roots of ostrom’s and complexity theorists’ thinking of self-organization can be found in the philosophical proposition that events and actions do not require external drivers, or hierarchically superior forces, to happen; they can happen for internal reasons, driven by the internal dynamics of systems (göktuğ morçöl, 2012a, p. 93). since the aristotle’s teleological proposition that causes of events in nature are in their final purposes (bogen, 1995, p. 868), philosophers and scientists debated whether events are caused internally or externally. the newtonian scientific understanding, which emerged in the 17th century, was based on the notion that natural events are caused by external factors (göktuğ morçöl, 2002, pp. 13–17). this newtonian understanding was countered by kant’s argument that organismic behaviors were driven by their “inner teleology” in the 18th century (taylor, 2001, pp. 84–93). in the 19th century, in response to auguste comte’s argument that external forces cause social events, phenomenologists argued that in order to understand social events properly, researchers have to understand the inner worlds of human beings (göktuğ morçöl, pp. 163–166). the logical positivists of the early 20th century refined the notion of external causation and made it the core of their vision for scientific explanation. the notion of self-causation, and self-organization, gained some popularity again in the middle of the 20th century, as general systems theory, cybernetic studies, mathematical models of neural networks, and the models of the nervous system adopted and refined it (capra, 1996, p. 22). complexity theorists made two significant contributions to our understanding of selforganization. first, they demonstrated that self-organization is the norm, not an exception, in nature (kauffman, 1993, 1995; nicolis & prigogine, 1989; prigogine, 1996; prigogine & article_14-14.indd 10 03/03/15 3:05 pm g. morçöl / self-organization in collective action 11 stengers, 1984; strogatz, 2003). second, they identified a set of self-organizational mechanisms of the emergence of aggregate patterns or macro structures in animal and human societies (axelrod, 1997; holland, 1995; strogatz, 2003). 2. ostrom’s conceptualization of self-organization and its foundations ostrom (2005) observes that most of the “contemporary policy recommendations” are based on the assumptions that governments are capable of objectively analyzing social problems, producing desired outcomes, and managing social resources and that people are not capable of managing themselves (p. 220). she counters these assumptions and argues that actually not all social problems can be solved centrally by governments and that people are capable of organizing themselves to solve these problems. she and her colleagues demonstrated that individual actors have self-organizational capabilities, but there are also conditions that determine whether or not they will organize themselves. ostrom identified and codified the conditions of self-organization in collective action processes, particularly in the management of common-pool resources. in her conceptualization of the conditions of self-organization, ostrom draws primarily on rational choice theory. although she tried to distance herself from some of the assumptions of this theory, she remained committed to some of its core assumptions. therefore, to understand the value of and the problems in ostrom’s conceptualization, one needs to understand the assumptions and positions of rational choice theorists. rational choice theorist challenge the liberal-democratic theory of government, particularly its proposition that public interest is determined or represented by public authorities, such as governments, who are populated by public servants whose actions are not self-seeking and who are capable of knowing what public interest is and act accordingly to protect and promote it (torfing, peters, pierre, & sorensen, 2012, p. 149). in the liberal-democratic theory, “government” is viewed as a singular actor and a “policy” as an action taken by this actor. a policy is like an object that “impacts” a “target population” to generate an “outcome” (i.e., a policy causes an outcome, as in the newtonian scientific image of external causation). then the role of government’s policy analysts is to trace the trajectory of a policy action and check whether the outcome is compatible with the predetermined goal of the policy (for a classical example of this view of policy analysis, see rossi, lipsey, and freeman, 2004). rational choice theorists argue that governments are not singular actors, that governmental employees are self-seeking actors, just like all other individuals, and that public interest is the aggregated outcome of the actions of all these self-seeking actors, not something protected or promoted by governmental actors only (tullock, 1979, p. 31). in the rational choice view, public policies are collective outcomes of individual actions. then the question is how do individual actions turn into collective outcomes? in answering this question, rational choice theorists make a set of simplifying assumptions. the limitations imposed by these assumptions are reflected in ostrom’s conceptualizations, as i will demonstrate in the next section. the first simplifying assumption is that rational actors act in all economic, political, and other social arenas to maximize their utilities. this assumption has three components: article_14-14.indd 11 03/03/15 3:05 pm 12 g. morçöl / self-organization in collective action the preferences, interests, and actions of individual actors should be the basis of understanding any social and economic phenomenon (methodological individualism); the interests, values, tastes, and preferences of these utility maximizing actors are “fixed” in the sense that they do not have any individual histories or cultural affiliations that would affect the way they make their decisions (the fixed interest assumption); and individuals are selfinterested and utility maximizing rational actors (the utility maximization assumption) (macdonald, 2003). the utility maximization assumption has been criticized by behavioral and cognitive psychologists, who demonstrated that individuals cannot maximize their utilities because of the many cognitive biases and behavioral predispositions they have (friedland & robertson, 1990; hogarth & reder, 1986). simon (1979, 1986) offered an alternative: because humans have “ineradicable” cognitive limitations, their rationality could only be “bounded.” a few scholars also criticized the fixed interest assumption. lichbach (2003, pp. 64–69), a proponent of rational choice, for example, argues that rational choice theorists should acknowledge that because each individual has a history and his/her behaviors are influenced by culture, individual interests and preferences cannot be not fixed. methodological individualism is criticized rarely, and it remains the bedrock principle in rational choice theory and in ostrom’s conceptualization. although her thinking is rooted in the simplifying assumptions of rational choice theory, one can observe that there are tensions, nuances, and contradictions in ostrom’s applications of them. ostrom adopts simon’s notion of bounded rationality (ostrom, 1990, p. 45; 2005, p. 118), but she also thinks that the assumption that individual actors try to fully maximize their utilities can still be useful in understanding economic markets and some other social situations (2005, pp. 99–133). she acknowledges that because of the different mental models individuals use and a variety of feedbacks they receive from the world and their shared culture (p. 105) and the emotional states they are in (pp. 112, 119), their interests, preferences, or behaviors cannot be fixed; they are complex and contextdependent and the culture of a society and institutional structures constitute the context in which individuals make decisions and act (ostrom, 1990, pp. 57–58; 2005, passim). however, in ostrom’s framework these institutional structures function as external inducements for action; they do not shape individual preferences or values (ostrom & parks, 1999: p. 292). because of this distinction between individual preferences and actions on the one hand and institutional structures on the other, ostrom’s conceptualization is static, and thus problematic, as i discuss later in this paper. 3. conditions of self-organization in common-pool resource situations ostrom’s commitment to rational choice assumptions is reflected in her conceptualization of the conditions of self-organization in the management of common-pool resources (cprs). she defines a cpr as a “natural or man-made resource system that is sufficiently large as to make costly (but not impossible) to exclude potential beneficiaries from obtaining benefits from its use” (ostrom, 1990, p. 30). she focused on cpr situations article_14-14.indd 12 03/03/15 3:05 pm g. morçöl / self-organization in collective action 13 in her studies, because, she notes, the processes of self-organization and selfgovernance are easier to observe in them compared to many others (p. 29). ostrom (1990) poses the following as the central question in understanding self organization in the management of cprs: “how [can] a group of principals who are in an interdependent situation…organize and govern themselves to obtain continuing joint benefit when all face temptations to free-ride, shirk, or otherwise act opportunistically”? (p. 29). the underlying assumption in this question is that all the individuals in cpr situations are self-seeking and boundedly rational actors who use a “calculus” to decide whether or not to engage in self-governance (pp. 245–247). ostrom and her colleagues studied empirically the conditions under which these actors act and codified them. ostrom identifies two sets of conditions that determine whether they will participate in self-organizational processes: those conditions that are conducive for individuals to begin governing themselves—what she calls the “attributes of resources and appropriators conducive to and increased likelihood that self-governing associations will form” (2005, pp. 244–245)—and those that will help them maintain self-governing institutions—what she calls the “design principles for long-enduring cpr institutions” (1990, p. 90) or the design principles for “long-enduring institutions for governing sustainable resources” (2005, p. 259). according to ostrom (2005), the attributes of resources that are conducive to appropriators’ self-organization are that there is a reasonable chance that the resources available to them can be improved, that there are sufficiently large resources and a good chance to improve them, that there are reliable and valid indicators of the conditions of resources available to appropriators at a relatively low cost, and that the resource system is sufficiently small so that appropriators can develop accurate knowledge of the boundaries of the system (pp. 247–248). the attributes of appropriators that would enable them to participate in self-governance in self-organization are that the cpr system is salient enough for their livelihood or their achievement of important social or religious values so that they will be motivated to self-organize, that they have sufficiently common understanding of the cpr system and how their actions will affect the system, that appropriators have “sufficiently low discount rate” in relation to their future benefits and costs for participating in a self-governing system, that they trust each other for keeping promises and reciprocating their actions, appropriators have enough autonomy to carry out their actions vis-à-vis “external authorities” (governmental authorities), and they should have organizational experience and leadership skills gained from earlier experiences (pp. 248–251). once a self-governing cpr system is established, it should be maintained. ostrom formulated the conditions that would help self-governing cpr institutions to endure for a long period of time (1990, 90–102), which she later broadened to apply to all types of “socio-ecological systems” (2005, pp. 258–288). the compliance by participants of self-governing system should be enforced and monitored and trust and reciprocity levels among the participants are important factors that will determine the level of compliance with the rules. ostrom (2005) observes that those systems in which participants select their own monitors of compliance—rather than external forces, such as governments, do article_14-14.indd 13 03/03/15 3:05 pm 14 g. morçöl / self-organization in collective action that—are more likely to survive for long periods of time (p. 265), as well as those in which “graduated sanctions” applied to rule breakers and those that have mechanisms of conflict resolution (pp. 265–268). 4. ostrom and complexity theory ostrom’s assumption that actors are capable of organizing themselves in collective action processes resonates well with complexity theorists’ observation that selforganization is natural. the affinity between her conceptualizations and those of complexity theorists go beyond this general resonance. particularly in her later writings and her nobel prize acceptance speech ostrom acknowledged the contributions by complexity theorists and adopted some of their concepts (e.g., ostrom, 2005, pp. 242–243, 256; 2007b; 2009). one might speculate about whether, when, and how complexity theory influenced her thinking. more important for the purpose of this paper is to acknowledge that with her conceptualizations and empirical findings, ostrom demonstrated, directly and indirectly, that the concepts of complexity theory can help advance our understanding of governance/ collective action processes. it is also important to highlight the parallels between her conceptualizations and those of complexity theorists, assess hers critically from the perspective of complexity theory, and draw lessons from them. one of ostrom’s positions that bring her close to complexity thinking is her rejection of simplified dichotomies like governmental intervention versus free markets (2005, p. 256). she rejects the propositions that either free markets or governments can generate the best solutions to all social and economic problems. an important contribution she makes is to dispel the myths that free markets equal to self-governance and that selfgovernance (or self-organization) is necessarily an egalitarian process. she points out that many studies demonstrated that local self-organization may be dominated by local elites (ostrom, 2005, p. 220), for example. ostrom also notes that although in many cases centralized governmental systems were not found to be as effective as locally self-organized common-pool resource governance systems, there is no common rule that would suggest that one or the other is always better (p. 222). she argues that there are different configurations of relationships among governmental, nonprofit, and private actors that will generate different solutions to social problems and the best configuration of the roles of governmental and other actors depends on the context (1990, pp. 8–15, 2009). she also observes that collective action processes take place in nested polycentric systems. drawing on vincent ostrom’s (1999a, 1999b) earlier conceptualizations of polycentric systems, ostrom (2005) observes that when common-pool resources are larger than the boundaries of a local community, many organizational actors will be involved and this will create “multiple layers of nested enterprises” (p. 270). in these multi-layered systems, not only there are multiple organizations and actors involved, such as generalpurpose and special-purpose governments and local community organizations, but also the authorities and functions of these organizations are layered: some have more general authorities and functions than others. these polycentric governance systems are complex article_14-14.indd 14 03/03/15 3:05 pm g. morçöl / self-organization in collective action 15 adaptive systems, and as such, they should adapt to their environments by reconfiguring their components in order to maintain their desired system characteristics for a long period of time (p. 258). she also notes that there are no simple, definitive, or universal rules that govern these systems, because we will always have incomplete knowledge of them and their relations with larger systems (p. 284). although she makes some generalizations about the “design principles” of self-organizing systems, as summarized in the previous section, she notes that there are many areas in which no conclusive design principles can be devised. therefore, a better approach is to develop a configurational understanding of these systems, to identify specific configurations of the variables for particular conditions, rather than trying to find out the optimal conditions for self-organization (p. 254). because all our analyses and knowledge of systems are partial, we should design our institutions adaptively (p. 254). having noted the parallels between ostrom’s thinking and complexity theory, i also should point out a major problematic area in her conceptualizations. as i mentioned earlier in this paper, ostrom remained committed to some of the basic assumptions of rational choice theory. because of this commitment, the conditions of self-organization she identified reflect a static conceptualization. what is particularly problematic in this assumption is its methodological individualist element. methodological individualists assume that individual behavior is a distinct and the only legitimate unit of analysis. although they may recognize that social structures exist, as ostrom does, in their conceptualization either these structures have independent existence from the behaviors of individuals, or they fail to explain how the social and individual processes may be related. in ostrom’s institutional analysis and development framework, individual actors act in “action arenas,” which social structures are created by external forces (e.g., governments) (2005, pp. 14–15). there are also “exogenous variables,” such as biophysical and material conditions, attributes of community (culture), and various rules set at different tiers of social existence. these rules are set for “operational,” “collective-choice,” “constitutional,” and “meta-constitutional situations” (pp. 58–62). ostrom uses the term “institutions” for these rule sets and action areas. it is important to stress that all these rule sets and action areas exist independently of individual actors in her conceptualization. ostrom and parks (1999) specifically mention that these institutions are external to individual decision makers; they function as external inducements for action and they do not shape individual preferences or values (p. 292). methodological individualism is evident also in ostrom’s descriptions of the conditions of self-organization in cpr systems. as i summarized in the previous section, in her conceptualization individual actors are utility maximizing individuals who make costbenefit calculations to such an extent that they calculate “discount rates” to make decisions whether or not to participate in a self-governing cpr system. she defines even “trust” in terms of the “expected cost” of one individual trusting another one. although she agrees with simon’s proposition that individuals do not have the cognitive capabilities to be fully article_14-14.indd 15 03/03/15 3:05 pm 16 g. morçöl / self-organization in collective action rational, they still are “boundedly rational” (i.e., imperfectly calculating actors), according to ostrom. then, where do the interests, tastes, preferences, and values of these calculating individuals come from? institutions do not shape them (ostrom & parks, 1999). is there anything else that shapes them? do the personal histories of individual actors shape them, for example? rational choice theorists have been criticized for assuming that the interests, values, tastes, and preferences of actors are “fixed” in the sense that they do not have any individual histories or cultural affiliations that would affect the way they make their decisions (macdonald, 2003), as i mentioned earlier in this paper. ostrom’s writings are not clear on whether she thinks that they are fixed. as i discussed in the previous section, she describes the “attributes of appropriators” that are conducive to making them participate in self-governance and by doing so she acknowledges that there are variations among actors. this acknowledgement might imply that their interests, values, etc. are shaped by their individual histories. however, she does not discuss specifically where the variations in the “attributes of appropriators” come from. because of the implicit, or explicit, assumption that the interests and values of individuals are fixed, ostrom’s framework is static. she ignores the dynamics of the relationships between individuals and social structures (e.g., institutions). the nature and dynamics of micro–macro relationships is a core problem for complexity researchers who study social systems in general, and policy/governance systems in particular (göktuğ morçöl, 2012b). in complexity theory, macro structures (the properties of a system) emerge from the relationships at the micro level and in turn they can help shape the relationships at the micro level. in other words, the micro and macro levels are interdependent (e.g., culture and individuals’ behaviors and interests are interdependent). this is why a researcher has to understand not only the properties of the components (e.g., individual actors), but also those of the system (e.g., institutions) and the dynamic relationships between the two (göktuğ morçöl, 2012a, chapter 3). cilliers’s (1998) conceptualization of self-organization is a good example of the dynamic understanding complexity theorists put forth. according to cilliers, “the capacity of self-organization is a property of complex systems which enables them to develop or change internal structure spontaneously and adaptively in order to cope with, or manipulate, their environment” (p. 90). in other words, a self-organizing system not only reacts to it environment, but also transforms itself through interactions with its environment (p. 108). “self-organization is a self-transforming process; the system acts upon itself ” (p. 108). the degree of openness of a system to its environment and the degrees to which the environment can influence the system and the system can influence the environment are open questions for complexity researchers (göktuğ morçöl, 2012a, pp. 115–119). another set of conceptual and empirical problems complexity researchers dealt with were about the boundaries of self-organizing complex systems and how the “self” should be defined in them. ostrom (2005) states that how the boundaries of cpr systems is an important problem (p. 248), but she does not provide an answer to it. complexity researchers aim to understand how boundaries of self-organizing complex systems are formed. article_14-14.indd 16 03/03/15 3:05 pm g. morçöl / self-organization in collective action 17 rhodes, murphy, muir, and murray’s (2011) conceptualizations of the dynamics of setting and re-setting the geo-political boundaries in the urban regeneration and healthcare systems in ireland and northern ireland are good examples (pp. 134–147). applying two complexity theory concepts, path dependency and bifurcation, they demonstrate that the urban boundaries in belfast and dublin were set and changed over time, both geographically and politically, based on the histories of the neighborhoods in these cities. if systems are composed of other systems and all systems are related to other systems, then how can we define the “self” in self-organizational processes? in moreno and ruiz-mirazo’s (2007) conceptualization, a system’s components define themselves as a system from which emerges the notion of self. in describing the process of defining self, they use the term “self-encapsulation”: sharply differentiating the organization of the system—the set of relations that constitute it as a distinct unity—from the environment. by establishing a distinct set of relationships among themselves, a group of elements define a system, a self, according to moreno and ruiz-mirazo. the best example of this is the process of developing the membrane of a living cell: through self-encapsulation, a group of molecules become a cell. in social systems self is defined through social construction processes, according to gerrits (2008) and gerrits, marks, and van buuren (2009). complex social systems are defined jointly by their participants and their observers (researchers) these authors point out, and they illustrate this point with examples from the netherland and germany. this social constructionist approach is insightful but not adequate, because it may imply that social systems are merely the products of the conjectures of their participants and/or observers and that a social system is merely an imaginary entity that exists in some individuals’ minds; also it does not adequately consider the dynamics of defining selves and systems. complexity researchers need to develop better conceptualizations, in my view. giddens’ (1984) view that systems are not imaginary entities, but they are “situated activities of human agents, reproduced across time and space” (p. 25) is a better conceptualization, as i argued elsewhere (göktuğ morçöl, 2012a). an important area of conceptualization and research in complexity studies is the nature of self-organizing agents. whether they self-encapsulate or socially construct their identities to form a system, agents should have the capacity to do so. teisman, van buuren, and gerrits (2009) point out that human agents have self-reflexive capacities and that “self-organization is the reflexive capacity of actors and (sub)systems who are able to receive, encode, transform and store information and use this to consider actions” (p. 9). in other words, self-organization requires some information processing capacity by reflexive actors. whether agents must indeed have self-reflexive capacities for them to be able to selforganize is an unsettled issue in complexity theory. the larger question is whether or not intelligence or knowledge is a precondition for self-organization. this question is closely related to the issue of the rationality, or bounded rationality, of actors. complexity theorists identified two kinds of agents: “cognitive agents” and “reactive agents.” cognitive agents are intentional and deliberative; they have “internal models,” or “schema,” which helps article_14-14.indd 17 03/03/15 3:05 pm 18 g. morçöl / self-organization in collective action them anticipate future events and thus guide their actions (holland, 1995, pp. 31–34). cognitive agents are similar to boundedly rational actors, who can process information and make predictions into future, which help them guide their actions. reactive agents do not have any internal models to guide their actions; they merely react to their immediate environmental conditions. for complexity researchers the question is does self-organization require cognitive agents, or could reactive agents self-organize as well? several researchers used agent-based simulations (abs) to answer these questions. in some of these abs studies, researchers demonstrated the self-organizational capabilities of cognitive agents (e.g., ahmad, 2008; castelfranchi, 1998; naveh & sun, 2006) or boundedly rational agents (e.g., axelrod, 1997; bednar & page, 2007; cohen, riolo & axelrod, 2001; epstein, 2006; holland, 1995; miller & page, 2007). as sawyer (2005) notes, however, other researchers demonstrated that self-organizational processes can occur even when there are only reactive agents and objective structures can emerge from the interactions of such agents (p. 161). epstein and axtell (1996) demonstrated, for example, that identifiable social structures can emerge from the interactions of artificial reactive agents that have limited knowledge of their environments and limited capabilities to react. the question of whether, or to what extent, cognitive capabilities (or rationality) are needed for self-organization is still an open question for complexity theorists and researchers. although she does not discuss this explicitly, ostrom’s conceptualization seems to imply that self-organizational processes yield more organization and orderliness in collective action processes. the conceptual problems she poses are whether and how selfregulating orderly regimes emerge through self-organization in cpr situations. complexity theorists show that the direction of self-organization may be toward more (higher levels of) organization and orderliness or toward disorderliness. according to anderson (1999) complex systems organize themselves toward order (p. 218). prigogine, on the other hand, shows that self-organizational processes may go in either direction. in his view, systems spontaneously evolve toward “far from equilibrium” conditions under which there are two possibilities: either systemic properties break down (disorderliness), or new systemic properties emerge (orderliness) (prigogine & stengers, 1984; nicolis & prigogine, 1989; prigogine, 1996). prigogine’s conceptualization of the two directions in self-organizational processes has implications for understanding governance/collective action processes. it can help researchers of these processes to conceptualize the emergence of failures and undesirable outcomes in such actions (disorderliness), as well as the emergence of successes and desirable outcomes (orderliness). 5. conclusions there is a major lesson in ostrom’s development of her institutional analysis and development framework for complexity researchers: in order for complexity theory to become a viable alternative in policy/governance/collective action studies, complexity researchers would need to take ostrom’s example of meticulous theory building and article_14-14.indd 18 03/03/15 3:05 pm g. morçöl / self-organization in collective action 19 empirical verification as an example. the institutional analysis and development framework has been recognized as the most developed of the theoretical frameworks explaining policy processes and the most widely applied framework in the united states and europe (sabatier, 2007, p. 9). ostrom’s contributions, which are encapsulated in this framework, were recognized with a nobel prize in economics in 2009 (ostrom, 2009). ostrom’s publications are among the the most frequently cited works. (a google scholar search in may of 2014, when this paper was written, yielded 77776 citations of her works, compared to albert einstein’s 94081 citations.) all in all it is clear that ostrom’s framework made significant impacts in the scholarly literature, which would be difficult to emulate, but it can be an example for the future works of complexity researchers. complexity researchers should take the theoretical articulation of their concepts seriously. i proposed elsewhere that there already are two theoretical frameworks that are emerging within complexity studies: the micro-macro framework and the socio-ecological systems framework (göktuğ morçöl, 2014). the concepts of these frameworks need to be refined and empirically verified in future studies. in doing so, complexity researchers could consider adopting and adapting ostrom’s conditions of self-organization. however, the researchers should keep in mind the limitations posed by the roots of ostrom’s framework in rational choice theory. because of her loyalty to methodological individualism, her framework is static. her descriptions of the layers of this framework and the relationships among them are conceptually and mathematically sophisticated, but they lack a conceptualization of the dynamics of these relationships. this deficiency can be seen clearly in her descriptions of the conditions of self-organization, particularly the attributes of resources and actors that are required for self-organization. in these descriptions, actors and social structures are conceptualized as distinct entities. she does not consider the effects of the interactions between actors and socials structures and possible transforming effects of one on the other. as i pointed out in the previous section, complexity researchers have already conceptualized, and investigated to some degree, various aspects of the dynamics of micro– macro interactions, but most of these conceptualizations lack the specificity of ostrom’s conceptualization of the conditions of self-organization in cpr situations. ostrom’s conditions could be reconsidered and re-conceptualized from a complexity theory perspective. for example, it could be asked whether utility maximization, as ostrom suggests, or social construction is the primary force that determines the decisions of the appropriators of common-pool resources to self-organize. if the existence of “cognitive agents” is not a necessary condition for self-organizational processes to occur (reactive agents can self-organize as well), to what extent do appropriators need certain kinds of information (information about feasibility of improvement, reliability and validity of indicators, predictability of future outcomes, etc.) to self-organize? how do path dependencies affect appropriators’ perceptions of the boundaries of common-pool resource systems? self-organization is an important concept that can help us rectify the newtonian external cause–effect way of thinking, which dominates contemporary policy analysis. the philosophical debates on whether events are caused externally or internally have a article_14-14.indd 19 03/03/15 3:05 pm 20 g. morçöl / self-organization in collective action long history, as i noted in this paper. these debates have affected the thinking in studies on policy, governance, and collective action processes. in general the newtonian external cause–effect thinking permeated most of the conceptualizations and empirical studies in these areas. ostrom and complexity researchers not only demonstrated that policy/governance/collective action processes can be self-organizational, but they also identified the conditions and mechanisms of self-organization in them. elizabeth ostrom’s and her colleagues’ contributions in this area are substantial, but they have shortcomings as well. with their systemic and dynamic understanding of natural and social phenomena, complexity theorists can build a better framework of understanding how complex governance networks work. references ahmad, r. (2008). a formal framework for engineering intelligent agents-based systems (doctoral dissertation). university at carbondale, carbondale, il. anderson, p. (1999). complexity theory and organization science. organization science, 10, 216–232. axelrod, r. (1997). the complexity of cooperation: agent-based models of competition and collaboration. princeton, nj: princeton university press. bednar, j., & page, s. (2007). the emergence of cultural behavior within multiple games. rationality and society, 19(1), 65–97. bogen, j. (1995). teleological explanation. in t. honderich (ed.), the oxford companion to philosophy (pp. 868–867). oxford, england: oxford university press. capra, f. (1996). the web of life: a new scientific understanding of living systems. new york, ny: anchor books. castelfranchi, c. (1998). simulating with cognitive agents: the importance of cognitive emergence. in j. s.  sichman, r. conte, & n. gilbert (eds.), multi-agent systems and agent-based simulation (proceedings of the first international workshop, mabs ’98, paris, france, july 4–6) (pp. 26–44). berlin, germany: springer. cilliers, p. (1998). complexity and postmodernism: understanding complex systems. london, england: routledge. cohen, m. d., riolo, r. l., & axelrod, r. (2001). the role of social structure in the maintenance of cooperative regimes. rationality & society, 13, 5–32. epstein, j. m. (2006). generative social science: studies in agent-based computational modeling. princeton, nj: princeton university press. epstein, j. m., & axtell, r. (1996). growing artificial societies: social science from the bottom up. washington, dc: brookings institution press. friedland, r., & robertson, a. f. (eds.). (1990). beyond the marketplace: rethinking economy and society. new york, ny: aldine de gruyter. gerrits, l. (2008). the gentle art of coevolution: a complexity theory perspective on decision making over estuaries in germany, belgium and the netherlands (doctoral dissertation). erasmus university, rotterdam, netherlands. retrieved from http://publishing.eur.nl/ir/repub/asset/11152 gerrits, l., marks, p., & van buuren, m. w. (2009). coevolution: a constant in nonlinearity. in g. r. teisman, m. w. van buuren, & l. gerritts (eds.), managing complex governance systems: dynamics, selforganization and coevolution in public investments (pp. 134–153). london, england: routledge. giddens, a. (1984). the constitution of society: outline of the theory of structuration. berkeley, ca: university of california press. hogarth, r. m., & reder, m. w. (eds.). (1986). rational choice: the contrast between economics and psychology. chicago, il: the university of chicago press. article_14-14.indd 20 03/03/15 3:05 pm g. morçöl / self-organization in collective action 21 holland, j. h. (1995). hidden order: how adaptation builds complexity. new york, ny: basic books. kauffman, s. (1993). the origins of order: self-organization and selection in evolution. new york, ny: oxford university press. kauffman, s. (1995). at home in the universe: the search for laws of self-organization and complexity. new york, ny: oxford university press. lichbach, m. a. (2003). is rational choice theory all of social science? ann arbor, mi: university of michigan press. macdonald, p. k. (2003). useful fiction or miracle maker: the competing epistemological foundations of rational choice theory. american political science review, 97(4), 551–565. miller, j. h., & page, s. e. (2007). complex adaptive systems: an introduction to computational models of social life. princeton, nj: princeton university press. morçöl, g. (2002). a new mind for policy analysis: toward a post-newtonian and postpositivist epistemology and methodology. westpost, ct: praeger. morçöl, g. (2012a). a complexity theory for public policy. new york, ny: routledge. morçöl, g. (2012b). the micro–macro problem as the core problem of complexity theory of policy processes. in l. gerrits, p. marks, & f. boons (eds.), compact 1: public administration in complexity (pp. 307–322). litchfield park, az: emergent publications. morçöl, g. (2014). complex governance networks: an assessment of the advances and prospects. complexity, governance & networks, 1(1), 1–12. moreno, a., & ruiz-mirazo, k. (2007). the maintenance and open-ended growth of complexity in nature: information as a decoupling mechanism in the origins of life. in f. capra, a. juarrero, p. sotolongo, & j. van uden (eds.), reframing complexity: perspectives from the north and south (pp. 60–76). mansfield, ma: isce publications. naveh, i., & sun, r. (2006). a cognitively based simulation of academic science. computational and mathematical organization theory, 12(4), 313–337. nicolis, g., & prigogine, i. (1989). exploring complexity: an introduction. new york, ny: w. h. freeman. ostrom, e. (1990). governing the commons: the evolution of institutions for collective action. cambridge, england: cambridge university press. ostrom, e. (2005). understanding institutional diversity. princeton, nj: princeton university press. ostrom, e. (2007a). institutional rational choice: an assessment of the institutional analysis and development framework. in p. a. sabatier (ed.), theories of the policy process (pp. 21–64). cambridge, ma: westview. ostrom, e. (2007b). biography of robert axelrod. political science & politics. 40(1), 171–174. ostrom, e. (2009). beyond markets and states: polycentric governance of complex economic systems (nobel prize lecture). retrieved from http://www.nobelprize.org/nobel_prizes/economics/laureates/2009/ostromlecture.html ostrom, e., & parks, r. b. (1999). neither gargantuan nor the land of lilliputs: conjectures on mixed systems of metropolitan organization. in m. d. mcginnis (ed.), polycentricity and local public economies: readings from the workshop in political theory and policy analysis (pp. 284–305). ann arbor, mi: the university of michigan press. ostrom, e., gardner r., & walker, j. (1994). rules, games, & common pool resources. ann arbor, mi: the university of michigan press. ostrom, v. (1999a). polycentricity (part 1). in m. d. mcginnis (ed.), polycentricity and local public economies: readings from the workshop in political theory and policy analysis (pp. 52–74). ann arbor, mi: the university of michigan press. ostrom, v. (1999b). polycentricity (part 2). in m. d. mcginnis (ed.), polycentricity and local public economies: readings from the workshop in political theory and policy analysis (pp. 75–92). ann arbor, mi: the university of michigan press. prigogine, i. (1996). the end of certainty: time, chaos, and the new laws of nature. new york, ny: the free press. prigogine, i., & stengers, i. (1984). order out of chaos: man’s new dialogue with nature. new york, ny: bantam books. article_14-14.indd 21 03/03/15 3:05 pm 22 g. morçöl / self-organization in collective action rhodes, m. l., murphy, j., muir, j., & murray, j. a. (2011). public management and complexity theory: richer decision-making in public services. london, england: routledge. rossi, p. h., lipsey, m. w., & freeman, h. e. ( 2004). evaluation: a systematic approach (7th ed.). thousand oaks, ca: sage. sabatier, p. a. (2007). the need for better theories. in p. sabatier (ed.), theories of the policy process (pp. 3–17). cambridge, ma: westview press. sawyer, r. k. (2005). social emergence: societies as complex systems. cambridge, england: cambridge university press. simon, h. a. (1979). rational decision making and business organizations. the american economic review, 69(4), 493–513. simon, h. a. (1986). rationality in psychology and economics. in r. m. hogarth & m. w. reder (eds.), rational choice: the contrast between economics and psychology (pp. 25–40). chicago, il: the university of chicago press. strogatz, s. (2003). synch: the emerging science of spontaneous order. new york, ny: hyperion. taylor, m. c. (2001). the moment of complexity: emerging network culture. chicago, il: the university of chicago press. teisman, g. r., van buuren, m. w., & gerrits, l. (eds.). (2009). managing complex governance systems: dynamics, self-organization and coevolution in public investments. london, england: routledge. torfing, j., peters, b. g., pierre, j., & sorensen, e. (2012). interactive governance: advancing the paradigm. oxford, england: oxford university press. tullock, g. (1979). public choice in practice. in c. s. russell (ed.), collective decision making: applications from public choice theory (pp. 27–45). baltimore, md: the johns hopkins university press. article_14-14.indd 22 03/03/15 3:05 pm complexity, governance & networks special issue: complexity, innovation and policy (2017) 1 message from the editors doi: http://dx.doi.org/10.20377/cgn-37 1 university of bamberg press message from the editors this special issue on the coevolution of innovation and public policy is the first issue of complexity, governance & networks to be published by our new publisher, the university of bamberg press (ubp). ubp, an academic publisher from germany, receives financial support from the ottofriedrich university of bamberg and the german science foundation to publish our journal through an open source, open access framework. because of this support, ubp will be able to publish it open access under the creative commons license and will not charge authors for the articles published in it. in other words: all the articles published in our journal will be freely available to anyone with an internet connection, nothing will be hidden behind paywalls. this way, the articles published in the journal will have greater visibility. complexity, governance & networks will be indexed in all relevant databases (e.g. scopus and web of science) and its contents will be searchable through them. open access publishing will also give authors greater control over their publications: they can distribute them freely without prior consent. the transition towards an open access journal will not diminish our focus on publishing high-quality papers. all manuscripts submitted to complexity, governance & networks will be subjected to a thorough double-blind review process to ensure that only the best publications will appear in the journal. the back catalogue of our journal has been taken over by ubp and all the articles published in previous issues will be available to download, for free, at the website of complexity, governance & networks: http://ubp.uni-bamberg.de/ojs/index.php/cgn ubp uses a different manuscript review and editing platform than the one our journal used before. it may take a few additional minutes for the authors and reviewers who are familiar with the old platform (manuscript tracker) to adjust to the layout out of the new platform (open journal systems: see http://ubp.uni-bamberg.de/ojs/index.php/cgn/about/submissions for the submission system, and https://pkp.sfu.ca/ for more information on ojs), but we believe that this will not create major difficulties in submitting manuscripts or reviews. in any case, authors and reviewers will receive clear instructions on how to proceed. many people have worked hard behind the scenes to bring complexity, governance & networks to this point. as the editors-in-chief of the journal, we want to thank everyone who has contributed to the journal: the editors of the special issues, the authors as well as the reviewers. we also want to express our gratitude to mrs. barbara ziegler, mrs. ulrike wiesner, mrs. silke raffel and mr. martin wirtz for their continuous efforts to make everything work, and the ottofriedrich university of bamberg for allocating the resources necessary to support the publication of complexity, governance & networks. the editors in chief, prof. dr. lasse gerrits prof. dr. göktuğ morçöl http://ubp.uni-bamberg.de/ojs/index.php/cgn http://ubp.uni-bamberg.de/ojs/index.php/cgn/about/submissions https://pkp.sfu.ca/ balancing the quantitative and qualitative aspects of social network analysis to study complex social systems complexity, governance & networks-vol 02, issue 01 (2015) 5–22 5 doi: 10.7564/15-cgn23 balancing the quantitative and qualitative aspects of social network analysis to study complex social systems danny schippera* and wouter spekkinkb adepartment of public administration, erasmus university rotterdam, the netherlands e-mail: schipper@fsw.eur.nl bdepartment of engineering systems and services, delft university of technology, the netherlands e-mail: w.a.h.spekkink@tudelft.nl social network analysis (sna) can be used to investigate complex social systems. sna is typically applied as a quantitative method, which has important limitations. first, quantitative methods are capable of capturing the form of relationships (e.g. strength and frequency), but they are less suitable for capturing the content of relationships (e.g. interests and motivations). second, while complex social systems are highly dynamic, the representations that sna creates of such systems are often static. these limitations can be overcome by balancing a quantitative approach to sna with a qualitative approach. in the article two different approaches that seek this balance are demonstrated. the illustrations show that in this combination quantitative sna is most useful for revealing system-level patterns, but that a deeper understanding of the mechanisms that produce these patterns is more easily achieved through the interpretation of qualitative data. keywords: social network analysis, complex systems, dynamic network analysis, qualitative analysis 1. introduction complex systems emerge through the interactions among their constituent elements (gerrits, 2012). the concept of emergence points to the fact that these systems cannot be reduced to the properties of their constituent elements (goldstein, 1999). complex systems are also dynamic, meaning that their structures and elements change over time (room, 2011). the investigation of complex social systems therefore requires methods that take into account their emergent and dynamic nature. traditional reductionist methods of scientific inquiry are unsuitable in this regard, as they tend to isolate properties of systems and analyze them separately, thereby losing sight of the system as a whole (clancy, effken, & pesut, 2008). social network analysis (sna) is capable of simultaneously taking into account higher-order system structures and their constituent elements (knoke & yang, 2008). what sets sna apart from other approaches in the social sciences * corresponding author. 62084-15-23.indd 5 18/08/15 10:30 am 6 d. schipper and w. spekkink / balancing the quantitative and qualitative aspects is its focus on relationships among actors instead of attributes of actors (cf. abbott, 1988). these relationships constrain individual behavior, while at the same time they are continuously remolded by that behavior (spreitzer & yamasaki, 2004). over the last decades the use of sna in the social sciences has grown exponentially (borgatti, mehra, brass, & labianca, 2009). this growing interest has resulted in major advances in the methods of sna. advances have been made primarily in quantitative methods such as the mathematical methods of mapping networks and measuring their properties (coviello, 2005; crossley, 2010; edwards, 2010; heath, fuller, & johnston, 2009). quantitative methods of sna can be used to study complex systems; they offer an efficient way to describe and analyze complex social structures (edwards, 2010). however, quantitative methods also abstract a great deal from the actual complexity of social systems in two ways. first, although they are capable of capturing the form of relationships (intensity, frequency, or strength), they are largely blind to their content (interests, purposes, drives, and motivations) (crossley, 2010; edwards, 2010; knoke & yang, 2008). second, with some exceptions, these quantitative methods are used to study static structures, and they largely neglect the changes that these structures undergo (knoke & yang, 2008; scott, 2013; wolbers, groenewegen, mollee, & bím, 2013). this hampers understanding the mechanisms responsible for the emergence and development of complex social structures. in this article we demonstrate how these limitations can be overcome by complementing quantitative methods of sna with methods that are primarily qualitative and focus explicitly on network dynamics. we introduce two methodological approaches and demonstrate their value in the investigations of complex social systems with two case illustrations. the two approaches help to establish a stronger link between patterns at the system level and micro-level behavior, while at the same time explicitly taking into account the dynamic nature of social systems. our approaches include quantitative methods through which broader system-level patterns and the changes in them are uncovered. these system-level patterns act as a starting point of a qualitative analysis to study how these patterns are linked to micro-level behavior, allowing us to uncover the mechanisms that are at the basis of the evolution of the social system. the central research question of the article is: what can qualitative and dynamic approaches to social network analysis contribute to our understanding of the emergent and dynamic properties of complex social systems? we start our article with a discussion of sna as a method to study complex systems, focusing primarily on the value of qualitative and dynamic methods of sna. we then offer a detailed discussion of two methods to overcome the limitations of quantitative sna and we demonstrate their workings through case illustrations. the illustrations are followed by the conclusions in the final section. 2. balancing quantitative and qualitative sna the primary conceptual starting point of this article is that complex social systems are emergent and dynamic. an understanding of the emergent nature of complex systems requires an investigation of micro-processes and their relations with macro structures 62084-15-23.indd 6 18/08/15 10:30 am d. schipper and w. spekkink / balancing the quantitative and qualitative aspects 7 (morçöl, 2014). these processes describe how the behaviors of individuals generate higher-order structures on the one hand, and how these higher-order structures affect individual behavior on the other hand (morçöl, 2012b). as knoke and yang (2008) point out, sna offers conceptual and methodological tools to explicitly link micro-level behavior to macro-level structures; both micro-level entities (i.e., actors and their relationships) and the macro-level structures that they form (i.e., the networks of relationships among actors) are included in the representations of social networks. although sna factors in both the structure of social networks and the nature of interactions between actors (jack, 2010), explanations for outcomes are sought almost exclusively in macro-level patterns (kilduff & tsai, 2003; knoke & yang, 2008). measuring and representing these structural patterns has become the central objective in sna, thereby neglecting the role of active individual actors in shaping the structures (kilduff & tsai, 2003). from a complexity perspective this is problematic, as micro-level behavior is understood to be at the basis of the emergence and development of social structures. neglecting micro-level behavior will lead to oversimplified descriptions (byrne & callaghan, 2013) and hampers our understanding of complex social systems. the emphasis on structural patterns in sna applications has also led to the generation of static pictures of networks (knoke & yang, 2008; scott, 2013; wolbers et al., 2013). however, networks are not static; they change over time as actors intentionally and unintentionally change relational structures. therefore, it is important to take into account the temporal dimensions of networks (doreian & stokman, 1997a, 1997b). however, collecting longitudinal network data can be very time consuming. in addition, commonly used methods of data collection, such as surveys and interviews, rely on the capacity of respondents to recollect interactions that have happened some time ago and could therefore be imprecise (morçöl, 2012a). nevertheless, there have been considerable developments in the collection of longitudinal data and the methods to analyze these data. several studies have incorporated a temporal dimension in sna. for instance, powell, koput, and smith-doerr (1996) and powell, white, koput, smith, and owen-smith (2005) study the evolution of interorganizational collaboration in biotechnology and the life sciences by mapping contractual agreements between companies in these sectors. ahuja (2000) performed a similar type of study for the chemical sector. zaheer and soda (2009) perform a longitudinal investigation of co-membership networks in the italian tv production industry to study the origin of structural holes. abbasi and kapucu (2012) have looked at the dynamic changes of interorganizational response networks during hurricane charley. wolbers et al. (2013) provide a toolset to measure network dynamics. this toolset combines time slices, two-mode analysis and information pathways to specifically follow the flows of information during an emergency response. other important work has been done by the center for computational analysis of social and organizational systems (casos) at carnegie mellon university. they have developed dynamic network analysis (dna), which offers a set of techniques and tools to investigate complex and dynamic sociotechnical systems (carley, diesner, reminga, & tsvetovat, 2007). one of the tools 62084-15-23.indd 7 18/08/15 10:30 am 8 d. schipper and w. spekkink / balancing the quantitative and qualitative aspects developed by casos, ora, is a dynamic meta-network assessment and analysis tool which allows researchers to visualize and analyze networks over time in a user-friendly environment. although these studies and tools help us to grasp the temporal dimension of networks, they are still mainly focused on macro-level patterns. to analyze the evolution of networks other researchers have developed a relatively new set of methods, which can be grouped under the heading of dynamic social network analysis (dsna). this approach usually entails the development of models that are tested through statistical analysis or simulations. hypothesized rules of action by actors are then used in a simulation to generate patterns of network evolution, which are compared to actually observed patterns (scott, 2013). a closely related approach is based on the use of exponentional random graph (erg) models, in which an empirically observed network is regarded as one possible outcome of some unknown social process (frank & strauss 1986; robins, pattison, kalish, & lusher, 2007a; robins, snijders, wang, handcock, & pattison, 2007b). a stochastic model is developed of the social process that is assumed to have generated the network, and the parameters of the model are estimated. typically, the aim of this approach is to assess whether certain structural characteristics occur more commonly in the observed network than would be expected by chance. a drawback of these approaches is that only a limited set of fairly general mechanisms (e.g. preferential attachment, homophily, transitivity) can be considered in the analysis. the question is whether such general mechanisms can truly account for the behavior that we encounter in complex social systems (morçöl, 2012a). although a more qualitatively oriented approach does not allow for the same statistical rigor, it also holds some advantages. as crossley (2010) points out, many different mechanisms are at play in complex systems and the interactions between mechanisms make it difficult to isolate specific mechanisms responsible for the observed outcomes. because we cannot always know in advance which mechanisms account for the emergence and development of complex social systems, it may be necessary to rely on qualitative observations to understand how mechanisms manifest and operate in a specific context (byrne & callaghan, 2013; crossley, 2010; teisman & gerrits, 2014). this can even lead to the discovery of new mechanisms (crossley, 2010). also, the effects that these general system-level patterns have on micro-level behaviors can be different across cases. for example, two actors may have structurally equivalent positions in two different networks, but respond to the conditions raised by this position differently based on their specific beliefs, desires, and the opportunities. thus, as a result of contextual differences general patterns are often accompanied by unique circumstances that invite different explanations across cases (buijs, eshuis, & byrne, 2009; teisman & gerrits, 2014). by combining the quantitative aspects of sna with qualitative observation and analysis it is possible to introduce contextual details that are otherwise lost in the abstractions that sna makes (crossley, 2010; morçöl, 2012a; morçöl, 2014). in order to use sna to its full potential in the investigation of complex social systems, a better balance needs to be sought between the quantitative and the qualitative aspects of sna both in terms of data collection and analysis (crossley, 2010; edwards, 2010; 62084-15-23.indd 8 18/08/15 10:30 am d. schipper and w. spekkink / balancing the quantitative and qualitative aspects 9 kapucu, hu, & khosa 2014). in our view the two primary ways in which qualitative methods and techniques can complement a quantitative approach are: 1. by offering a more detailed account of the micro-level behavior from which systemlevel patterns emerge, thereby allowing the researcher to consider not only the form of relationships (intensity, frequency or strength), but also their content (interests, purposes, drives an motivations); 2. by taking into account the temporal dimension of the emergence and development of networks, without restricting attention to a limited set of mechanisms in advance. thus, the discovery of new mechanisms is a possibility. we emphasize the complementary nature of these additions because in our view quantitative methods and techniques are still most useful for capturing the higher-order structures of complex social systems. qualitative methods and techniques tend to emphasize idiosyncrasies, and thereby run the risk of losing sight of the whole. we believe sna to be at its strongest when qualitative and quantitative methods and techniques are combined (edwards & crossley, 2009). in the next section we introduce and demonstrate two approaches that can be used to balance the quantitative and qualitative components of sna1. in the first example a mixed-method approach is applied. qualitative data are analyzed using both quantitative and qualitative methods. this makes it possible to adopt a multi-staged methodology in which quantitative sna is a preliminary stage that informs qualitative research (or vice versa) (edwards, 2010). this approach is applied to a case study on inter-organizational coordination in the dutch railway system. in the second example the focus is primarily on using qualitative data to introduce a temporal dimension to the analysis of the network. the data that are at the basis of the study consist out of chronologically ordered, qualitative descriptions of interactions. quantitative methods are used to abstract system-level patterns from the data, which are subsequently qualified by offering further interpretations based on the underlying data. the approach is demonstrated through an illustrative case study on the emergence and development of a collaborative process on sustainable cluster development in the canal zone of the netherlands. 3. exploring two approaches to balance quantitative and qualitative sna 3.1. combining dna with sensemaking to study coordination in a complex system in the first case illustration we used a multi-staged approach in which a quantitative approach to dynamic network analysis informs the qualitative analysis of sensemaking at the micro-level. sensemaking is used as a lens to explain how the actors, in and through 1 the two studies have been performed by the authors independently from each other, and they are brought together here purely for illustrative purposes. 62084-15-23.indd 9 18/08/15 10:30 am 10 d. schipper and w. spekkink / balancing the quantitative and qualitative aspects interactions with each other, frame situations as a basis for (coordinated) action. in this particular case the organizations were confronted with the news that four double-switches and two rail tracks were deemed no longer fit for use. this was decided by the responsible track manager and track inspector, who also set the deadline at six o’clock in the evening for the switches and tracks to be taken out of service. the other organizations had roughly three hours to prepare for the operation. by the end of the study period, almost three hours after the involved organizations received the news, the process ended with the decision to stop all train movement during rush hour in one of the busiest parts of the netherlands. the case study aims to reconstruct and explain the dynamics that led up to this decision. more specifically, the study focuses on the changes that occur in the information flows between the involved actors, and the micro-level dynamics underlying these changes. we applied dynamic network analysis to visualize and analyze the flows of information that followed the decision to take the switches and tracks out of service. we reconstructed the information flows from qualitative data. we obtained all available recordings of telephone conversations between actors involved in the process. these recordings offer rich and complete network data. we transcribed the recordings and then translated them into numerical data. this could be done quite easily as most of the files included information on the specific actors communicating and the time of communication. in addition, we used interviews and shift reports as additional data for the dna as not all (telephone) conversations were recorded. visualizing the flows of information, and how these changed over time allowed us to determine where the most striking changes in the structure of information flows occurred. the qualitative analysis aimed specifically at developing an understanding of these changes. using sensemaking as a theoretical lens, we performed a qualitative analysis of the telephone conversations and interviews to study how the involved actors framed the situations with which they were confronted throughout the process, and how these frames informed the (coordinated) actions that they engaged in. we created an edge list from the qualitative data, containing the sources and targets of information flows and the time of communication. each row in an edge list represents a single tie in the network, and it is possible to attach variables (such as the time of occurrence) to the ties in separate columns (see table 1). it thus allows for the inclusion of information that cannot be included in, for example, an adjacency matrix. we split the entire process into time slices of thirty minutes, such that the interactions of a specific time interval can be visualized and analyzed (wolbers et al., 2013). in addition, it allows us to show the development of the communication patterns over time. we also created a two-mode table 1 an example of an edge list nodetype id nodetype id2 value time actor train dispatcher actor riib 1 1 yyyy-mm-dd hh:mm actor wd 2 am actor wd 3 am 1 yyyy-mm-dd hh:mm actor wd 2 am actor wd rbi 1 yyyy-mm-dd hh:mm 62084-15-23.indd 10 18/08/15 10:30 am d. schipper and w. spekkink / balancing the quantitative and qualitative aspects 11 network2 that shows which actors were involved in which time slice of the process. the two-mode network is recorded as an incidence matrix in which the presence (1) or absence (0) of actors in the different time slices is marked. we used two-mode variants of network centrality metrics to identify important actors (faust, 1997). in this case we used fifteen minute time slices in order to get a more detailed picture of the network development. the software package ora was used to visualize and analyze the dynamic network. figure 1 shows the number of actors involved and the instances of information sharing during each time slice. we identified a total of 205 instances of information sharing among 42 actors (or nodes). each actor represents an individual performing a specific role in the process. the number of actors involved and the interactions increases with a peak in the number of interactions at time slice 6 (when the decision was made to stop the train service). there is however a dip in the number of actors and interactions at time slice 4. perhaps even more remarkable is the spike in the number of actors and interactions between time slices 2 and 3. to explain this sudden increase between time slices 2 and 3 we have to take a look at the first time slice. figure 2 shows the network graph for the first time slice. from the network graph we can tell that the track manager (wd 2 am) gave an early warning to the actors (riib 1) in the occr3. given what we were told in various interviews, this is somewhat surprising, as most of our interviews consider the train dispatchers to be crucial for the process (they figure 1. number of actors and interactions throughout the process. 2 for detailed discussions on two-mode networks see chapter 8 in borgatti and everett (1997), borgatti and halgin (2011), scott (2013), and wasserman and faust (1994) 3 the occr is the national control room of the railway sector in which the infrastructure manager prorail and the train operating companies monitor the railway traffic flows on the main corridors. 62084-15-23.indd 11 18/08/15 10:30 am 12 d. schipper and w. spekkink / balancing the quantitative and qualitative aspects figure 2. network graph showing the interactions in the first half hour of the process. control the switches and are responsible for safe railway operations), and should therefore be among the first to be informed by the track manager. however, the track manager assumed that the occr would coordinate the process and inform the train dispatchers and traffic controllers in the regional control centers4. he thus assumed to have informed the train dispatchers indirectly. the national traffic manager (rlvl) in the occr on the contrary, decided to wait for further details and didn’t want to alarm the regional control centers just yet to avoid any unnecessary panic. as is illustrated in the time slice, the train dispatchers did receive some rumors on the decision to take the switches out of service from one of the train operating companies (lbc). in the second time slice the actors in the occr discussed the additional information on the switches and tracks they received from the track manager. this new information 4 in the regional control centers local railway traffic is monitored by regional traffic controllers and train movement is controlled by train dispatchers using switches and signaling. 62084-15-23.indd 12 18/08/15 10:30 am d. schipper and w. spekkink / balancing the quantitative and qualitative aspects 13 however once again uncoordinatedly spread through the network and caused a chain of reactions. train dispatchers and regional traffic controllers were approached by the train operating companies for confirmation on the information they had just received. however, the train dispatchers couldn’t confirm these ‘rumors’, as they were still not officially notified about the situation. the regional control center therefore felt as if they were excluded from the process and blamed the rlvl for not following procedures. as a result communication became more conflictual than problem solving. our qualitative analysis also pointed out that the abrupt changes of the network gave respondents in the occr the feeling that no one was in control and that everyone held only small pieces of information. we made a quantitative assessment of this statement by looking at the centralization of the networks5. we calculated the betweenness centralization of the networks (figure 3) to assess the potential of an actor or small group of actors to control the flows of information. we normalized our measures to compensate for changes in the size of the network. overall, the betweenness centralization remains low. this indicates that there wasn’t a central actor or group of actors, but that information was dispersed throughout the different clusters and locations in the network. thus, no one seemed to have been able to provide an overall direction to the flows of information. this resulted in much confusion among the different actors on the switches and tracks that had to be taken out of service, the reason behind this decision, and the procedure being followed. the two-mode analysis does show that the riib and the rlvl show the highest consistency over time in terms of distributing information (table 2)6. the betweenness scores of actors can be seen as an indication for their potential to digest and distribute figure 3: betweenness centralization of the network over time. 5 a network centralization measure indicates how tightly the network is organized around its most central nodes (abbasi & kapucu, 2012). 6 degree centrality of an actor indicates the percentage of time events the actor was actively communicating in. with betweenness centrality we looked at the presence of actors in unique time events. 62084-15-23.indd 13 18/08/15 10:30 am 14 d. schipper and w. spekkink / balancing the quantitative and qualitative aspects information in other time periods and thus their importance in providing communication opportunities between actors (wolbers et al., 2013). the analysis shows that the riib and the rlvl had the highest potential to offer these opportunities. the qualitative analysis shows that the rlvl and the riib framed the situation as a routine procedure and they therefore tried to restore normal practices, which explains the consistency of their activities. this involves making sure that the train dispatchers receive an official notification from a contractor, so they can take the lead in the process. thus, although the rlvl and the riib had full details on the switches and tracks, they did not feel that they were in the position to provide the train dispatchers with this information, nor that they had the power to tell them to take the switches and tracks out of service. as a result the important task of officially informing the train dispatchers was not taken up by anyone. the drop in the number of interactions during time slice 4 (figure 1) can be explained by the conflict of perceptions between the track team and the occr, as both were unaware of this situation and therefore focused on their own tasks. around half past five the track team provided the train dispatchers with the exact details on the switches and tracks along with the deadline of six o’clock. the train dispatchers decided not to cooperate with the track inspector, as he is officially not an authority that can forbid them to stop using the switches and tracks if there isn’t an immediate threat. in the perception of the train dispatchers cooperating would also give a wrong signal to the track inspector and would even pose a threat to their role during similar situations in the future. instead they demanded confirmation from a contractor along with an official reference number. yet, moments later, when a contractor also proved to lack the full details, the train dispatchers decided to stop all rail traffic, as they could no longer guarantee safe operations. 3.2. reconstructing networks from event sequences in our second case illustration we used a methodological approach that starts with the reconstruction of social processes as sequences of discrete events (boons, spekkink, & jiao 2014). the events represent theoretically relevant actions and interactions among  actors. table 2 actors overall degree and betweenness centrality in two-mode network agent  normalized degree centrality normalized betweenness 1 riib 2 0.94 0.134 2 rlvl 0.75 0.076 3 traffic control a 0.75 0.054 4 train dispatcher a3 0.63 0.036 5 tl a2 0.56 0.032 6 lvl 1 0.56 0.026 7 train dispatcher a1 0.56 0.026 8 traffic control b1 0.56 0.023 9 wd 2 am 0.50 0.031 10 smc 0.50 0.020 62084-15-23.indd 14 18/08/15 10:30 am d. schipper and w. spekkink / balancing the quantitative and qualitative aspects 15 the data on these actions and interactions are typically gathered from archival sources, including newspaper articles, documents produced by involved actors, as well as web pages. information from these sources is then recorded as chronologically ordered incidents in an event sequence dataset (cf. poole, van de ven, dooley, & holmes, 2000). the incidents are brief qualitative descriptions that include information on (1) the action or interaction performed, (2) the actors involved, (3) the date at which the action/interaction occurred, and (4) the source of the data. some incidents may be grouped together if they can be understood to refer to the same event. in this respect, incidents can be understood as indicators of events (abbott 1988, 1990). an event sequence dataset can be analyzed in different ways and each type of analysis requires further preparation of the data (boons et al., 2014; poole et al., 2000). to use the data as the basis for the analysis of network dynamics, we coded the actors that participate in the events. once the data have been coded, the information retrieved can be represented in an incidence matrix that shows the affiliation of actors to events (see figure 4). an incidence matrix can easily be converted into a valued adjacency matrix by multiplying it with a transposed version of itself. the adjacency matrix shows the direct relationships between actors, where the relationships represent the joint participation of actors in events (see figure 5)7. the more events that actors feature in together, the stronger the relationship will be, indicated by the value that is reported in the corresponding cell of the matrix. as figure 4 demonstrates, the incidence matrix can list the events in their chronological order. thus, the information about the temporal order of the event data is maintained. this quality of the matrix can be exploited by dividing the matrix into different “frames,” where each frame represents the network of actors at a different point in the process. by having these frames overlap with each other, the gradual changes that occur in the social network from one event to the next can be studied. for example, if we divide the matrix into frames of 30 events each, our first frame will contain events 1 to 30, our second frame will contain figure 4. an incidence matrix (left) and its corresponding two-mode network (right). 7 see doreian (1979–1980) and stadtfeld and geyer-schulz (2011)for similar approaches. 62084-15-23.indd 15 18/08/15 10:30 am 16 d. schipper and w. spekkink / balancing the quantitative and qualitative aspects events 2 to 31, our third frame will contain events 3 to 32, and so on. we can then convert each separate frame into an adjacency matrix, and make measurements on the resulting matrices that are relevant to sna, thereby creating time series of sna measures. to illustrate this approach, we present a case study of the evolution of biopark terneuzen, a collaboration between governments, companies, and knowledge institutes that aims to develop a sustainable industrial cluster in the canal zone of the province of zeeland in the netherlands8. more specifically, companies in the cluster want to improve their environmental performance by exchanging by-products to replace their “normal” inputs, and by sharing utilities. the initiative also aims to strengthen the local economy by attracting new economic activities to the region. we reconstructed the collaborative process using the procedure described above. we excluded informal interactions from the study, as they leave too few traces behind to be reconstructed in a reliable way. in total, 220 events were reconstructed for the collaboration. we coded all events in the dataset to identify the actors involved in them (at the organizational level). we produced several time series that represent changes in different sna measures of the collaborative network (see figures 6 to 8). producing and inspecting the time series is only the starting point of analysis. the time series offer a description of the evolution figure 5. adjacency matrix (top) and corresponding network graph (bottom). this matrix was created by multiplying the incidence matrix of figure 4 by a transposed version of itself. 8 the initiative is described in more detail in spekkink (2013) and spekkink (2014). 62084-15-23.indd 16 18/08/15 10:30 am d. schipper and w. spekkink / balancing the quantitative and qualitative aspects 17 figure 6. the size (number of actors) of the collaborative network over time. figure 7. the density of the collaborative network over time. of the collaborative network over time, but a visual inspection offers no information about the mechanisms that are at the basis of this evolution. to offer an interpretation of the network dynamics a qualitative analysis is performed of the underlying event data, using the qualitative descriptions of events to developing an understanding of the network dynamics visualized in the time series. figures 6 and 7 show the size and density of the collaborative network respectively. both time series show a sudden increase in the size around segment 45. the increased size is maintained for quite some time (although there is a dip around segment 95), but the density of the network drops again relatively quickly. as multiple explanations for these patterns are possible (including correlations between the size and density of the network) we rely on our qualitative analysis of the underlying event data to develop an understanding of the mechanisms behind these patterns. before the point where the size of the network and the network density suddenly increase the collaboration has not formally started yet, although several actors are already working on relevant projects more or less independently from each other. the sudden increase occurs when these actors are brought together for a formal process in which they discussed the benefits that could be achieved by combining their efforts; this marks the formal start of the biopark terneuzen initiative. the projects that the actors were working on independently before the start of the formal process are included in the new, overarching initiative. 62084-15-23.indd 17 18/08/15 10:30 am 18 d. schipper and w. spekkink / balancing the quantitative and qualitative aspects the fact that a large number of actors is brought together from different projects (and therefore different actor constellations that are already present in the network) is reflected in the increased density of the network. new actors, primarily knowledge institutes, were also temporarily introduced into the network to provide additional support to the collaborative process, which leads to an increase of the number of actors involved in the network. a few meetings take place in which a rather large group of actors discuss the aims of the biopark terneuzen initiative and the studies of the knowledge institutes. after a few meetings, biopark terneuzen was officially launched with a public event. the decrease in the density of the network occurs briefly after this event. a closer inspection of the sequences of events reveal that the actors involved in the collaboration worked on the implementation of the vision for biopark terneuzen in parallel projects. the projects were carried out by small constellations of actors with only limited overlap, which explains why the density of the network is relatively low during the implementation phase. in the later stages of the process density increases slightly, which coincides with a decrease in the number of actors involved in the network. at this point, some of the projects that are part of the collaboration have been successfully implemented, while others are abandoned. some companies involved in the collaboration had anticipated on governmental support for the production of biofuels, but the national government ended up deciding otherwise in fear of the negative consequences that biofuel production might have for food production. one of the companies decided to cancel its plans for a biofuel factory. another company had already constructed a biofuel factory and went bankrupt, which was also partly due to the bad economic circumstances at the time. at this stage, the activities are focused primarily on one project, which concerns the supply of co2 and residual heat from a fertilizer factory to several newly developed greenhouses. the realization of this exchange is one of the main successes of the biopark terneuzen collaboration. the shift in focus is reflected in the decrease in the size of the network, which in this case also accounts for the increase of density. figure 8 reveals that in the early stages of development the network is characterized by a relatively high degree of betweenness centralization, which indicates that there is a small number of actors that function as a bridge between different parts of the network. the pattern in figure 8 is almost entirely accounted for by position of three public actors that were involved in the process: the municipality of terneuzen, the province of zeeland, and the port authority zeeland seaports (the province being the most prominent). they obtained their high betweenness centrality through their involvement in various projects at the same time, largely due to their administrative responsibilities in these projects. they were thus in a good position to observe the commonalities that existed in the activities that were being carried out in the different projects, and to broker the relationships between the actors involved in the projects (cf. burt, 2000, 2001). these three actors are also at the basis of the biopark terneuzen collaboration. thus, in this case the mobilization for collaborative action was initiated by actors that had the best opportunity to do so, because of the special position that they took in their network. from the qualitative 62084-15-23.indd 18 18/08/15 10:30 am d. schipper and w. spekkink / balancing the quantitative and qualitative aspects 19 data we know that these public actors brought the other actors together for a deliberative process in which the vision for biopark terneuzen was developed, which marked the start of the collaboration. in the literature on collaborative governance this mechanism is known as facilitative leadership (ansell & gash, 2008; bryson, crosby, & stone, 2006; emerson, nabatchi, & balogh, 2007; vangen & huxham, 2003). figure 8 reveals that the betweenness centralization of the network decreases as different members of the network are brought together for collaboration and start interacting with each other directly. around frame 100 betweenness centralization increases again. here too, the pattern is accounted for almost entirely by a position of relatively high betweenness centrality of the three public actors. this is the phase of the process where the vision for biopark terneuzen is implemented through several parallel projects. the public organizations are typically the only organizations that are involved in all these projects, and thus they again function as bridges between the groups of actors involved in the projects. betweenness centralization decreases again near the end, as some projects are finished, while one project is abandoned. despite the setbacks that the collaborating actors faced biopark terneuzen is still in progress. 4. conclusion in this article we have developed the argument that a balanced application of quantitative and qualitative sna can contribute to a better understanding of the emergent and dynamic properties of complex social systems. we illustrated two approaches in balancing the quantitative methods of sna with qualitative methods. the case studies show that the quantitative methods of sna offer important benefits when studying complex social systems. quantitative analyses reveal system-level patterns that would remain obscure in a completely qualitative approach. for example, certain system-level patterns entail numerous indirect relationships between actors, such as the centrality of actors in communication flows and the bridging position of actors between different parts of the network. this is where the quantitative tools for sna are superior. we however took a different approach than traditional quantitative methods of sna, which figure 8. betweenness centralization of the network over time. 62084-15-23.indd 19 18/08/15 10:30 am 20 d. schipper and w. spekkink / balancing the quantitative and qualitative aspects solely look for effects of the network structure on observed behavior. we used the quantitative measurement of network structures and dynamics as useful starting points for a more in-depth, qualitative analysis of the mechanisms behind these system-level patterns. as we discussed in section 2, there are other approaches that achieve similar results through statistical modeling and simulation. although these approaches allow for more statistical rigor in the analysis, the type of mechanisms that can be considered through them are typically of a very general nature. qualitative data and analysis can guide the interpretation of findings by offering a more vivid picture of the sequences of events that are at the basis of the observed system-level patterns, as has been shown in the second case. in addition, as the first case has shown, qualitative data and analysis can offer insight in the quality and meaning of ties for actors (edwards, 2010). for instance, it showed that the actors in the occr did not use their central positions in the network to provide others with crucial information, because they did not believe that it was their role. qualitative data and analysis therefore offer important contextual details to gain a good understanding of the observed system level patterns and contributes to a greater understanding of the changes in these patterns. we believe that the combination of a quantitative and qualitative analysis does not only offer important value for those researchers who mainly apply traditional quantitative sna. a mixed-method approach can also be beneficial for qualitative researchers, who are used to doing interviews and writing thick case descriptions. as has been shown in this article, sna can be an interesting tool to structure these case descriptions, by abstracting and visualizing system-level patterns and by offering useful starting points for a more in-depth, qualitative analysis. in addition, visualizations of (changes in) networks of relationships can also serve as aid in the collection of additional qualitative data by discussing the visualizations with respondents. for example, actors in the dutch railway system declared that they were unaware of their relatively central position in the network and thus their potential to steer the process. as edwards (2010) notices, there is no one best way to integrate quantitative and qualitative methods in sna. in this regard, we see our own approach as complementary to existing approaches in the sna literature, and not as a replacement. overall, both approaches introduced in this article helped us to improve our understanding of complex social systems by establishing a stronger link between micro-level patterns and emergent patterns at the system level without completely reducing one to the other. in addition, both approaches relatively easily deal with the highly dynamic nature of complex social systems, by making possible a relatively fine-grained reconstruction of the dynamics that have occurred. references abbasi, a., & kapucu, n. (2012). structural dynamics of organizations during the evolution of interorganizational networks in disaster response. journal of homeland security and emergency management, 9(1). doi:10.1515/1547-7355.1975 abbott, a. (1988). transcending general linear reality. sociological theory, 6(2), 169–186. 62084-15-23.indd 20 18/08/15 10:30 am d. schipper and w. spekkink / balancing the quantitative and qualitative aspects 21 abbott, a. (1990). a primer on sequence methods. organization science, 1(4), 375–392. ahuja, g. (2000). collaboration networks, structural holes, and innovation: a longitudinal study. administrative science quarterly, 45(3), 425–455. ansell, c., & gash, a. (2008). collaborative governance in theory and practice. journal of public administration research and theory, 18(4), 543–571. boons, f., spekkink, w., & jiao, w. (2014). a process perspective on industrial symbiosis. journal of industrial ecology, 18(3), 341–355. borgatti, s. p., & everett, m. g. (1997). network analysis of 2-mode data. social networks, 19(3), 243–269. borgatti, s. p., & halgin, d. s. (2011). analyzing affiliation networks. in j. scott & p. j. carrington (eds.), the sage handbook of social network analysis (pp. 417–433). thousand oaks, ca: sage. borgatti, s. p., mehra, a., brass, d. j., & labianca, g. (2009). network analysis in the social sciences. science, 323(5916), 892–895. bryson, j. m., crosby, b. c., & stone, m. m. (2006). the design and implementation of cross-sector collaborations: propositions from the literature [special issue]. public administration review, 66, 44–55. buijs, j. m., eshuis, j., & byrne, d. (2009). approaches to researching complexity in public management. in g. teisman, a. van buuren & l. gerrits (eds.), managing complex governance systems: dynamics, selforganization and coevolution in public investments (pp. 37–55). new york, ny: routledge. burt, r. s. (2000). the network structure of social capital. research in organizational behavior, 22(2000), 345–423. burt, r. s. (2001). structural holes versus network closure as social capital. in n. lin, k. cook, & r. s. burt (eds.), social capital: theory and research (pp. 31–56). new brunswick, new jersey: transaction publishers. byrne, d., & callaghan, g. (2013). complexity theory and the social sciences: the state of the art. london, england: routledge. carley, k. m. (2006). a dynamic network approach to the assessment of terrorist groups and the impact of alternative courses of action. in visualising network information (pp. kn1-1–kn1-10). meeting proceedings rto-mp-ist-063, keynote 1. neuilly-sur-seine, france: rto. carley, k. m., diesner, j., reminga, j., & tsvetovat, m. (2007). toward an interoperable dynamic network analysis toolkit. decision support systems, 43(4), 1324–1347. clancy, t. r., effken, j. a., & pesut, d. (2008). applications of complex systems theory in nursing education, research, and practice. nursing outlook, 56(5), 248.e3–256.e3. coviello, n. e. (2005). integrating qualitative and quantitative techniques in network analysis. qualitative market research: an international journal, 8(1), 39–60. crossley, n. (2010). the social world of the network: combining qualitative and quantitative elements in social network analysis. sociologica, 1. doi:10.2383/32049 doreian, p. (1979–1980). on the evolution of group and network structure. social networks, 2(3), 235–252. doreian, p., & stokman, f. n. (1997a). evolution of social networks. amsterdam, the netherlands: routledge. doreian, p., & stokman, f. n. (1997b). evolution of social networks: processes and principles. in f. n. stokman & p. doreian (eds.), evolution of social networks (pp. 233–250). amsterdam, the netherlands: routledge. edwards, g. (2010). mixed-method approaches to social network analysis. ncrm working paper [manuscript in preparation]. edwards, g., & crossley, n. (2009). measures and meanings: exploring the ego-net of helen kirkpatrick watts, militant suffragette. methodological innovations online, 4(1), 37–61. emerson, k., nabatchi, t., & balogh, s. (2012). an integrative framework for collaborative governance. journal of public administration research and theory, 22(1), 1–29. faust, k. (1997). centrality in affiliation networks. social networks, 19(2), 157–191. frank, o., & strauss, d. (1986). markov graphs. journal of the american statistical association, 81(395), 832–842. gerrits, l. (2012). punching clouds: an introduction to the complexity of public decision-making. litchfield park, az: emergent publications. goldstein, j. (1999). emergence as a construct: history and issues. emergence, 1(1), 49–72. 62084-15-23.indd 21 18/08/15 10:30 am 22 d. schipper and w. spekkink / balancing the quantitative and qualitative aspects heath, s., fuller, a., & johnston, b. (2009). chasing shadows: defining network boundaries in qualitative social network analysis. qualitative research, 9(5), 645–661. kapucu, n., hu, q., & khosa, s. (2014, november). the state of network research in public administration. administration & society. doi:10.1177/0095399714555752 knoke, d., & yang, s. (2008). social network analysis. london, england: sage. morçöl, g. (2012a). a complexity theory for public policy. london, england: routledge. morçöl, g. (2012b). the micro–macro problem as the core problem of complexity theory of policy processes. in l. gerrits, p. k. marks, & f. a. a. boons (eds.), compact i: public administration in complexity (pp. 307–322). litchfield park, az: emergent publications. morçöl, g. (2014). complex governance networks: an assessment of the advances and prospects. complexity, governance & networks, 1(1), 5–16. poole, m. s., van de ven, a. h., dooley, k., & holmes, m. e. (2000). organizational change and innovation processes: theory and methods for research. new york, ny: oxford university press. powell, w. w., koput, k. w., & smith-doerr, l. (1996). interorganizational collaboration and the locus of innovation: networks of learning in biotechnology. administrative science quarterly, 41(1), 116–145. powell, w. w., white, d. r., koput, k. w., smith, j. o., & owen-smith, j. (2005). the life sciences network dynamics and field evolution: the growth of interorganizational collaboration in the life sciences. american journal of sociology, 110(4), 1132–1205. robins, g., pattison, p., kalish, y., & lusher, d. (2007a). an introduction to exponential random graph (p*) models for social networks. social networks, 29(2), 173–191. robins, g., snijders, t., wang, p., handcock, m., & pattison, p. (2007b). recent developments in exponential random graph (p*) models for social networks. social networks, 29(2), 192–215. scott, j. (2013). social network analysis (3rd ed.). los angeles, ca: sage. snijders, t. a. b. (2001). the statistical evaluation of social network dynamics. sociological methodology, 31(1), 361–395. snijders, t. a. b. (2011). network dynamics. in j. scott & p. j. carrington (eds.), the sage handbook of social network analysis (pp. 501–513). thousand oaks, ca: sage. spekkink, w. (2013). institutional capacity building for industrial symbiosis in the canal zone of zeeland in the netherlands: a process analysis. journal of cleaner production, 52, 342–355. spekkink, w. (2015). building capacity for sustainable regional industrial systems: an event sequence analysis of developments in the sloe area and canal zone. journal of cleaner production, 98, 133–144. spreitzer, a., & yamasaki, s. (2004). beyond methodological tenets: the worlds of qca and sna and their benefit to policy analysis. compass research group working paper [manuscript in preparation]. stadtfeld, c., & geyer-schulz, a. (2011). analyzing event stream dynamics in two-mode networks: an exploratory analysis of private communication in a question and answer community. social networks, 33(4), 258–272. teisman, g., & gerrits, l. (2014). the emergence of complexity in the art and science of governance. complexity, governance & networks, 1(1), 1–12. vangen, s., & huxham, c. (2003). enacting leadership for collaborative advantage: dilemmas of ideology and pragmatism in the activities of partnership managers. british journal of management, 14(1), 61–76. wasserman, s., & faust, k. (1994). social network analysis: methods and applications. cambridge, england: cambridge university press. wolbers, j., groenewegen, p., mollee, j., & bím, j. (2013). incorporating time dynamics in the analysis of social networks in emergency management. journal of homeland security and emergency management, 10(2), 555–585. zaheer, a., & soda, g. (2009). network evolution: the origins of structural holes. administrative science quarterly, 54(1), 1–31. 62084-15-23.indd 22 18/08/15 10:30 am creating and maintaining innovative space a framework for unraveling the complexities of entrepreneurial systems complexity, governance & networks (2014) 23–40 23 doi: 10.7564/14-cgn16 creating and maintaining innovative space— a framework for unraveling the complexities of entrepreneurial systems kevin s. marshall university of la verne colleges of law, business and public management e-mail: kmarshall@laverne.edu this article proposes a theoretical framework for analyzing, designing, and managing innovative institutional space. innovative spaces generate macro-level, emergent outcomes. a fundamental challenge of the administrative and managerial sciences is to promulgate, adopt and enforce rules relevant and appropriate to specific goal-oriented spaces, while fostering spatial conditions that nurture the emergent, entrepreneurial properties of complex systems. emphasizing the revelations of complexity theory, this article argues that differentiation and integration are essential components of an adaptive, progressive space, and that administrative and managerial rules not only define the parameters of space in question, but also influence the innovative, progressive nature of the space. the proposed theoretical framework suggests a two-part analytical process germane to the creation and maintenance of innovative space consisting of 1) the identification and comprehension of the space to be administered or managed, and 2) the identification and analysis of relevant rules (existing or proposed) that positively or negatively impact the emergent properties of such space. keywords: complexity, decision-making, entrepreneurship, innovation, law, managerial, organizational theory, public administration, rule-making. “rules define the private spaces within which each of us carry on our own activities.” geoffrey brennan and james buchanan (2000/1985, p.5) 1. the conventional political economy approach regarding the conditions necessary for the creation of entrepreneurial systems political economy has long concerned itself with the operational conditions necessary for entrepreneurialism to thrive and flourish. for example, the early theoretical sentiments of neo-classical political economy prognosticated the “slow progress of opulence” in environments oppressed by a stifling, heavy-handed protectionist, and regulatory authority (smith, 1766/1982, p.521, 521–542). abbe` de condillac (1776/2008) proffered article_14-16.indd 23 03/03/15 3:05 pm 24 k. s. marshall / informed governance entrepreneurial productivity requires that “trade be fully and permanently free” (p. 326). madison (1787) implicated the blessings of republicanism and a multiplicity of commercial factions, (carey, 2001), while de tocqueville noted the productive nature of voluntary associations (de tocqueville, 1835/2010, pp. 302–304). the austrian school of economics suggested that the entrepreneurial function depends upon an individual’s ability to “act” (von mises, 1949/2007, p. 11) and “transcend the boundaries of his ignorance” (hayek, 1960, p. 22). and let us not forget the fundamental roles played by clearly defined property and contract rights (marshall, 2010). the conventional political economy approach is to organize its philosophical pseudopropositions regarding the causality of entrepreneurialism upon conditions of individual, commercial and political liberties. in contrast, this article seeks to reorganize the causality question upon the operational conditions necessary for the creation and maintenance of innovative spaces; spaces within which ideas particulate, maturate, evolve, and even combine into new and improved manifestations. 2. a proposed revised approach to understanding the creation and maintenance of entrepreneurial systems based on complexity theory while the political economy literature is replete with positive and normative references to the conditions necessary for a thriving entrepreneurial function, it remains archaically expressed and organized due to the absence of a unifying framework. the derivation of such a framework is promising, especially in light of recent developments in complexity theory. the unifying nature of such a framework is even more promising given that complexity theory seemingly invites interdisciplinary approaches and methods. this article proposes a unifying framework from which to pursue an understanding of entrepreneurialism. the framework itself is essentially an interdisciplinary reconciliation of complexity theory, physics and political economy. the interdisciplinary character of the proposed framework is important to the frameworks unifying potential. austrian economist, ludwig von mises (1949/2007), instructed that “[o]ne must study the laws of human action and social cooperation as the physicist studies the laws of nature.” (p.2) in the philosophical tradition of descartes, we must start by revisiting, and perhaps even revising our understanding of “initial conditions” (hawking & mlodinow, 2010, p. 26). with a given set of properly specified initial conditions, “the laws of nature determine how a system will evolve over time” (hawking & mlodinow, 2010, p. 27). accordingly, if we seek to explain and predict schumpeterian entrepreneurial outcomes (schumpeter, 1942/2008), we must specify, (and perhaps even re-specify) the initial conditions from which such outcomes maturate. drawing from past and recent theoretical advancements in complexity theory, political economy, and newtonian/quantum physics, this article elaborates an organizing theoretical construct from which to catalogue the conditions necessary for igniting and sustaining innovative action. the organizing and theoretical construct is space. properly article_14-16.indd 24 03/03/15 3:05 pm k. s. marshall / informed governance 25 identifying and specifying the unit-of-analysis is the first and perhaps the most important step in the science of observation. complexity theory, political economy and physics all involve the analysis of emergent forces within a defined space. all three disciplines suggest, but perhaps take for granted, that emergence requires space, and the dimensions and characteristics of such space ultimately nurture or obstruct emergent activities. this article proceeds as follows: 1) it articulates the fundamental tenets and contextual relevancy of complexity theory; 2) it reiterates the platonic and socratic methodological approaches to knowledge, drawing on past observations in political economy emphasizing the limitations and fragility of knowledge manifested within complex systems; 3) it observes that, despite the growing acceptance of complexity constructs, there remains a place for newtonian reductionism in the administrative sciences; 4) it suggests (by way of analogy) that entrepreneurialism is a system of emergent forces: and 5) it proffers a unifying theoretical framework from which the administrative sciences should proceed to analyze, design, (re)configure and manage institutions. and it is from this proposed framework that this article emphasizes that the fundamental challenge of the administrative sciences lies in its ability to promulgate, adopt and enforce rules relevant and appropriate to specific spaces, while remaining cognizant of the rules necessary to nurture the emergent, entrepreneurial properties of complex systems. whether the unit of analysis is a household, neighborhood, community, municipality, state, country, strategic multi-jurisdictional alliance, a classroom, place of worship, an art studio, a manufacturing facility, office complex, etc., the proposed framework indicates that the creation and maintenance innovative space requires that the administrative authorities begin with identification and specification of the spatial parameters of the system in question. it is important to determine the emptiness of the space and its conduciveness to the entrepreneurial force of human action. the magnitude of “emptiness” implicates the issue of rules and their relevancy to creating and maintaining a complex, catallactic space. 3. the complexity construct and its contextual relevancy in the administrative and management sciences “complexity represents social systems behavior in which the dynamic synergy of both individual autonomy and social responsibility can produce adaptive emergent behaviors” (kiel, 2000, p. 68). complexity theory generalizes that complex systems energized by autonomous and responsible behavior yield surprising macro-level effects (kiel, 2000, p. 67) ranging from the idyllic to the catastrophic (taleb, 2010, p. xvi). while cause, effect and dependency are not often linearly predictive in complex, dynamic systems (taleb, 2010, pp. 358–359), the emergence of evolving and adaptive macro-patterns is nonetheless regularly experienced and expected (newell & meek, 2000, p. 83; taleb, 2010, p. xxvi, p. 358). autonomous behavior is manifested through the acts of “independent micro-level agents” (kiel, 2000, p. 67) in pursuit of independent micro-level interests. auto nomous behavior is “responsible” to the extent it conforms to system standards, article_14-16.indd 25 03/03/15 3:05 pm 26 k. s. marshall / informed governance norms or expectations that promote and nurture advantageous emergent outcomes. it is “radical” to the extent it obstructs the system’s emergent properties. given its potential for transformative and advantageous emergent outcomes, complexity is tolerated, revered, and even deliberately pursued. and yet, complexity also has the potential of producing catastrophic macro-level effects, and consequently, complexity is also often feared, discouraged, and even opposed. it merits noting that the complexity construct is not new to the social sciences. economics has long recognized the recursive nature of market activities characterized by their multilevel and dimensional feedback loops which fundamentally diminish the ability to conclude linearity with respect to cause and effect. this is the very nature of neoclassical economics’ general equilibrium model which ultimately was addressed by the introduction of the partial equilibrium model. while economic/social life is complex given its nonlinearity and its unpredictability, it nonetheless has long been acknowledged to possess and exhibit emergent properties. smith summarized this observed complexity by referencing the “invisible hand” (smith, 1776/1910, p. 400) hayek (1960) observed that “[w]e know little of the particular facts to which the whole of social activity continuously adjusts itself in order to provide what have learned to expect. we know even less of the forces that bring about this adjustment by appropriately co-ordinating individual activity.” (p. 25) the renewed attention to complexity theory is triggering a paradigm shift with respect to exploring, critiquing, or designing social system constructs. it is “shifting attention from individual components and relationships to overall pattern[s] or motif[s] created by the system” (newell & meek, 2000, p. 83) this shift in focus is driven by an expectation that complex systems generally demonstrate long-run stability with respect to producing advantageous emergent patterns. it is this long-run stability that overshadows the system’s associated risks. and the expected long-run advantageous patterns are attributed to be dynamically creative, “evolutionary, entrepreneurial, and ultimately sustainable. while stability is comforting given its perceived stasis, interim instability is disruptive and distressing. environments of disequilibria, however, frequently fuel innovative and adaptive solutions. fluctuation and recursive non-linearity often begets progressive order (kiel, 2000, p. 72). austrian economist and nobel laureate, friedrich hayek (1960), instructed: it would be more correct to think of progress as a process of formation and modification of the human intellect, a process of adaptation and learning in which not only the possibilities known to us but also our values and desires continually change. as progress consists in discovery of the not yet known, its consequences must be unpredictability. it always leads to the unknown, and the most we can expect is to gain an understanding of the kind of forces that bring it about (p. 40) the desirability of progressive order lies in its revealing and creative attributes. accordingly, complexity theory offers insight to understanding how to create and maintain innovatively sustainable and progressive social systems: “a good society, one that encourages individuals to realize their potential and permits complexity to evolve, is one that provides room for growth. its task is not to build the best institutions, article_14-16.indd 26 03/03/15 3:05 pm k. s. marshall / informed governance 27 create the most compelling beliefs, for to do so would succumb to an illusion. institutions and beliefs age rapidly; they serve our needs for a while, but soon begin to act as brakes on progress . . . the task of a good society is not to enshrine the creative solutions of the past into permanent institutions; it is rather, to make it possible for creativity to keep asserting itself.” (csikszentmihalyi, 1993, p. 276, as cited in kiel, 2000, p.72) the arts and sciences are in a constant pursuit of unraveling and understanding the complexities of entrepreneurial systems. neoclassical political economy has long revered entrepreneurial action to be socially responsible and beneficial. from a purely economic perspective, entrepreneurial activity addresses and temporarily frees society from the constraints of scarcity. from a metaphysical perspective, entrepreneurial activity provides gateways to higher levels of understanding and being. from a complexity perspective, entrepreneurialism is system within which atomistic forces are capable of producing emergent outcomes—in which eureka moments are manifested and experienced. and this entrepreneurial experience is not exclusive to only the economic and commercial realms, but rather it abounds throughout all aspects of human activity. the entrepreneurial spirit (and it innovative outcomes) recursively fuels hope of a better tomorrow despite today’s constraints. the administrative act of creating and maintaining entrepreneurial and innovative space, therefore, is individually and social advantageous. complexity theory, although with little specificity, suggests that within certain systems “it is possible for creativity to keep asserting itself” (csikszentmihalyi, 1993, p. 276 as cited by kiel, 2000, 72). the challenge for the administrative sciences is determining the attributes of such a system, and then purposefully designing a system within which such innovative and creative attributes thrive. 4. the complexity and fragility of knowledge dynamic complexity defies conventional forecasting, design, planning and analytical methods (newell & meek, 2000, p. 87). complexity theory acknowledges, and even emphasizes the “severe limitations to . . . learning from . . . observations or experience and the fragility of our knowledge” (taleb, 2010, p. xxi). complexity cautions against platonicity. platonicity is the “tendency to mistake the map for the territory, to focus on pure and well-defined ‘forms’, whether objects, like triangles, or social notions, like utopias (societies built according to some blue print of what ‘makes sense’) . . . .” (taleb, 2010, p. xxix). “platonicity is what makes us think that we understand more than we actually do” (taleb, 2010, p. xxx). complexity, on the other hand, embraces socraticity. socraticity is “the recognition that our ignorance is the beginning of wisdom” (taleb, 2010, p. xxx). it is from such ignorance that emergent knowledge is nurtured and realized. although the allure of platonicity dominates the pursuits of the social science academy, it merits noting that such concepts of complication, ignorance and the fragility of knowledge have been long acknowledged within the academy, and particularly within the field of political economy. nevertheless, the social science academy generally, regularly and cavalierly glosses over this ignorance as but a minor imperfection in its analytical processes (hayek, 1960, p. 25). article_14-16.indd 27 03/03/15 3:05 pm 28 k. s. marshall / informed governance adam smith (1776/1910) conceptualized the complexification of an “invisible hand” with emergent force capable of producing macro-level patterns and outcomes without specific intention or knowledge of doing so (p. 400). social philosopher and historian adam ferguson (1768) (in his essay on the history of civil society) identified a “state of complication” from which successive and progressive social improvements emerged not by human design but by a variety of situations within which mankind was placed, which could not be knowingly predicted nor fully comprehended. ferguson (1768) explained: every step and every movement of the multitude even in what are termed enlightened ages, are made with equal blindness to the future; and nations stumble upon establishments, which are indeed the result of human action, but not the execution of any human design. (p. 187) * * * the artifices of the beaver, the ant and the bee, are described to the wisdom of nature. those of polished nations are ascribed to themselves, and are supposed to indicate a capacity superior to that of rude minds. but the establishments of men, like those of every animal, are suggested by nature, and are the result of instinct, directed by the variety of situations in which mankind are placed. those establishments arose from successive improvements that were made, without any sense of their general effect; and they bring human affairs to a state of complication, which the greatest reach of capacity with which the human nature was ever adorned, could not have projected; nor even when the whole is carried into execution, can it be comprehended in its full extent. [emphasis added] (p. 279) political economist and social philosopher ludwig von mises (1949/2007) observed: since time immemorial men have been eager to know the prime mover, the cause of all being and all change, the ultimate substance from which everything stems and which is the cause of itself. science is more modest. it is aware of the limits of the human mind and of the human search for knowledge. it aims at tracing back at every phenomenon to its cause. but it realizes that these endeavors must necessarily strike against insurmountable walls. there are phenomena which cannot be analyzed and trace back to other phenomena. they are the ultimate given. the progress of scientific research may succeed in demonstrating that something previously considered as an ultimate given can be reduced to components. but there will always be some irreducible and unanalyzable phenomena, some ultimate given (p. 17). * * * concrete value judgments and definite human actions are not open to further analysis. we may fairly assume or believe that they are absolutely dependent upon and conditioned by their causes. but as long as we do not know how external facts—physical and physiological—produce in a human mind definite thoughts and volitions resulting in concrete acts, we have to face an insurmountable methodological dualism. in the present state of our knowledge the fundamental statements of positivism, monism and panphysicalism are mere metaphysical postulates devoid of any scientific foundation and both meaningless and useless for scientific research. reason and experience show us two separate realms: the external world of physical, chemical, and physiological phenomena and the internal world of thought, feeling, valuation, and purposeful action. no bridge connects—as far as we see today—these two spheres. identical external events result sometimes in different human responses, and different external events produce sometimes the same human response. we do not know why (p. 18). article_14-16.indd 28 03/03/15 3:05 pm k. s. marshall / informed governance 29 hayek (1960) further acknowledged the relevance and inter-relatedness of complexity, ignorance and the fragility of knowledge by observing: the misleading effect of the usual approach stands out clearly if we examine the significance of the assertion that man has created his civilization and that that he therefore can also change its institutions as he pleases. this assertion would be justified only if man had deliberately created civilization in full understanding of what he was doing or if he at least clearly knew how it was being maintained. in a sense it is true, of course, that man has made his civilization. it is the product of his actions or, rather, of the action of a few hundred generations. this does not mean, however, that civilizations is the product of human design, or even that man knows what its functioning or continued existence depends upon (p.23) the “inescapable [multidimensional] interdependence of [social] phenomena” (von mises, 1949/2007 p. 2) imposes significant limitations on one’s ability to comprehend and address the separate and differentiated realities of social life; it renders knowledge fragile. although the social science academy has been, and continues to be warned of the errors associated with reductionism and its overly simplistic causal models, it continues to offer explanatory narratives grounded upon directional, causal conjectures and probabilities. the legitimacy of such narratives seems circularly dependent upon the legitimacy of the power to engineer social solutions to social problems. since a causes b, we can socially yield b by engineering the condition of a. and since we can socially yield b, by engineering the conditions of a, then a must indeed cause b. and yet, social systems are complex, recursive, non-linear, and ultimately incapable of being reduced to a simple, linear causal model. it is interesting that economics, for example, speaks of “general equilibrium analysis,” and yet acquiesces to the complexity of market phenomena and its interdependent and recursive natures by adopting “ceteris paribus” principles or by resorting to more simplistic and manageable “partial equilibrium analysis.” economics theorizes that patterns of general equibria emerge from certain economic markets. but given the dynamic nature of such markets, such patterns of equilibrium are unobservable at any instant in time. moreover, given the complexity of such market phenomena, it is necessary for the economist to resort to abstractions and partial equilibrium analytics for the purpose of comprehending market realities. and it is from such analytical and modeling processes that social administrators find themselves engineering policy directives purposefully designed to achieve specific policy objectives. 5. the relevance and limitations of newtonian reductionism in a complex world certainly newtonian reductionism is of value in assessing, analyzing and organizing one’s journey through the complex socio-physical world (marion, 1999, p. 13). despite the challenges of complexity, reductionist approaches and modeling advance knowledge. “when models are restricted to their intended uses, they are reasonably appropriate” (marion, 1999, p. 11). article_14-16.indd 29 03/03/15 3:05 pm 30 k. s. marshall / informed governance reality is an elusive commodity, and the best one can hope for are closer and closer approximations of it. we may never arrive at absolute and total reality, and even if we did our blindness would likely prevent us from being able to confirm our arrival. nonetheless, our models serve us well, for we learn eventually to ride our elephants and use them for work. model building is a process of building and rebuilding with each cycle pushing back the darkness a bit farther than did the previous one (marion, 1999, p. 12). notwithstanding the revelations of complexity theory, the newtonian world of calculus and mechanistic design remains relevant with respect to the administration of complex systems. despite the existence of complexity, social science observers are indeed capable of measuring, causally establishing, and probabilistically predicting social phenomena, especially at the micro-level. and this reductionist approach remains relevant because it contributes to one’s ability to observe and comprehend, especially at the micro-level. given the relevancy and contributory value of both conventional reductionism and complexity theory, each has a place in the administrative and managerial sciences. the academic conversation must, and will continue to explore the appropriate relevancy of each. the following proposed framework acknowledges and reconciles these two theoretical paradigms (i.e., platonicity/reductionism and socraticity/complexity) such that they jointly contribute to the design and comprehension of complex social phenomena. the application of newtonian mechanics for the purpose of abstracting the forces driving a non-linear, recursive, complex social system reveals a framework that recognizes the contributions of both complexity theory and reductionism. moreover, it provides insight with respect to designing entrepreneurial space, at least in as much as such space exhibits characteristics of emergence, a fundamental characteristic of entrepreneurialism. 6. entrepreneurial forces, newtonian forces and innovative space ludwig von mises (1949/2007) maintained that the traditional [political] economic framework of analysis proffered by the classical school was constrained by its theoretical nexus to economic phenomena and constructs (e.g., wealth, selfishness, profit, etc.) (p. 2). he further explained that from this classical school of political economy emerged a “general theory of human action, praxeology. the economic or catallactic problems [were] embedded in more general science, and can no longer be served from its connection to economic action” (p. 3). mises (1949/2007) argued: it is no longer enough to deal with the economic problems within the traditional framework. it is necessary to build the theory of catallactics upon the solid foundation of a general theory of human action, praxeology. this procedure will not only secure it against many fallacious criticisms but clarify many problems hitherto, not even adequately seen, still less satisfactorily solved. there is, especially, the fundamental problem of economic calculation. (p. 7) mises observed that “[i]n the course of social events there prevails a regularity of phenomena to which man must adjust his actions if he wishes to succeed” (p. 2). such article_14-16.indd 30 03/03/15 3:05 pm k. s. marshall / informed governance 31 action, while purposeful, need not be viewed from arbitrary normative perspectives of being good or bad, but rather may be viewed from the positive perspective that such action simply takes place. social action is an aggregation of human action, and “human action is purposeful behavior” (von mises, 1949/2007, p.12). action is “will put into operation” (von mises, 1949/2007, p.11) it is a force that drives outcomes, and ultimately emergent outcomes. it is in this regard that newton’s laws of motion are relevant and instructive. it was newton’s wish to develop a “quantitative yet comprehensive description of the physical world that accorded with factual descriptions of how things behave[d]” (lederman & teresi, 1993, p. 97). it is understood, even experienced, that nature creatively evolves in a schumpeterian manner, provided that it has the space to do so. lederman & teresi (1993) artfully provided the following quantum illustration: have only a few different styles of atoms, like smooth, rough, round, angular, and have a selected number of different sharps, but have an infinite supply of each kind. then put them in empty space [emphasis added]. . . . let those atoms move about at random. let them incessantly, occasionally colliding, sometimes sticking and collecting together. then one collection of atoms makes wine, another makes the glass in which it is served, ditto feta cheese, baklava, and olives (p. 42). from a quantum perspective, space is quintessential; without it there can be no force, no motion; and without motion there can be no creative atomic evolution (lederman & teresi, 1993, p. 42–45). thus, the laws of nature dictate that motion (and thereby forces) are relevant for evolutionary change. newton’s definition of momentum (definition two) and his second law of motion (law ii), informs that f = ma, where “f” represents force (the causative agent for change of motion), “m” represents mass, and “a” represents acceleration (lederman & teresi, 1993, p. 42–45). in nature, such force is the causative agent for change, and understanding the components of such force is critical to understanding the natural order and its infinite potential. by analogy, entrepreneurialism is a force; it is characteristically innovative in nature for which e = ma, where “e” represents the schumpeterian entrepreneurial function, “m” represents physical matter (of both earthly and cosmological origin), and “a” represents the innate, psychological, spiritual, intellectual, physical, and imaginary energies of definition ii the quantity of motion is the measure of the same, arising from the velocity and quantity of matter conjunctively. law ii the alteration of motion is ever proportional to the motive force impressed; and is made in the direction of the right line in which that force is impressed. isaac newton, the principia, 1642–1727 (p. 19) article_14-16.indd 31 03/03/15 3:05 pm 32 k. s. marshall / informed governance mankind (human capital). similar to newton’s vision of the mechanical world, schumpeterian entrepreneurialism is driven by a specified relationship between “m” and “a”, all of which requires that there be space within which this relationship is expressed, experienced, and maturated. and consistent with our evolving understanding of the dynamic, quantum world, both “m” and “a” are complex, each with immense possibilities, of which their interaction implicates infinity—as long as there exists sufficient interactive space. it is proposed herein that entrepreneurialism, as a force (i.e. an agent of change), requires space, and the creation and maintenance of such space is an important social and public policy goal. it is upon this theoretical construct that the entrepreneurial function may be better articulated and conceptualized. 7. empty space—the initial condition of entrepreneurialism and innovation empty space is paramount for the exertion of force. space is realm within which intent is actuated. it can emerge from nature or artificial design. it is multi-dimensional. it likely has both a vertical and horizontal axis. it can be unbounded and infinite; or prescribed, circumscribed or specified. it’s is as important as the matter and energy contained within its dimensions. it is the playground in which force exerts itself and motion reveals itself. space, and in particular, empty space is the initial condition within which human action, entrepreneurial action (or force) manifests itself. thus, understanding space and its dimensions is the starting point for administrative comprehension. all human action takes place within a given space, and it is in this regard, that human action is spatial. whether we are investigating output in the classroom, a firm, an agency, empty space: article_14-16.indd 32 03/03/15 3:05 pm k. s. marshall / informed governance 33 municipality, state or national sovereignty, all such investigations begin with the identification or specification of relevant spatial parameters. although often taken for granted by the administrative sciences, such identification and specification is the necessary first step in the analysis of human action. social externalities are spatial in nature. they manifest themselves within spatial certain parameters. administrative design, hierarchies, and networks are implemented with regard to spatial horizons and constraints. agency and firm strategies evolve within spatial contexts. knowingly, and often unknowingly, the relevance of spatial contexts or parameters is often taken for granted, under-estimated, or even purposively ignored within administrative design processes. 8. the premise of active and self-organizing space newtonian mechanics teaches: [t]hings move because that’s all they can do. motion is natural. . . . all things in the universe move, taking part in the great flow of time. if everything in the universe was left unattended and alone, all things would move just as they had eons ago. whatever motion they had then, they would have now. everything would flow in continuous way. but things interact with each other. and each interaction causes their motion to ‘bend’ or ‘accelerate’ (wolf, 1989, p. 38–39). similarly, classical political economy speaks of a “spontaneous,” “self-organizing” and “self-generating order” in which “every single piece has a principle motion of its own. . . .” (hayek, 1973, p. 35). praxeology teaches that human action is “an element of cosmic activity and becoming” (von mises, 1949/2007, p. 18). and that “[h]uman life is an unceasing sequence of single actions” (von mises, 1949/2007, p. 45). and thus, xi xi xi xi xi xi xi xi xi xi xi xi xi xi xi xi xi active space: article_14-16.indd 33 03/03/15 3:05 pm 34 k. s. marshall / informed governance social action, manifested through action of individuals and the aggregation of individuals, constitutes force, which if left alone moves through social space in a spontaneous, self-organizing and self-organizing manner, sometimes linearly, sometimes non-linearly. because human action is analogous to newtonian force, and because the exertion of such force requires empty space, the identification and comprehension of the relevant spatial realm within which such action (or force) is manifested should dominate the initial stages of any administrative analysis. 9. the austrian school’s catallaxy and its newtonian relevancy hayek (1976, pp. 106–109), in explaining the nature of spontaneous order, and drawing on the work of whately (1832, p. 4–6) and von mises (1949/2007, p 3), introduced and developed the term catallaxy. a “catallaxy” is the special kind of spontaneous market order” emerging from the mutual and reciprocating adjustments of many market participants within a given social space (hayek, 1976, p. 108). a catallaxy is pluralistic without a common hierarchy of particular ends (hayek, 1976, p. 109). it is a space in which human action differentiates and integrates, provided the space is empty and conducive to differentiation and integration. newton mechanics teaches that force is in a constant state of motion: “things move because that’s all they do” (wolf, 1989, p. 38). and so it is in a catallaxy. force is manifested through human action, naturally so, in pursuit of a more satisfactory state of affairs (von mises, 1949/2007, p.13). such force is natural and spontaneous, provided there is space for to exert itself, uninterrupted and unobstructed. catallactic space is xi xi xi xi xi xi xi xi xi xi xi xi xi xi xi xi xi the catallaxy of exchange: article_14-16.indd 34 03/03/15 3:05 pm k. s. marshall / informed governance 35 complex space. it is non-linear, uncertain, recursive, and yet produces emergent outcomes at the macro-level. absolute empty catallactic space is likely analogous to social anarchism—theoretical envisioned to be infinitely unbounded and unobstructed (although as a practical matter there is no “absolute empty space” in any system to the extent it consists of particles or participant-actors capable of colliding and whose collision ultimately redirects, obstructs and even eradicates the forces that emanate from their active state). 10. catallaxy with rules—using rules to define space newton illustrated that “[t]hings move because that’s all they can do. motion is natural. it’s only when motion is disrupted that we should wonder what is going on” (wolf, 1989, p. 38). human action is an initial condition of the catallaxy of exchange. it “must be considered as an ultimate given and must be studied as such” (von mises, 1949/2007, p. 18). as in physics, when spontaneous motion (or action) is disrupted, we should wonder why? what lies in the path of expected action? from the administrative science’s perspective attentive to revelations of complexity, the inquiry should focus on identifying and addressing the existence of spatial clutter that jeopardizes the attributes of empty space, a necessary condition for the force of human action to freely and infinitely manifest itself. from a social engineering perspective, such clutter will likely consist of institutionalized rules. moreover, such clutter is often addressed by the adoption of offsetting rules intended to marginalize the effect of the spatial clutter and ultimately further nurture a social space within which a catallaxy of exchange will thrive. xi xi xi catallaxy with rules: xi xi xi xi xi xi xi xi xi xi xi xi xi xi xi xi xi xi article_14-16.indd 35 03/03/15 3:05 pm 36 k. s. marshall / informed governance austrian philosopher (and economist) geoffrey brennan and political economist (and nobel laureate) james m. buchanan maintained that: “[r]ules define the private spaces within which each of us carry on our own activities” (brennan & buchanan, 2000, p. 5): [t]he rules that coordinate the actions of individuals are important and are crucial to any understanding of the interdependence process. the same individuals with the same motivations and, capacities will interact to generate quite different aggregate outcomes under differing sets of rules, with quite different implications for the well-being of every participant. . . . at least since the eighteenth century, and notably since adam smith, the influence of rules (smith’s term was “laws and institutions”) on social outcomes has been understood, and this relationship has provided the basis for a central theme in economics and political economy, particularly as derived from their classical foundations. if rules influence outcomes and if some outcomes are “better” than others, it follows that to the extent that rules can be chosen, the study and analysis of comparative rules and institutions become proper objects of our attention. without an understanding of how the individuals who make up a social order interact, and how different sets of rules affect these interactions, it is impossible for participants to make informed changes in existing rules or even to behave prudently with respect to the preservation of those rules that have proved essential to the tolerably efficient functioning of the society as such (brennan & buchanan, 2000, p. 4). 11. the challenge of the administrative and managerial sciences: promulgating and administering rules relevant and appropriate to “catallactic” spaces the fundamental challenge of the administrative sciences, therefore, lies in its promulgation, adoption, and administration of rules relevant and appropriate to specific catallactic spaces, while remaining cognizant of the rules necessary to nurture the emergent, entrepreneurial properties of complex systems. herein lays the value and contribution of complexity theory with respect to the administrative sciences. on the one hand, it reveals the limits of administrative design given the unavoidable ignorance of its participant-actors and the fragmentation of knowledge at the micro-levels. on the other hand, it reveals self-organizing and self-generating entrepreneurial and adaptive forces capable of producing emergent and progressive patterns within systems bounded by scarcity. these juxtaposing revelations ultimately provide insight into the creation and maintenance of creative and innovative space. it lies in creating and maintaining complex space conducive to spontaneous, self-organizing and self-generating order. for the administrative sciences, it reveals that the creation and maintenance of innovative space is accomplished through the promulgation, adoption and administration of rules designed to facilitate the emergence of adaptive, macro-level patterns. although the contribution of this paper springs from its emphasis on the role spatial foundations and parameters play with respect to nurturing and maintaining, thriving emergent entrepreneurial systems, it necessarily recognizes and adopts both hayekian principles of law and classical economic principles of spontaneous order: once we see that, in the absence of a unified body of knowledge of all the particulars to be taken into account, the overall order depends on the use of knowledge possessed by the individuals and article_14-16.indd 36 03/03/15 3:05 pm k. s. marshall / informed governance 37 used for their purposes, it becomes clear that the role of government in that process cannot be to determine particular results for particular individuals or groups, but only to provide certain generic conditions whose effects on the several individuals will be unpredictable. it can enhance the chances that the efforts of unknown individuals towards equally unknown aims will be successful by enforcing the observance of such abstract rules of conduct as in the light of past experience appear to be most conducive to the formation of a spontaneous order1 (hayek, 1976, p. 11). 12. the proposed framework of analysis in its most generally stated form, the proposed research question is: what drives innovation? or what conditions are necessary for innovative markets to thrive? in its more specified form: can innovative space be socially engineered (or impaired) by rules and laws? guided by complexity theory, the proposed framework of analysis directs the administrative sciences to examine the causal relationship between innovative space (the dependent variable) and a variety of explanatory variables falling within the following general categories of rules: rules of nature, rules of virtue and ethics, rules of law, and rules of management. the framework proposes that rules be examined with respect to their impact on the property of emergence experienced within complex systems, grounding such examination on the newtonian concept that emergence is ultimately force exerted within empty space: (h1): innovative space is a function of governing rules. (h0): innovative space is not influenced by governing rules. using this framework, the administrative sciences should attempt to examine the relationship between innovative space and rules from both a theoretical and quantitative perspective. as demonstrated above, the theoretical component of the framework is interdisciplinary in nature, identifying and organizing the rules of nature, virtue, law, and management upon the concept of creating and designing innovative space. the quantitative component will seek data that operationally measures the dependency of innovative space on the implementation of appropriate rules conducive to emergent outcomes. for example, the dependent variable (innovative space) might be operationally measured by the number of patents recognized by a particular jurisdiction. hypothesized corresponding explanatory variables (e.g., constitutional rules of law) might be operationally measured by the dichotomous provision and protection of such constitutional rights as freedom of assembly or speech, or the right to due process, etc. or by the provision of statutory regulations nurturing or constricting the flow of information in the market place. the figure below offers a graphic overview of potentially relevant variables, although certainly not exhaustive, to be considered in the proposed framework of analysis. 1 friedrich a. hayek, law legislation and liberty, volume 2 the mirage of social justice (university of chicago press, 1976), 11. article_14-16.indd 37 03/03/15 3:05 pm 38 k. s. marshall / informed governance 13. implications and anecdotal applications relevant to the administrative and managerial sciences: the proposed framework has significant implications with respect to the administrative and managerial sciences. first is suggests a two-part analytical process consisting of 1) the identification and comprehension the complex space in question, and 2) the identification and analysis of relevant rules (existing or proposed) that positively or negatively impact the emergent properties of such space. to the extent that entrepreneurialism constitutes a force, newtonian mechanics teaches that there is a profound relationship between the manifestation of force and the space in which it exerts itself. such space, depending on its cosmic clutter, may facilitate, modulate, exacerbate, accentuate, or even terminate the force driving entrepreneurial outcomes. while it is important to understand the directional relationships between rules (i.e. , language, cultural traditions and virtues, formal law, organizational mission statements, management standards, principles and prescriptions) and their impact on complex, catallactic space, it is first and foremost important that the scope, dimensions, and purpose of the space be identified and understood. complexity theory has long acknowledged that differentiation and integration are essential components of an adaptive, complex system (marion, 1999, p. 143–150; kiel, 2000, p. 70). integration suggests motion, and motion again implicates space. to the extent that differentiation is dynamically sustainable process, it too implicates motion, again implicating the space in which it manifests. it is reasonable then that the management and administration of creative and innovative space requires the maintenance of a space conducive to both differentiation and integration. innovative space (y) rules of nature (x1) scarcity behavior propensities newtonian mechanics quantum theory rules of law (x3): constitutional rules freedom of assembly freedom of speech contract rights property rights statutory rules regulatory rules executive orders executive enforcement rules of the firm (x4): compensation policies hiring practices non-discriminatory policies labor diversification capital diversification product diversification joint venturing globalization policies virtues (x3): respect optimism truthfulness openness innovative space ← f(naturerules + virturerules + lawrules + firmrules + xi) research framework: engineering space with rules article_14-16.indd 38 03/03/15 3:05 pm k. s. marshall / informed governance 39 both the scope and dimensions of the space, as well as the endogenous and exogenous rules relevant to the space have significant import to managers and administrators. is the catallactic space in question homogeneous in nature? or populated with diverse participant-actors? is the space bounded? unbounded? circumscribed but permeable? what rules within the space facilitate differentiation? suppress differentiation? facilitate or suppress integration? and the inquiry as to rules is an inquiry in general with broad implications. the inquiry includes rules emanating from or relating to the state of nature, linguistics, culture, psychology, tradition, habit, artificial design, statutory or constitutional promulgations, firm structure, duties, architectural design, and so on. whether the unit of analysis is a household, neighborhood, community, municipality, state, country, strategic multi-jurisdictional alliance, a classroom, place of worship, an art studio, a manufacturing facility, office complex, etc., the proposed framework indicates that the creation and maintenance innovative space requires that the administrative authorities begin with identification and specification of the spatial parameters of the system in question. it is important to determine the emptiness of the space and its conduciveness to the entrepreneurial force of human action. the magnitude of “emptiness” implicates the issue of rules and their relevancy to creating and maintaining a complex, catallactic space. references brennan, g., & buchanan, j. (2000). the collected works of james m. buchanan, vol. 10 (the reason of rules: constitutional political economy) foreword by robert d. tollison (indianapolis: liberty fund, 1999). retrieved 1/26/2015 from the world wide web: http://oll.libertyfund.org/titles/1826. carey. g. w. (2001). federalist no. 10 by james madison, 1787. in g. w. carey & j. mcclellan (eds.), the federalist (the gideon edition) [edited with an introduction, reader’s guide, constitutional cross-reference, index, and glossary]. indianapolis, in: liberty fund. retrieved from http://oll.libertyfund.org/titles/788 csikszentmihalyi, m. (1993). the evolving self: a psychology for the third millennium. new york, ny: harpercollins. de tocqueville, a. (2010). of political associations (chap. 4, vol. 2) (j. t. schleifer, trans.). in e. nolla (ed.), democracy in america (1835): historical-critical edition of de la démocratie en amérique (vols. 1–4). a bilingual french–english edition. indianapolis, in: liberty fund. (reprinted from de la democratie en amerique by e. nolla, ed., 1990, paris, france: librarie philosophique) étienne bonnot & abbe de condillac (2008). commerce and government considered in their mutual relationship (s. eltis, trans. with an introduction to his life and contribution to economics, by s. eltis & w. eltis). indianapolis, in: liberty fund. (reprinted from commerce and government considered in their mutual relationship by s. eltis & w. eltis, eds., 1997, cheltenham, england: edward elgar. originally published 1776) ferguson, a. (1768). an essay on the history of civil society (2nd ed.). london, england: printed for a. millar and t. cadell in the strand, and a. kincaid and j. bell, edinburgh. hawking, s., & mlodinow, l. (2010). the grand design. new york, ny: bantam books. hayek, f. a. (1960). the constitution of liberty. chicago, il: university of chicago press. hayek, f. a. (1973). law legislation and liberty, volume 1: rules and order. chicago, il: university of chicago press. hayek, f. a. (1976). law legislation and liberty, volume 2: the mirage of social justice. chicago, il: university of chicago press. kiel, d. (2000). the sciences of complexity and public administration, complexity as challenge and goal. in g. morcol & l. f. dennard (eds.), new sciences for public administration and policy connections and reflections (pp. 63–80). burke, va: chatelaine press. article_14-16.indd 39 03/03/15 3:05 pm 40 k. s. marshall / informed governance lederman, l., & teresi, d. (1993). the god particle: if the universe is the answer, what is the question? new york, ny: first mariner books. madison, james. federalist no. 10: “the same subject continued: the union as a safeguard against domestic faction and insurrection.” new york daily advertiser, november 22, 1787. marion, r. (1999). the edge of organization, chaos and complexity theories of formal social systems. thousand oaks, ca: sage. marshall, k. s. (2010). free enterprise and the rule of law: the political economy of executive discretion (efficiency implications of regulatory enforcement strategies). william and mary business law review, 1(2), 235–321. newell, w. n., & meek, j. w. (2000). what can public administration learn from complex systems theory? in g. morcol & l. f. dennard (eds.), new sciences for public administration and policy connections and reflections (pp. 81–105). burke, va: chatelaine press. newton, i. (1995). the principia. amherst, ny: prometheus books. (reprinted from newton’s principia, 1st american ed., carefully rev. and corr. with the life of the author by n. w. chittenden, 1848, new york, ny: d. adee) schumpeter, j. a. (2008). capitalism, socialism and democracy (with introduction by thomas k. mccraw) (first harper perennial modern thought edition). new york, ny: harper perennial. (reprinted from capitalism, socialism and democracy, 1942, 2nd ed. 1947, 3rd ed. 1950, new york, ny: harper & brothers) smith, a. (1910). the wealth of nations. london, england: j. m. dent (originally published 1776) smith, a. (1982). lectures on jurisprudence (report dated 1766). indianapolis, in: liberty fund. (reprinted from the glasgow edition of the works and correspondence of adam smith vols. 1–5, by ronal l. meek et al., eds., 1978, oxford, england: clarendon press) taleb, n. n. (2010). the black swan, the impact of the highly improbable. new york, ny: random house trade paperbacks. von mises, l. (2007). introduction: economics and praxeology (vol. 1). in b. b. greaves (ed.), human action, a treatise on economics (vols. 1–4). indianapolis, in: liberty fund. (reprinted from foundations for economic education, 4th rev. ed. 1996, new york, ny: irvington-on-hudson. originally published 1949 by yale university press) whately, r. (1832). introductory lectures on political economy, delivered in easter term 1831 (2nd and enlarged ed.). london, england: b. fellowes. retrieved from http://oll.libertyfund.org/titles/1377#whately_0208_27 wolf, f. a. (1989). taking the quantum leap: the new physics for non-scientist (perennial ed.). new york, ny: harper & row. article_14-16.indd 40 03/03/15 3:05 pm the complexity and instability of policy conditionality and transfer: imf interventions in the political economy of south america 14 university of bamberg press the complexity and instability of policy conditionality and transfer: imf interventions in the political economy of south america authors: david alemna1, kepa artaraz2, philip haynes3, shadreck mwale4 the university of brighton, united kingdom email: 1d.alemna@brighton.ac.uk, 2k.artaraz@brighton.ac.uk, 3p.haynes@brigton.ac.uk, 4s.mwale@bsms.ac.uk, international monetary fund (imf) interventions have evolved in the last sixty years based on the predominant orthodoxy in world political economy with a focus in recent decades on encouraging liberal market conditions to secure inward investment and capital flows. this has resulted in a dominant model of policy conditions and transfer, but with a debate about the contextual relevance. this paper uses an innovative approach to longitudinal research, called dynamic pattern synthesis, to compare the economic performance of south american nations between 2000-2015. the results from using this method illustrates multifinality in the imf outcome of encouraging foreign direct investment. a complex configuration of influences on this outcome are evidenced. complexity theory is used to explain the results, with the continent defined as a complex system that does not respond to simple causal policy mechanisms, but rather displays different patterns of political and economic influence in the context of global market instability. different foreign direct investment configurations result, and these illustrate that international monetary and policy interventions need to be contextual and cannot make simplistic and universal assumptions about policy problems and their mechanistic solutions. keywords: dynamic pattern synthesis; international monetary fund; south america; policy conditionalities; macroeconomic instability introduction the imf is an international financier that lends to its member countries when they experience balance of payments problems. its funding comes from deposits from all member countries. the purpose of its loans are to facilitate an individual country being able to accumulate adequate international reserves and to stabilize their currency. this should allow a continuation with payments for imports and to successfully engage in international trade. the imf assumes that its financing will create macroeconomic stability that drives growth. they also impose conditionalities on loans: for example, demanding reductions in government current accounts and spending (thacker, 1999). conditionalities are designed to cover the scope and detail of an imf intervention and the specific tools used to monitor progress (imf, 2019). these measures are intended to preserve imf funds by guaranteeing that the country’s balance of payments will be enough to permit it to repay the loan. it has been argued by a previous imf chief economist that the primary goal of the imf lending is to enable a country to borrow from other sources, for example, to attract capital from international investors, as foreign direct investment (fdi): “the imf is…tasked with helping countries hit by financial crises regain access to private credit markets” (rogoff, cited in inman, 2019). indeed, recent academic scholarship has attempted to identify this ‘catalistic effect’ of imf interventions in stimulating fdi (breen & egan, 2019). here, it is suggested that the presence of an imf intervention may serve as ‘a seal of approval’ accrediting recipient nations, thereby encouraging fdi inflows (bauer, cruz, & graham, 2012). this approach to imf activities is what is often regarded as the ‘neo-liberal turn’ in international relations, occurring after the 1970s, and where international organisations turned increasingly to facilitating open global markets as the fundamental paradigm to drive activity (jessop, 2002; stiglitz, 2002; vreeland, 2006). in this paper, we consider how the presence or complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 mailto:d.alemna@brighton.ac.uk mailto:p.haynes@brigton.ac.uk 15 university of bamberg press absence of imf interventions has influenced the fdi profile of south american countries, while also considering the influence of other key economic indicators, and situate this policy system in the framework of complexity theory. background the 1980s provided the scenario for imf intervention and application of policy conditionalities in exchange for loans during the latin american region’s debt crisis that saw the wide imposition of structural adjustment programmes (pastor, 1989). these programmes pushed national governments towards open, global market economic approaches. authors such as naomi klein have explored the lineage of these economic policies, traced it back to the economic 'shock therapy' introduced in chile by the pinochet regime in the 1970s, and referred to this in terms of 'disaster capitalism' (klein, 2007). thus, the imf has been described by critical literature as one of the global institutional overseers of a broader neoliberal globalization project that aims to enforce market-based forms of development policies, a concentration of power in the hands of transnational corporations, and the establishment of freedom of movement for capital in search of investment opportunities (greig, hulme & turner, 2007; aalbers, 2013). the south american experience in the 1980s resulted in significant increases in inequality and poverty (klein, 2007; kohl & farthing, 2009). as a result, some authors have referred to the imf as part of the ‘unholy trinity’ (along with the world bank and the world trade organisation) (peet, 2010), or referred to this organisation as a predator in a global system of economic oppression that reflects the interests of capital and of the global north (chossudovsky, 1999). the historical experience with the imf of many countries in south america is partly responsible for explaining the attitude of public officials and why some left-leaning nationalist governments were elected to office in the period of our study (2000-2015): from brazil (2003-2016), to argentina (2001-2015), ecuador (2007-2017), or bolivia (2006-2019). often following popular uprisings, these governments were referred to as ‘the pink tide’ (spronk, 2008). a common theme of these governments during this period was the rejection of neoliberal economic principles and of the international institutions that promote them (de la barra & dello buono, 2009). here is a different influence on international relations, one that we might see as synonymous with neo-marxism and the idea of a nation state that still has some relative autonomy from the power of global capital and finance. the argentinian imf relationship hiatus was marked by the collapse of the economy in 2001 and a long lapse in associated international financial relations until the recent 2018 facilitation of a $50bn rescue package (mander, 2018). all this illustrates an opposition to the work of the imf in some of the countries that need international assistance and distrust in its underpinning ideology and associated policy agendas. if the imf has moved towards making sure developing countries operate within a certain global economic framework underwritten by powerful economies with comparative advantage and currency supremacy, it is not surprising that many poorer countries have been through periods when they have resisted working with it. but to cast the international relations of the imf approach into a simple duality of global capitalism versus global anti-capitalism is too simplistic. for this reason, we turn to the framework of complexity theory. theoretical perspective: complexity of imf interventions complexity theory has had an increasing impact on approaches to social science research with a growing number of sources citing this approach (byrne & callaghan, 2013). in this paper, we argue the evaluation of an international policy instrument like imf loans requires such a new theoretical framework. this is because the cases involved (nation states) and the external context (the global economy), are always complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 16 university of bamberg press changing, and any intervention mechanism, like conditional lending, is likely to be highly dependent on time and space diversity in the underlying context and mechanisms (gerrits, 2012). there are multiple sub systems of interaction, including intergovernmental cooperation, and capital, trade and labour flows influenced by the market, but also regulated by governments. one of the best summaries of the contribution that complexity makes to these policy challenges came from the late paul cilliers (1998, 3-4). for the purposes of this research article, we have paraphrased this to define the socio-political-economic system of national countries sharing the continent of south america. in our research, the environment is the global economy, and the sub system of interest is the majority of the largest south american national economies. many have experienced at least one imf intervention in the period 2000-2015. this historical period was chosen to reflect a substantial period before and after the global financial crisis of 2008. the system is defined by many different levels, cases, and elements, in accordance with cilliers’ narrative. at the top level is the global economic system with its flows of finance, trade, and labour and the international financing institutions of the imf with its economic and political agenda, as dictated by the dominant world economic powers, like the united states and european union. although we are focused in this paper on the international role of the imf, there are other global institutions of importance at this level, such as the world bank and united nations. beneath this global ‘super system’ is the level of the nation state, and in this study, we compare nations and whether they have experienced imf interventions. we refer to these nations as ‘cases’ and apply a case-based research method, as explained in our methods section below. cilliers stated that cases (in our research, these are the nation states) are defined by the relationships and interactions between them and across the system. also, in our research we observe the influence in international relations of the powers and functions of global institutions like the imf. cilliers asserted that the relationship between cases in a system is not static, but dynamic, so there is no fixed state in the relationships between nation states and the international institutions that influence them. there may be periods of stability in these relationships, but also periods of instability, and over time we must expect different changes and paces of change. public policy is designed to promote aspects of certainty and stability and to limit instability when the market is seen as potentially destabilizing. such a view of the need for international policy was at the heart of the birth of the imf in international relations, at the bretton woods post war agreement. in this sense, the imf should operate in what snowden & boone (2007) describe as the ‘zone of complexity’. it seeks to increase economic stability rather than instability and potential chaos. another observation of cilliers is that the dynamic interactions of cases in complex systems vary in volume and frequency, but are more likely to be influenced by interactions with other cases in their locality. in international relations, we can think of local system relations being the reality that neighbours are important (haynes & haynes, 2016; collier, 2008). nations are likely to be influenced by those on their geographical borders. similarly, interactions and communications are quite likely to get scaled up from these localised relations. this is one important influence on a nation in addition to the capital flows dominated by the wealthiest nations and institutions. behaviour (like the type of communication and requests made to the imf, or government decisions to ignore the imf) are quite likely to be copied, leading to ‘positive feedback’ where nations, their institutions, and elites copy certain approaches to policy, governance, and investment. governments may copy policy ideas, including approaches negotiated with monetary lenders like the imf, or political decisions to ignore the imf. similarly, for example, large investors may exit capital from certain countries and regions of the world and copy each other’s behaviour. interventions, including international policy, can be used to check such feedback and to create what (meadows, 2009) calls ‘balancing feedback’. imf policy interventions might be thought of in this way. finally, cilliers recognizes the importance of history for a system and its behaviour, as cases (nations) are always influenced by their previous experience. there is no ‘year zero’ for nation states. this complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 17 university of bamberg press means that while countries are changing, they are always to some extent influenced by their history. yet countries have some agency as they move forward. this makes complexity research interested in how different or similar countries are over time, as cases are not pre-determined alone by history but also subject to numerous internal and external factors, not least the interactive relations that take place between them (byrne, 2005). historical comparative advantage is important in the working of global economics. the supreme economic power of the dollar in the global capital market place is a major distorting factor for poorer countries. this makes them vulnerable to capital movements that are unhelpful to their own ability to raise capital. history dictates that there is no stable economic ‘level playing field’. ideally, global institutions would build positive international relations by mitigating such extremes of comparative disadvantage (stiglitz, 2006). the task of individual nations to find the right domestic economic policy in such shifting international dynamics is extremely difficult. in their seminal review of complexity theory and public policy, geyer and rihani (2010, p. 141) conclude on an optimal policy environment: ‘there are some rules to follow; create a stable institutional framework, encourage decentralized local interactions, avoid civil strife and a stifling state structure… however, there is no perfect pattern for exactly how a country should stay within these boundaries.’ given the above theoretical framework, we argue that complexity theory is an appropriate approach through which to try and understand the interventions of the imf in the workings of nation states and their political economies in south america. method social scientists who seek to apply a complexity theoretical framework to their applied research use a variety of methods. often, they are critical of cross-sectional, single time point modelling and linear based multi-variate models, fearing that these methods are likely to give a simplistic, partial, and static view of complex dynamic circumstances (harvey & reed, 1996; reed & harvey 1992). as a result, we have seen in the ‘turn to complexity’ (urry, 2005) the promoting of mixed methods, longitudinal, and casebased approaches. longitudinal case-based approaches show the dynamics of case patterns over time (byrne, 1998). gerrits & verweij (2016. p. 8) comment, “there is a need for methods that retain the complex details of particular cases and that compare those cases in a systematic and transparent manner”. in casebased methods, cases are “a complex combination of properties, a specific whole that should not be lost or obscured in the course of the analysis – this is a holistic perspective” (berg-schlosser, et al, 2009. p. 6). these methods should assist in avoiding over-simplification, where understanding the diversity of cases is lost at the expense of over-emphasizing their similarity. in short, cases may be similar in some respects, but they are also likely to be different in others. a longitudinal case-based approach enables the researcher to better ascertain whether cases are becoming more similar or more different over time. one recent innovation to achieve such complex understanding was the integration of agent-based modelling with the case-based approach (castellani, barbrook-johnson, & schimpf, 2019). pagliarin & gerrits (2020) have recently developed a longitudinal approach to qualitative comparative analysis. one method for examining the dynamic similarities and differences in cases over time in complex circumstances is dynamic pattern synthesis (dps). this method combines exploratory cluster analysis with explanatory configurational approaches to case analysis and association (haynes, 2018). dynamic pattern synthesis examines the similarity and difference of cases at the different time points selected and the extent to which cases are similar and different over time. this can also be related to a particular outcome variable. in this research the outcome variable is foreign direct investment (fdi). complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 18 university of bamberg press the reason for this is that the imf has stated their ambition to intervene in nation states to enable them to attract new investment from global capital markets. the dps method is used in this paper to explore and compare the case interaction with some core economic indicators. the cases are twelve countries in south america, including those that experienced imf interventions in the period 2000-2015. these countries are: argentina, bolivia, brazil, chile, colombia, ecuador, guyana, paraguay, peru, suriname, uruguay, and venezuela. a hierarchical cluster analysis (hca) explores the influence of selected key economic variables on the similarity and differences of these countries (pastor, 2010). the choice of variables is influenced by the core macroeconomic variables monitored by the imf and similar international organisations. the full list of variables analyzed in the exploratory cluster analysis is in table 1. cluster dendrograms (figures 1 – 3 in the results section) reveal the similarity groupings at each time point and can be used to begin to understand changes over the timescale used. code explanation fdi foreign direct investment, net inflows (annual percentage of gdp) gdppc gross domestic product, growth per capita (annual percentage) ggnlb general government net lending/borrowing (annual percentage) iacp inflation, average consumer prices percentage change nbtt net barter, terms of trade index (2000 = 100) oda net overseas development assistance received per capita (current us$) imf prior intervention was there an imf loan, in the preceding historical period? (this variable is only used in the configurational stage of analysis, after the initial cluster explorations) table 1 list of political economy variables included1 configurational tables can then be used with the same dataset to examine the influence of variables on cluster formulation and how these configurations change over time. to do this, the variable individual case scores are also considered with regard to their point on the central tendency of the variable distribution, either being above threshold, or below threshold. the central tendency point used is the median. the mean and standard deviation statistics are also provided in these tables, as points of reference. tables of case configurations are used to identify any clusters of cases that share the same variable threshold. an additional variable is added to this configurational modelling to evaluate the influence of imf interventions. this is a dichotomous variable – countries that did experience prior imf loans, and countries that did not. when a cluster, or group of countries of interest, share the same threshold score on a given table, these similar scores are either shaded in black with white text to represent a shared score pattern above threshold, or shaded in grey to represent a shared score pattern below threshold. when analyzing the tables in the results, the boolean convention of uppercase equals above the median, lowercase equals below the median, is used (ragin, 1987; rihoux & ragin, 2009). the results tables can be used to identify which threshold scores are consistently shared by all members of a cluster or group. 1 our variable selection assumes that, given the mission statement of the imf, its activities are more likely to have an influence on certain macroeconomic undertakings of a country. for this reason, we selected six macroeconomic indicators likely to be impacted by an imf intervention. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 19 university of bamberg press results figure 1 shows the clusters resulting from the 2000 data analysis where six macro-economic variables (table 1) have been entered into the analysis. the resulting clusters of similarity are: cluster one – brazil, colombia, peru and chile; cluster two – argentina, uruguay, paraguay, bolivia; cluster 3 ecuador and venezuela. guyana (4) and suriname (5) are outliers. figure 1 hca dendrogram: core economic characteristics, south american countries, 2000 table 2 shows the data format, including the addition of the cluster membership variable and the new imf prior loan variable. the configurational approach helps to validate the clusters by explaining variable influences on each cluster membership. the evidence of shared shaded threshold scores suggests that some of the clusters are best subdivided into a and b pairs, as in this format they then share more threshold shaded evidence of similarity. in table 2, an additional categorical variable is added (imf) not used in the cluster analysis. this indicates which countries have had an imf loan intervention for the period before 2000. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 20 university of bamberg press gdppc ggnlb iacp nbtt oda cluster imf fdi brazil 2.61 -3.32 7.04 99.94 1.38 1a 1 5.03 colombia 2.91 -2.94 9.22 96.9 4.62 1a 0 2.44 peru 1.29 -2.09 3.76 98.51 15.49 1b 0 1.56 chile 4.09 -0.06 3.84 92.75 3.19 1b 0 6.24 argentina -1.89 -3.33 -0.94 96.91 1.63 2a 0 3.67 uruguay -2.28 -3.34 4.76 103.73 5.6 2a 1 1.15 paraguay -4.26 -0.95 8.98 107.28 15.74 2b 0 1.32 bolivia 0.58 -3.73 4.6 110.34 58.28 2b 1 8.77 venezuela 1.75 7.53 16.21 81.63 3.25 3 0 4.01 ecuador -0.75 -0.31 96.1 91.23 11.66 3 1 -0.13 guyana -0.97 2.83 6.11 98.71 155.25 4 1 9.42 suriname -1.21 -8.31 29.56 97.03 77.56 5 0 -16.59 mean 0.16 -1.50 15.77 97.91 29.47 2.24 median -0.09 -2.52 6.58 97.77 8.63 3.06 st dev 2.36 3.74 25.35 7.19 44.53 6.36 table 2 country case configurations ranked by cluster membership results, 2000 cluster one is split between two distinct pairs. cluster 1a, brazil and colombia, share similar variable characteristics for gdppc, ggnlb, iacp, and oda. cluster 1b, peru and chile, share similar variable characteristics for gdppc, ggnlb, iacp, and no imf interventions. cluster two is split between two similar pairs, a and b. cluster 2a, argentina and uruguay, share shaded scores for gdppc, ggnlb, iacp, and oda. cluster 2b, paraguay and bolivia share scores for nbtt and oda. the third cluster, venezuela and ecuador share ggnlb, iacp, and nbtt. guyana and suriname are outliers. the clusters indicate several different economic patterns that characterise the groupings, but none of the similarity groupings have similar fdi characteristics. one of the outliers, guyana, as the highest fdi score and did experience imf intervention. there is little evidence overall of the diverse economic patterns influencing fdi in 2000 for these twelve countries. table 3 ranks the configurations differently to table 2, to demonstrate more clearly the configurations of imf interventions with fdi outcomes. as a result, the economic cluster grouping are dispersed. the first group has higher fdi outcomes and prior imf interventions (brazil, bolivia, and guyana). the other shared shaded variable for this group is higher nbtt. the second group (chile, argentina, and venezuela) also has higher fdi outcomes, but they have achieved these without prior imf interventions. this group shares shaded scores for lower nbtt and oda. the third group (uruguay and ecuador) have lower fdi, despite experiencing imf interventions. the other shared shaded score for this pair is lower gdppc. the final group (colombia, peru, paraguay, and suriname) have lower fdi outcomes and did not experience prior imf interventions. they do not share any other economic characteristics. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 21 university of bamberg press gdppc ggnlb iacp nbtt oda cluster imf fdi guyana -0.97 2.83 6.11 98.71 155.25 4 1 9.42 bolivia 0.58 -3.73 4.6 110.34 58.28 2 1 8.77 brazil 2.61 -3.32 7.04 99.94 1.38 1 1 5.03 venezuela 1.75 7.53 16.21 81.63 3.25 3 0 4.01 chile 4.09 -0.06 3.84 92.75 3.19 1 0 6.24 argentina -1.89 -3.33 -0.94 96.91 1.63 2 0 3.67 ecuador -0.75 -0.31 96.1 91.23 11.66 3 1 -0.13 uruguay -2.28 -3.34 4.76 103.73 5.6 1 1 1.15 paraguay -4.26 -0.95 8.98 107.28 15.74 2 0 1.32 peru 1.29 -2.09 3.76 98.51 15.49 1 0 1.56 suriname -1.21 -8.31 29.56 97.03 77.56 5 0 -16.59 colombia 2.91 -2.94 9.22 96.9 4.62 1 0 2.44 mean 0.16 -1.50 15.77 97.91 29.47 2.24 median -0.09 -2.52 6.58 97.77 8.63 3.06 st dev 2.36 3.74 25.35 7.19 44.53 6.36 table 3 country case configurations ranked by fdi outcome and prior imf interventions, 2000 figure 2 shows the clusters resulting from the 2008 data analysis. the first cluster comprises of argentina, colombia, brazil and ecuador. the second cluster comprises of bolivia, paraguay, and chile. the third cluster is peru and uruguay. the fourth cluster is suriname and guyana. venezuela is an outlier. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 22 university of bamberg press figure 2 hca dendrogram: core economic characteristics, south american countries, 2008 table 4 shows the cluster configurations and provides evidence of which variables are contributing the most to each cluster construction. again, an additional categorical variable is added. this dichotomous categorical variable indicates which countries have had one or more imf prior loan interventions for the seven-year period before 2008. cluster one is divided into two sub-groups based on the configurational evidence of similarity. cluster 1a comprises of colombia and argentina and they share shaded scores of gdppc and ggnlb. cluster 1b, brazil and ecuador, share lower scores for fdi, higher scores for gdppc, and lower scores for iacp and oda. imf prior interventions are confirmed for both countries. cluster two is also divided into two homogenous sub-groups. cluster 2a, bolivia and paraguay, share shaded scores for gdppc, ggnlb, iacp, oda, and imf prior interventions. cluster 2b, is a partial outlier in this cluster, chile, which shares shaded scores with the cluster for ggnlb and iacp. cluster 3, peru and uruguay, share shaded scores for gdppc, iacp, and fdi, with both having experienced imf interventions. cluster 4, guyana and suriname share shaded scores for gdppc, nbtt, and oda. venezuela is an outlier. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 23 university of bamberg press country gdppc ggnlb iacp nbtt oda cluster imf fdi colombia 2.33 -0.25 7.00 148.86 21.75 1a 0 4.33 argentina 3.00 0.23 8.59 139.48 2.97 1a 1 2.69 brazil 4.03 -1.53 5.68 109.78 2.34 1b 1 2.99 ecuador 4.57 0.56 8.40 146.06 15.86 1b 1 1.71 bolivia 4.40 3.57 14.01 173.34 65.45 2a 1 3.07 paraguay 4.94 2.96 10.19 92.70 22.29 2a 1 1.83 chile 2.43 3.58 8.72 169.63 6.37 2b 0 10.28 peru 7.80 2.68 5.79 148.07 16.36 3 1 5.74 uruguay 6.82 -1.59 7.88 85.42 9.87 3 1 7.05 guyana 1.93 -1.92 8.10 102.82 218.34 4 1 9.29 suriname 3.04 1.61 14.67 119.02 197.30 4 0 -6.55 venezuela 3.80 -2.02 31.44 446.61 2.20 5 0 0.66 mean 4.09 0.66 10.87 156.82 48.43 3.59 median 3.92 0.40 8.50 142.77 16.11 3.03 st dev 1.71 2.08 6.76 91.62 73.25 4.22 table 4 country case configurations ranked by cluster membership results, 2008 table 5 reconfigures the 2008 data to compare the fdi outcomes with prior imf interventions. the first group have shared shaded higher fdi outcomes and prior imf interventions (bolivia, peru, uruguay, and guyana). no other economic characteristics are similar for all four. another group (colombia and chile) also have higher fdi outcomes, but they have achieved these without prior imf interventions. this pair also shares scores for gdpcc and nbtt. argentina, brazil, ecuador, and paraguay have lower fdi, despite experiencing imf interventions. the four do not share other similar economic characteristics. the final group (suriname and venezuela) have lower fdi outcomes and did not experience prior imf interventions. this pair share the shaded scores of gdpcc and icap. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 24 university of bamberg press country gdppc ggnlb iacp nbtt oda cluster imf fdi guyana 1.93 -1.92 8.10 102.82 218.34 4 1 9.29 uruguay 6.82 -1.59 7.88 85.42 9.87 3 1 7.05 peru 7.80 2.68 5.79 148.07 16.36 3 1 5.74 bolivia 4.40 3.57 14.01 173.34 65.45 2 1 3.07 brazil 4.03 -1.53 5.68 109.78 2.34 1 1 2.99 argentina 3.00 0.23 8.59 139.48 2.97 1 1 2.69 paraguay 4.94 2.96 10.19 92.70 22.29 2 1 1.83 ecuador 4.57 0.56 8.40 146.06 15.86 1 1 1.71 chile 2.43 3.58 8.72 169.63 6.37 2 0 10.28 colombia 2.33 -0.25 7.00 148.86 21.75 1 0 4.33 venezuela 3.80 -2.02 31.44 446.61 2.20 5 0 0.66 suriname 3.04 1.61 14.67 119.02 197.30 4 0 -6.55 mean 4.09 0.66 10.87 156.82 48.43 3.59 median 3.92 0.40 8.50 142.77 16.11 3.03 st dev 1.71 2.08 6.76 91.62 73.25 4.22 table 5 country case configurations ranked by fdi outcome and prior imf interventions, 2008 the cluster analysis for 2015 (figure 3) data produced the following clusters. cluster 1 has four countries with two pairs: colombia and guyana; uruguay and peru. cluster 2 has three members: argentina, paraguay, and ecuador. chile (cluster 3), bolivia (cluster 4), and venezuela (cluster 6) are outliers. cluster 5 comprises suriname and brazil. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 25 university of bamberg press figure 3 hca dendrogram: core economic characteristics, south american countries, 2015 table 6 shows the case configurations again based on the transformations of the variables to above and below threshold scores. because there were no new imf loans to these south american countries between 2007 and 2014, the imf priors categorical variable has been reconfigured in this analysis of 2015 data. the imf prior variable now comprises of three ordinal categories to reflect imf prior interventions between 1996 and 2007. category 0 means that zero loans were made. category 1 means between 1 and 2 loans were made; category 2 means that 3 or more loans were made. cluster one can be divided into two similar pairs. group 1a, colombia and guyana, share shaded scores for fdi, gdppc, ggnlb, nbtt, and oda. group 1b, peru and uruguay, share scores for fdi and ggnlb and share the highest rating for prior imf interventions. cluster 2 includes argentina, ecuador, and paraguay. the shared shaded scores are for fdi and prior imf interventions are in the mid-range. cluster 5, brazil and suriname share scores for gdppc, ggnlb, and iacp. the remaining countries are outliers. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 26 university of bamberg press country gdppc ggnlb iacp nbtt oda cluster imf fdi colombia 2.12 -3.41 5.00 110.35 28.11 1a 0 4.03 guyana 2.62 -0.21 -0.87 115.42 41.88 1a 1 4.31 peru 1.92 -2.20 3.55 159.10 10.67 1b 2 4.36 uruguay 0.02 -3.58 8.67 106.12 6.75 1b 2 4.57 argentina 1.70 -5.79 10.62 146.93 -0.43 2 1 1.98 ecuador -1.40 -5.26 3.97 118.97 19.71 2 1 1.33 paraguay 1.62 -2.41 3.13 108.57 9.06 2 1 1.94 chile 1.11 -1.86 4.35 188.90 3.02 3 0 8.63 bolivia 3.27 -6.90 4.06 94.85 73.78 4 1 1.68 brazil -4.37 -10.33 9.03 108.38 4.87 5 2 4.15 suriname -3.53 -9.36 6.90 128.05 28.89 5 0 3.81 venezuela -4.73 -9.01 121.74 268.29 1.22 6 1 0.24 mean 0.03 -5.03 15.01 137.83 18.96 1.00 3.42 median 1.37 -4.42 4.68 117.20 9.87 1.00 3.92 st dev 2.71 3.16 32.32 46.95 20.76 0.64 2.10 table 6 country case configurations ranked by cluster membership results, 20152 . this final 2015 table 7 is a reconfigured version of table 6 to rank fdi outcomes. this makes it easier to see if any overall configurational patterns are influencing fdi outcomes for all the possible variable influences, including prior imf interventions. the first group are three countries (peru, uruguay, and brazil) that have experienced 3 or more prior imf interventions and experience higher fdi in 2015. the second group includes six countries that have experienced a mid-range of imf interventions. with the expectation of guyana, these countries have lower fdi. the third group is the pair of colombia and chile who have no prior record of imf interventions in the twenty years before 2015 but are experiencing higher fdi. they also share higher ggnlb. . finally, suriname has no record of previous imf interventions and is lower on fdi. 2 note: imf variable is an ordinal variable, based on the number of prior loans from 1996 – 2007. there were no imf loans made to south american countries between 2007 – 2015. category of 0 = 0 loans, 1 = 1 or 2 loans, 2 = 3 or more loans complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 27 university of bamberg press country gdppc ggnlb inflation nbtt oda cluster imf fdi uruguay 0.02 -3.58 8.67 106.12 6.75 1 2 4.57 brazil -4.37 -10.33 9.03 108.38 4.87 5 2 4.15 peru 1.92 -2.20 3.55 159.10 10.67 1 2 4.36 guyana 2.62 -0.21 -0.87 115.42 41.88 1 1 4.31 argentina 1.70 -5.79 10.62 146.93 -0.43 2 1 1.98 paraguay 1.62 -2.41 3.13 108.57 9.06 2 1 1.94 bolivia 3.27 -6.90 4.06 94.85 73.78 4 1 1.68 ecuador -1.40 -5.26 3.97 118.97 19.71 2 1 1.33 venezuela -4.73 -9.01 121.74 268.29 1.22 6 1 0.24 colombia 2.12 -3.41 5.00 110.35 28.11 1 0 4.03 chile 1.11 -1.86 4.35 188.90 3.02 3 0 8.63 suriname -3.53 -9.36 6.90 128.05 28.89 5 0 3.81 mean 0.03 -5.03 15.01 137.83 18.96 1.00 3.42 median 1.37 -4.42 4.68 117.20 9.87 1.00 3.92 st dev 2.71 3.16 32.32 46.95 20.76 0.45 2.10 table 7 case and variable configurations ranked by fdi outcomes, 2015 discussion table 8 shows the variable averages when comparing the three time points studied (2000, 2008, 2015). this gives some indication of the trajectory of each indicator in the period studied, but it is not a trend analysis that includes each intervening year. fdi gdppc ggnlb iacp nbtt oda 2000 3.06 -0.09 -2.52 6.58 97.77 8.63 2008 3.03 3.92 0.40 8.50 142.77 16.11 2015 3.92 1.37 -4.42 4.68 117.20 9.87 table 8 variable trends: averages for 12 south american countries, 2000, 2008, 2015 foreign direct investment as a percentage of gdp is stable between 2000 and 2008, with an incremental upward trajectory after the financial crisis in the period when the imf lending is not active in south america. gross domestic product per capita is negative at the time of the south american debt crisis in 2000, and peaks at the beginning of the financial crisis. this is because the global financial crisis had a mild effect on many south american countries with economies dominated by the export of commodities—many of which were quite high in prices. these countries were relatively disconnected from financial flows between some of the global financial centres (i.e. wall street, city of london, hong kong, etc). complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 28 university of bamberg press gross government borrowing turns negative in 2015, perhaps indicating an ease for governments in raising borrowing in the post 2008 crisis period when they are not dependent on the imf. inflation also shows some downward consolidation after the 2008 financial crisis. net barter terms of trade, as a measure of export earnings over import earnings, is at its most positive at the beginning of the global financial crisis in 2008, as the dollar begins to strengthen making exports more competitive. this indicator remains positive in 2015. overseas development assistance as a percentage of gdp is at its highest in 2008. when the influence of imf interventions is configured with fdi outcomes there is no evidence of universal related patterns with the other economic data (tables 3, 5, 7). small groups of countries are periodically influenced by certain variable patterns, but the patterns are time limited and dynamic over the medium and longer term. in 2000, guyana, bolivia, and brazil experience considerably above average fdi and above average terms of trade. inflation is close to, or below, average for these countries and government borrowing is below -4% (where the maximum lend is -8%). their economies are stable and imf lending seems to contribute to this stability with resulting inward investment. but other configurations are rather different and do not follow this trend. venezuela, chile, and argentina achieve lower rates of fdi (still above average) but without imf loans. terms of trade, and overseas development assistance are below average for these three economies. they are a dissimilar group of countries and all were situated in different economic clusters in the hca stage of the analysis (table 2). this is most likely a reflection of the differing conditions of their specialist trade markets at that time. such markets can be a key magnet for inward investment when a specific market is growing, but subject to limited growth periods and fluctuations. in contrast, ecuador and uruguay have considerably below average fdi, despite receiving imf loans. they share below average (and negative) gross domestic product growth. this suggests their economic performance is not sufficient to appeal to inward investment at a time when other south american countries have economies performing better. the exploratory cluster analysis of economic performance (table 2) confirms four countries in cluster 1 with considerably above average gross domestic product growth (ranging from 1% to 4%) and three achieve above average fdi, but only one has received imf support (brazil). the lack of generality to the link between growth and fdi is evidenced when one sees that the highest fdi performance at 9% is from two countries bolivia and guyana, both with weaker growth below 1%. these nations also share above average terms of trade. here there is a different attraction for inward investment, again likely to be specific to the unique trade each country offers and the context of that trade market. in 2008, the highest growth countries, peru and uruguay (rates above 7%) in cluster 3 (table 4) also have some of the highest rates of fdi (above 6%) and experienced imf interventions. they share below average inflation at a time when the average is relatively high at 8%. nevertheless, the highest rates of fdi (above 9%) are from countries with their own very different economic characteristics and below average rates of growth (chile, guyana). table 5 also evidences that chile and colombia are lower growth countries achieving above average fdi and without imf loans. brazil and ecuador in cluster 1 (table 4) are two countries that share above average growth and experienced imf interventions but, in contrast, only achieve below average fdi. the overall picture is complex, with the likelihood of numerous contingencies impacting countries economic performance and the imf only one influence amongst numerous others. the limited case patterns are temporary and contingent and do not settle into homogeneous and generalizable relationships over time. after the financial crisis, in 2015, all the nations have not experienced further imf interventions. the reference in tables 6 and 7 to imf loans is historical, summarizing interventions between 1996 and 2007. in table 6, the relationship between growth and fdi shows equifinality with outcomes of above average fdi (>4%) resulting from both above average positive growth (colombia and guyana) in cluster 1a, and shared negative below average growth (brazil and suriname) in cluster 5. the latter two have an complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 29 university of bamberg press economic performance that is noticeably worse than many of their peers (substantial pressures on government borrowing, and higher rates of inflation). peru and uruguay (cluster 1b) have a history of greater frequency of previous imf loans compared to other countries and share above average restraint on government borrowing at a time when all countries have moved into negative territory with this indicator (average = -4%). possibly the restraint on government borrowing is linked to previous reinforcement of imf discipline. they continue to attract above average fdi. table7 shows that brazil shares the interventionalist imf history of peru and uruguay, and yet continues to attract fdi despite the evidence on borrowing and inflation that it has neglected the imf’s classical economic disciplines. in 2015, colombia and chile are the other economies with above average fdi performance, yet with no recent history of imf involvement. nevertheless, government borrowing, and inflation also show some evidence of ‘economic discipline’, demonstrating close to average or below average achievements when compared to the other nations. when imf interventions in the sample are examined over the period 2000 -2015 there is no evidence of a core mechanistic and predictive effect with regard to imf interventions and fdi. there is some evidence of geographical location driving partial economic similarity as several pairings in the cluster analysis dendrograms show evidence of this type of similarity in 2000 and 2008 (e.g. colombia and brazil; bolivia and paraguay; peru and uruguay; guyana and suriname). after this, such similarity between neighbour countries is not evident. as regards, considering the possible influence of imf interventions on future fdi there is a multifinality of results. imf interventions look to be related to higher foreign direct investment in some countries, but not in others. levels of growth also look to be a significant influence on fdi in some circumstances, but not others. after 2008 the imf is much less active in the 12 countries studied, but average foreign direct investment as a proportion of gdp remains stable (table 8). similarly, there is little evidence of a consistent pattern of intervening economic factors in these fdi patterns. there is some limited evidence in 2015 that countries with less negative outcomes for government borrowing are more likely to secure better performance for fdi (table 7: see peru, uruguay, guyana, colombia, chile) where brazil is an exception with government debt of 10% of gross domestic product. this pattern is temporal and not evidenced in 2008 and 2000. the complexity and case configurational approach highlights the indeterminate outcomes of imf interventions. neoliberal accounts see the imf as a global banking institution that restructures national economies towards a more liberal market place which will prove attractive for future investors (homedes & ugalde, 2005; yeldan, 2006). they argue only market investment and activity can bring prosperity. countries must depend on private finance rather than government and central bank expansion of the economy if they are to become prosperous. the alternative neo-marxist accounts of the imf argue instead that this is a global capitalist institution seeking to spread the doctrine of ‘hyper marketization’ where capital must flow freely in and out of countries and extract maximum profit, regardless of the social and environmental cost to the local population. imf conditions to privatize state assets and services are seen as the key evidence of this pro-market ‘anti-social’ policy (thacker, 1999). the complexity perspective lies somewhere between these two approaches, but arguably closer to the structuralism of marxism than the market liberalism of the imperative of open capital markets. the imf is not an ‘honest broker’ of global finance as international relations pluralism might suggest. its policy approach and agenda are driven by the most powerful, richest, and influential nations. the complexities of the global political economy are therefore structured by these dominant countries and their economic orthodoxy. nevertheless, complexity demonstrates that there is some agency within individual nation states as they pursue their own agendas and partner with others against the dominant ideologies that sometimes feel against their national interests. the nature of the global market place is that it is politically steered by the dominant ideologies, but not totally controlled by authoritarian dictates. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 30 university of bamberg press therefore, at times, individual countries pursue their own interests and appear to be resisting the economic orthodoxy represented by the imf (spronk, 2008). the nature of the global market system is that it is relatively unstable, with flows of money that are very difficult to predict with any high degree of accuracy. while the flows follow reinforced patterns, such as the ‘flight of capital’ at certain times, from certain countries, these patterns are dynamic (stiglitz, 2004). they are not static and cyclic with any precise temporal repetition. this is complex system instability that is difficult to manage and predict. there is much uncertainty in the operation of international relations in such unstable market conditions and while national politics seeks to bring order to such chaos, to provide stability for a population, the ‘trial and error’ of policy interventions is more an art form than a science (geyer & rihani, 2010; haynes, 2012). it is not surprising that in such circumstances some governments behave ambivalently towards the imf. they seek to relate to it in some circumstances and to resist it in others. since 2015, several south american countries have begun to return to imf lending for economic support. there continues to be discussions about the imf changing the focus of their lending towards a notion of macroeconomic stability that might include social goals like reduced inequality and sustainability, but no clarity about how international policy interventions will adjust to achieve this (inman, 2019). conclusion as for the international role of the imf in south america, there is little evidence that it has delivered economic stability in the last twenty years. its language and rhetoric may have changed with more recent mention of goals such as macro-economic stability, reducing inequality, and lending to promote sustainability, but the nature of the conditions imposed on its lending suggest it will be many decades before such change in language changes the imf’s actual behaviour. complexity explains the limited success of the imf in south america. there is no single pattern or economic mechanism that the imf can transfer to nations that guarantee it will achieve success with its lending. in addition, global political instability reflects global economic instability. if the highest international political goals are the free operation of the capital markets, politics merely mirrors the instability and chaos of these goals. complexity implies that the global international arena needs to do better at negotiating shared values and a core and adaptable policy framework for intervention if there is to be more political and economic stability. such a form of international political system, policy transfer, and financial environment has not been achieved in the current era of globalization. references abbott, p., andersen, t. b., & tarp, f. (2010). imf and economic reform in developing countries. the quarterly review of economics and finance, 50(1), 17-26. aalbers, m. b. (2013). neoliberalism is dead … long live neoliberalism! international journal of urban and regional research, 37(3), 1083–1090. https://doi.org/10.1111/1468-2427.12065 bauer, m. e., cruz, c., & graham, b. a. (2012). democracies only: when do imf agreements serve as a seal of approval? the review of international organizations, 7(1), 33-58. breen, m., & egan, p. j. (2019). the catalytic effect of imf lending: evidence from sectoral fdi data. international interactions, 45(3), 447-473. berg-schlosser, d., de meur, g., & rihoux, b. r. (2009). qualitative comparative analysis (qca) as an approach. in b. rihoux, & c. ragin, configurational comparative methods: qualitative comparative analysis (qca) and related techniques (pp. 19-32). london: sage. broome, a. (2015). back to basics: the great recession and the narrowing of imf policy advice. governance, 28(2), 147-165. byrne, d. (1998). complexity theory and the social sciences. london: routledge. byrne, d. (2005). complexity, configurations and cases. theory, culture & society, 22(5), 95-111. byrne, d., & callaghan, g. (2013). complexity theory and the social sciences: the state of the art. abingdon: oxon: routledge . castellani, b., barbrook-johnson, p., & schimpf, c. (2019). case-based methods and agent-based modelling: bridging the divide complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 https://doi.org/10.1111/1468-2427.12065 31 university of bamberg press to leverage their combined strengths. international journal of social research methodology. doi:10.1080/13645579.2018.1563972 chossudovsky, m. (1999) the globalisation of poverty: impacts of imf and world bank reforms. london: zed books. cilliers, p. (1998). complexity and postmodernism: understanding complex systems. london: routledge. collier, p. (2008). the bottom billion: why the poorest countries are failing and what can be done about it. oxford university press, usa de la barra, x., and dello buono, r. a. (2009) latin america after the neoliberal debacle: another region is possible. lanham: rowman and littlefield. gerrits, l. (2012). punching clouds: an introduction to the complexity of public decision-making. emergent publications. gerrits, l., & verweij, s. (2016). qualitative comparative analysis as a method for evaluating complex cases. zeitschrift fur evaluation, 15(1), 7-22. geyer, r., & rihani, s. (2010). complexity and public policy: a new approach to 21st century politics, policy and society. london routledge. greig, a., hulme, d., turner., m. (2007) challenging global inequality: development theory and practice in the 21st century. london: palgrave. harvey, d. l., & reed, h. (1996). social science as the study of complex systems. in d. l. kiel & e. elliot (eds.), chaos theory in the social sciences. foundations and applications (pp.95–323). ann arbor. haynes, p. (2018). social synthesis: finding dynamic patterns in complex social systems. oxon: routledge. haynes, p., & haynes, j. (2016). convergence and heterogeneity in euro based economies: stability and dynamics. economies, 4(3), 1-16. haynes, p. (2012) public policy beyond the financial crisis: an international comparative perspective. oxon: routledge. homedes, n., & ugalde, a. (2005). why neoliberal health reforms have failed in latin america. health policy, 71(1), 83-96. imf (2019). imf conditionality. retrieved 16 november 2019, from https://www.imf.org/en/about/factsheets/sheets/2016/08/02/21/28/imf-conditionality inman, p. (2019, november 29th). georgieva's brave new agenda at imf threatened by economic storm clouds. the guardian, pp. https://www.theguardian.com/business/2019/oct/12/kristalina-georgieva-imf-washington-climate-crisis. jessop, r. (2002). the future of the capitalist state. polity press klein, n. (2007) the shock doctrine: the rise of disaster capitalism. london: allen lane. kohl, b. and farthing, l. (2009) “less than fully satisfactory development outcomes": international financial institutions and social unrest in bolivia. latin american perspectives, 36(3): 59-78. mander, b. (2018). argentine investors welcome $50bn imf rescue deal. financial times, june 8 2018. available on https://www.ft.com/content/40372974-6b2c-11e8-8cf3-0c230fa67aec (accessed 15 november 2019). meadows, d. (2009). thinking in systems: a primer. london: earthscan. pagliarin, s. & gerrits, l. (2020) trajectory-based qualitative comparative analysis: accounting for case-based time dynamics. methodological innovations 13(3) https://doi.org/10.1177/2059799120959170 pastor, d. a. (2010). cluster analysis. in g. r. hancock, & r. mueller (eds.), the reviewer's guide to quantitative methods in the social sciences. london: routledge. pastor, m. (1989) latin america, the debt crisis, and the international monetary fund. latin american perspectives, 16(1): 79 110. peet, r. (2010) unholy trinity: the imf, world bank, and wto. london: zed books. ragin, c. (1987). the comparative method: moving beyond qualitative and quantitative strategies. berkeley, ca: university of california press. reed, m., & harvey, d. l. (1992). the new science and the old: complexity and realism in the social sciences. journal for the theory of social behaviour, 22(4), 353–380. https://doi.org/10.1111/j.1468-5914.1992.tb00224.x rihoux, b., & ragin, c. (2009). configurational comparative methods: qualitative comparative methods (qca) and related techniques. london: sage. snowden, d., & boone, m. (2007, november). a leader's framework for decision making. harvard business review, 69-77. spronk, s. (2008) ‘pink tide? neoliberalism and its alternatives in latin america’, canadian journal of latin american and caribbean studies, 33(5): 175-89. stiglitz, j. e. (2002). globalization and its discontents. norton: new york. stiglitz, j. e. (2004). capital-market liberalization, globalization, and the imf. oxford review of economic policy, 20(1), 57-71. stiglitz, j. e. (2006). making globalisation work. esri. thirty-fifth geary lecture. thacker, s. c. (1999). the high politics of imf lending. world politics 52(1) 38-75 urry, j. (2005). the complexity turn. theory, culture and society, 22(5), 1-15. vreeland j.r. 2006. the international monetary fund: politics of conditional lending, new york, ny: routledge. yeldan, e. (2006). neoliberal global remedies: from speculative-led growth to imf-led crisis in turkey. review of radical political economics, 38(2), 193-213. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-103 p. 14-31 https://www.theguardian.com/business/2019/oct/12/kristalina-georgieva-imf-washington-climate-crisis https://www.ft.com/content/40372974-6b2c-11e8-8cf3-0c230fa67aec https://doi.org/10.1177%2f2059799120959170 https://doi.org/10.1111/j.1468-5914.1992.tb00224.x the “super-network”: fostering interaction between human rights and climate change institutions 32 university of bamberg press the “super-network”: fostering interaction between human rights and climate change institutions author: andrea schapper university of stirling, division of history, heritage, and politics, united kingdom email: andrea.schapper@stir.ac.uk this paper contributes to understanding unique forms of actor constellations and their tactics in fostering institutional interaction. it explores interaction processes between the human rights and the climate regime, and more specifically, the incorporation of human rights in the 2015 paris climate agreement. during the paris negotiations, an interconstituency alliance comprised of environmental movements, human rights organizations, gender activists, indigenous peoples’ representatives, trade unions, youth groups, and faith-based organizations successfully lobbied for the incorporation of rights principles into the new climate instrument. i argue that this alliance can be grasped as a “super-network”, a network above several individual transnational advocacy networks (tans), that works across policy fields and uses information, symbols, stories, and accountability and leverage politics to foster interaction between a source institution (human rights regime) and a target institution (climate regime). by employing a package approach, which reiterates a core message of common principles that individual networks have agreed on, the “super-network” changed the practices of governments in international negotiations and fostered inter-institutional interaction. empirically, my research is mainly based on expert interviews and participatory observations at the strategic meetings of tans at three different climate negotiations in warsaw (2013), paris (2015), and bonn (2017), including follow-up skype interviews with key experts between 2013 and 2020. keywords: transnational advocacy networks; institutional interaction; human rights; climate change; paris agreement introduction in today’s complex world, emerging institutional arrangements are never isolated but embedded in a web of already existing institutions originating in different policy fields. thus, overlap, conflict, and cooperation between institutions are at the heart of many studies in international relations. the research programme of institutional interaction analyses how the development or effectiveness of one institution (source institution) affects another institution (target institution) (gehring & oberthür, 2006, 6; oberthür & prozarowska, 2013). it explores connections between policy fields and the causal mechanisms explaining regime interaction. but who are the agents behind interinstitutional relations? who fosters interlinkages between regimes from different policy arenas? and what are the politics and dynamics behind these interaction processes? so far, our knowledge of actors and actor constellations initiating interaction between source and target institution is still limited. in previous studies, the focus has clearly been on analyzing overlap, conflict, and cooperation between governmental institutions (e.g. kent, 2014, keohane & victor, 2010; oberthür & gehring, 2006). even though non-governmental actors play an important part in shaping international institutions, there is still a gap in understanding their role in institutional interaction. therefore, i address the following research question in this paper: which forms of nongovernmental actor constellations initiate institutional interaction and how do they foster interaction processes? in line with earlier studies by orsini (2013, 2016), who investigates the influence of “multiforum” non-state actors on regime complexes, this article will reveal new insights in this strand of research by shedding light on the tactics employed by transnational advocacy networks (tans) (keck & sikkink, 1998) fostering institutional interaction through commitment (gehring & oberthür, 2009). i argue that new forms of collaborative tans, so-called “super-networks”, build a network above several individual networks, combine their strengths and work across policy fields, exchange crucial information, and use symbols, stories, and accountability and leverage politics towards state governments. by employing a package approach, which repeats a core message of common principles and values that are shared across all individual networks, such a super-network can foster institutional complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 mailto:andrea.schapper@stir.ac.uk 33 university of bamberg press interaction. the term “super-network” reflects new empirical insights from novel forms of tans (keck & sikkink, 1998) that are emerging in and across various policy fields, including climate change, human rights, biodiversity politics, and nuclear disarmament. the objective of this paper is to explore these actor constellations, as well as the politics and dynamics behind institutional interaction, focusing on the specific interlinkages between the human rights and the climate regime. this paper contributes to the special issue on institutional complexity by adding a social constructivist actor perspective to the rational choice-oriented study of institutional interaction and by exploring new forms of non-state actor constellations fostering these interaction processes. it also reveals the tactics of actor networks transporting norms from one regime (human rights) into another (climate policy), creating a regime complex, i.e. a loosely coupled set of institutions (raustiala & victor 2004). tans can be grasped as communicative structures, in which a range of activists guided by principled ideas and values interact. these ideas and values are central to the networks and they determine criteria for evaluating whether particular actions and their outcomes are just or unjust (keck & sikkink 1998, 1). tans create new linkages, multiply access channels to the international system, make resources available to new actors, and help to transform practices of national sovereignty by changing governmental policies. within these networks, international and local civil society organizations (csos), foundations, the media, churches, trade unions, academics, and even members of regional or international organizations collaborate. this means several csos build a tan and several tans collaborate in a “super-network”. the overall objective of tans and super-networks is to change the policies of states and international organizations (ibid: 9). in order to do so, they revert to a range of tactics, i.e. their ideational power (orsini, 2013), including information politics, symbolic politics, leverage politics, and accountability politics (keck & sikkink, 1998, 16-25). my paper focuses on institutional interaction between the human rights and the climate change regime. the case study i investigate comprises the advocacy activities of the inter-constituency alliance for incorporating human rights in the 2015 paris agreement. i have selected to investigate this as a unique, politically relevant, and significant case to study. it is unique because it demonstrates how a new form of transnational advocacy network, a super-network – operating above individual networks and comprised of human rights organizations, including civil society and international organizations (ios), like the ohchr and unicef, environmental movements, religious actors, representatives of indigenous peoples, youth activists, gender groups, trade unions, and land rights organizations – is built. it is politically relevant because the super-network has influenced governmental decision-making at the 2015 conference of the parties (cop) where member states of the united nations framework convention on climate change (unfccc) negotiated a new climate agreement. it is significant because the super-network has initiated and fostered further institutional interaction between the human rights and climate change, resulting in the incorporation of human rights in the preamble of the 2015 paris agreement. empirically, the research for this paper is based on 20 expert interviews and participatory observations at the strategic meetings of the inter-constituency alliance, the human rights and climate change working group (hrccwg), and other tans at three different cops, i.e. cop 19 in warsaw, cop 21 in paris, and cop 23 in bonn (hosted by the fiji islands), including follow-up skype interviews with key experts between 2013 and 2020. these key experts are representatives of the interconstituency alliance, and i interviewed at least two experts from each constituency, the two initiators/coordinators of the inter-constituency alliance and its ohchr and unicef representatives. the interviews conducted were semi-structured in-depth expert interviews (witzel & reiter, 2012), and i evaluated them using a qualitative content analysis (mayring, 2014). i also draw on a content analysis of primary documents and data, comprising policy documents of member organizations of the interconstituency alliance, strategy papers of the hrccwg, unfccc outcome agreements, twitter campaigns relating to #stand4rights used by the inter-constituency alliance during the paris negotiations, information circulated via the email lists of the inter-constituency alliance and the hrccwg, and conference calls. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 34 university of bamberg press this paper is structured in the following way: i will first elucidate my analytical framework integrating insights on institutional interaction and on transnational advocacy networks. second, i will introduce the case study on the human rights and climate change working group as the main initiator of the inter-constituency alliance fostering the incorporation of human rights in the 2015 paris agreement. i will then move on to explain how exactly tans foster institutional interaction between the human rights and the climate change regime. finally, i will reflect on how the inter-constituency alliance built in paris is different from other networks and why understanding these new forms tans is relevant for the study of institutional interaction and institutional complexity, before i conclude. analytical framework for developing a better understanding of institutional interaction, the micro-macro link (buzan, jones & little, 1993), i.e. the mechanisms at play between the micro level of actors and the macro level of institutions, needs to be further established. here, constructivist scholarship highlighting the mutually constitutive character of actors and structures might be able to enrich rational choice-oriented institutionalist theories, that is scholarship on institutional interaction. in this piece, i aim to combine these two approaches, which have previously remained disconnected from each other, to build an analytical framework that helps to understand the underlying dynamics and politics of interaction between the human rights and the climate regime. institutionalization, institutional interaction, and interplay in international relations (ir) scholarship, institutions are grasped as rules and practices prescribing behavioral roles, constraining activity, and shaping expectations (keohane 1989: 3), and as functional entities managing complex issues (zürn & faude, 2013, 123). institutionalization can be understood as the process toward establishing sets of rules and practices to guide the behavior of state and non-state actors. increased institutionalization processes have, in the past, led to “material and functional overlaps between international institutions” resulting in fragmentation (zelli & van asselt, 2013, 1). an arrangement of overlapping and non-hierarchical institutions governing an issue area can develop into a regime complex, i.e. a loosely coupled set of institutions (raustiala & victor, 2004; keohane & victor, 2010, 3-4; abbott, 2014). decision-making in fragmented regime complexes can be a challenge as certain institutional arrangements often are initially made for a single sector only but then turn out to be relevant for others as well. those who make decisions, however, are often only trained in one sector and are solely familiar with structures, processes, and actors in their own policy field. institutionalization of human rights in the climate regime can best be explained with insights from scholarship on institutional interaction (young, 2002; oberthür & stokke, 2011). institutional interaction means that the institutional development or effectiveness of one institution becomes affected by another institution (gehring & oberthür, 2006, 6; oberthür & prozarowska, 2013). interaction between source and target institution can be understood as the cause and the consequence of political decisions within respective institutions (oberthür & gehring, 2009). interplay management is a key term grasping intentional efforts by actors and actor constellations addressing or improving institutional interaction (oberthür & pozarowska, 2013, 103). this involves governance decisions on institutional complexes, in which structural elements and agency are interlinked (oberthür & pozarowska, 2013, 13). one focus in the literature on institutional interaction is the identification of causal mechanisms of influence exerted from one source institution to a specific target institution (gehring & oberthür, 2006, 6-7). these first comprise cognitive interaction, or learning. here, the source institution has insights that it feeds into the decision-making process of the target institution (gehring & oberthür, 2009, 133). second, interaction through commitment means that the member states of a source institution have agreed on commitments that might be relevant for the members of the target institution as well. if there is an overlap of membership and issues to be dealt with, the commitments made in the source institution can lead to differing decision-making in the target institution (gehring complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 35 university of bamberg press & oberthür, 2009, 136). with increasing institutionalization processes, norms and standards in one policy field might also affect another, leading to overlaps and fragmentation (zelli & van asselt, 2013). interaction through commitment is particularly important for understanding interaction processes between the human rights and the climate regime. third, behavioral interaction comes into play if the source institution has obtained an output initiating behavioral changes that is meaningful for the target institution. in cases like this, the initiated changes in behavior can foster further behavioral changes (gehring & oberthür, 2009, 141144). and fourth, impact-level interaction is based on a situation of interdependence, in which a “functional linkage” (young, 2002) between the governance objectives of the institutions can be observed. if the source institution obtains an output that affects the objectives of the source institution, this impact can also influence the objectives (and effectiveness) of the target institution (gehring & oberthür, 2009, 143-144). actors, actor constellations and transnational advocacy networks so far, our knowledge on actors and actor constellations initiating and/or fostering institutional interaction is still limited. i argue that research on transnational advocacy networks (keck & sikkink, 1998) provides useful insights that help us understand how unique actor constellations use established norms in one policy field to motivate (more powerful) ios and their member states in a different policy field to change their policies accordingly. tans are usually comprised of civil society actors, including, but not limited to, nongovernmental organizations, media actors, representatives of churches and faith-based groups, foundations, and trade unions. individual state actors and international organizations can also organize in tans (keck & sikkink, 1998, 9). tans mobilize around common objectives. based on certain ideas and values, such as human rights or environmental standards, they evaluate whether policies are just or unjust – and consequently lobby for change. tans create new linkages, multiply access channels to the international system, make resources available to new actors, and help to transform practices of national sovereignty by changing inter-governmental decisions and policies. in order to do so, they revert to a range of tactics, i.e. their ideational power (orsini, 2013). keck and sikkink (1998, 16-25) have developed a typology of tactics tans use in order to initiate policy change. in this context, they highlight (1) information politics understood as strategically using information, (2) symbolic politics as drawing on symbols and stories to highlight a situation to a target audience that might be geographically distant, (3) leverage politics as network actors being able to gain moral or material leverage over state actors and ios, and (4) accountability politics referring to formerly adopted norms and policies of governmental actors and obligations to comply with them (keck & sikkink, 1998, 16-25). in this article, i use established insights on the politics of transnational advocacy to explain important dynamics of institutional interaction between two different regimes. figure 1 illustrates the analytical framework highlighting the role of tans in the interaction between institutions. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 36 university of bamberg press figure 1 analytical framework (source: own compilation based on gehring & oberthür, 2009) case study: transnational human rights advocacy in the climate arena my empirical research departs from the work of one particular tan, the human rights and climate change working group (hrccwg), that aims at institutionalizing human rights in the climate regime. the hrccwg initiated collaboration between seven different constituencies building the inter-constituency alliance in paris 2015. the activities of the working group are inspired by two main links between human rights and climate change. first, climate change impacts, like droughts, flooding, changes in precipitation, and extreme weather events can adversely affect social rights, like the rights to health, food, water and sanitation, adequate shelter or, in extreme cases, the right to selfdetermination and the right to life. second, international climate policies, like projects under the clean development mechanism (cdm) or the reducing emissions from deforestation and forest degradation (redd+) scheme, have previously led to human rights infringements on the ground. when, under the roof of these policies, local population groups cannot enter their ancestral lands anymore or are relocated, territories are flooded, and communities loose access to the river they need for fishing and agricultural purposes, social, cultural, and indigenous peoples’ rights are at risk (schapper & lederer, 2014). the hrccwg can be described as a hybrid link of predominantly civil society and some state actors operating at various scales – from the local to the global (interview indigenous rights organization, cop 19 in warsaw, 16 november 2013). among the network’s members are prominent international csos, such as the center for international environmental law (ciel), earthjustice, friends of the earth and carbon market watch, human rights watch, and amnesty international, local csos from various developing countries, gender advocates, indigenous peoples’ representatives, academics, and representatives from ios, as well as single actors from state delegations (interview human rights watch, cop 21 in paris, 8 december 2015). ios operating within the hrccwg (and within the entire inter-constituency alliance) were the office of the high commissioner of human rights (ohchr) and unicef. membership in the network is rather informal: participants can be present at one negotiation meeting but miss the next one (interview coordinator hrccwg, cop 19 in warsaw, 17 november 2013). source institution target institution cognitive interaction interaction through commitment behavioral interaction impact-level interaction transnational actor constellations complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 37 university of bamberg press since 2010, the hrccwg has initiated and contributed to several rights institutionalization processes in the climate regime, including adding rights references in the cancun agreement demanding that all climate action should be in accordance with human rights (unfccc, 2010) and reviewing the modalities and procedures of the cdm with a focus on procedural rights for those affected by cdm policies, like access to information, transparency, participation in decision-making and access to justice (kuchler, 2017). demands for strengthening procedural rights in projects under the framework of the sustainable development mechanism (sdm) are brought forward by the hrccwg in the negotiations as well. more broadly speaking, the human rights and climate change working group has also made important contributions to the 16 framework principles on human rights and the environment published by the un special rapporteur on human rights and the environment in 2018 (ohchr, 2018). specifically referring to climate change, the framework principles formulate states’ human rights obligations in the face of environmental challenges and have been understood as a precursor to adopting an international human right to a healthy environment (atapattu & schapper, 2019). in this article, i will particularly focus on the incorporation of the commitment to respect, protect and consider human rights in the preamble of the paris agreement (unfccc, 2015). even though the hrccwg as part of the entire inter-constituency alliance had unsuccessfully advocated for including rights in article 2 of the paris climate agreement determining the objective of the accord, rights references in the preamble are still meaningful due to two main reasons. first, this is the first time human rights have been incorporated in a binding environmental instrument (knox, 2015; atapattu & schapper, 2019) and this is considered to be an important step in institutionalizing human rights in the climate regime (interview quaker united nations office, 16 march 2020, via skype). second, there is now increasing evidence that the paris agreement, including its human rights reference, is used for climate litigation (adelman, 2020; wegener, 2020), e.g. in the “the people’s climate case”, in which ten families tried to compel the european union to substantially reduce greenhouse gas emissions (climate case, 2020). the “super-network” in preparation for the negotiations of a new climate instrument, two key actors within ciel, the leading coordinating cso within the human rights and climate change working group, brought together different constituencies to discuss ideas for a common advocacy strategy. as both had personal ties to other constituencies and previous experiences with cross-constituency coordination, they had the idea of bringing most constituencies together to discuss their advocacy plans in light of the prospect that a new climate agreement would be passed in paris (interview initiator interconstituency alliance, 26 march 2020, via skype). the constituencies at the unfccc negotiations are clustered groups of officially registered csos sharing certain interests and acting as observers in the process. as of 2014, the unfccc secretariat reported more than 1,600 admitted csos organized into nine constituencies, including environmental csos (engo), business and industry (bingo), indigenous peoples’ representatives (ipo), youth groups (youngo), women and gender (women and gender), and trade unions (tungo,) as well as research and independent organizations (ringo). participation in a constituency comes with several advantages: it allows observers to make interventions at certain points in the course of the state negotiation process, it facilitates the use of focal points for better coordination with the unfccc secretariat, and it enhances flexible information-sharing (unfccc, 2014). opportunities for civil society to engage through constituencies and be granted official speaking time demonstrates that environmental and climate politics can be regarded as unique policy fields facilitating advanced institutional mechanisms for access and participation (bäckstrand, 2012). at the initial meeting discussing a common cso strategy prior to the start of the paris negotiations, some officially registered constituencies, including engo, ipo, youngo, women and gender, and tungo, as well as other organized networks that are not officially registered constituencies but are observers to the unfccc process, like faith-based organizations and the lands working group (climate land ambition and rights alliance, clara), decided to establish the new intercomplexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 38 university of bamberg press constituency alliance. the alliance comprises different layers of organization ranging from the individual initiators who are representatives of one civil society organization (ciel), which is organized in a transnational advocacy network (the human rights and climate change working group) and encourages several networked constituencies (engo, ipo, youngo, women and gender, tungo, clara, and faith-based groups) to collaborate and speak with one voice. one of the key initiators of the inter-constituency alliance described its emergence highlighting the idea of formulating a common message: “we tried to see if we could do some collective planning; like identify what are the next steps and who wants to do what […] but what people seemed really interested in was the idea of having, first of hearing each other, and then having one unified message that we could all use.” (interview initiator inter-constituency alliance, 26 march 2020, via skype.) this means representatives of the constituencies themselves came up with the idea of aligning their advocacy strategy and formulating one key message; this was not planned or encouraged by the initiators. communicating one key message across such a diverse range of networks is unprecedented and unusual. previously, human rights groups would, for example, rather focus on protecting human beings (anthropocentric perspective), whereas environmental actors would prioritize ecosystem integrity (ecocentric perspective). nevertheless, the alliance managed to draft one key message with seven principles – reflecting the priorities of the seven participating constituencies and their overlapping interests – using a broad human rights framework (interview representative of indigenous peoples’ caucus, 8 april 2020, via skype). these principles included human rights (hrccwg), indigenous peoples’ rights (ipo), gender equality (women and gender), intergenerational equity (youngo), just transition of the workforce (tungo), food security (clara), and ecosystem integrity (engo). one main reason for this unity among otherwise often diverse networks was the unique opportunity to shape the next climate agreement and integrate crucial civil society concerns (interview brahma kumaris world spiritual university un representative, 20 march 2020, via skype). focusing on one key message did not necessarily mean that the abovementioned contradictions were overcome but that there was a consensus on what to focus on during the 2015 paris negotiations. during the intersessionals (pre-negotiations) in bonn in june 2015, the alliance managed to lobby for all seven principles to be placed in the operative part of the negotiation text in article 2. this was quite meaningful because article 2 was to determine the objective of the paris agreement. including human rights in this article would have required states to acknowledge that the agreement’s purpose is the protection of basic rights in the face of a changing climate (interview care, women and gender, cop 23 in bonn, 8 november 2017). two days before the final text was adopted, state representatives moved human rights to the preamble of the agreement. on 12 december 2015, the states unanimously adopted the paris agreement including human rights as part of its preamble, although a number of states, including the usa, saudi arabia, norway, the african states, and others, opposed this over longer periods of the negotiation process. thus, the paris agreement is the first legally binding climate instrument containing human rights (carazo, 2017: 114). the key sentence of the preamble of the 2015 paris agreement now reads: “[…] parties should, when taking action to address climate change, respect, promote and consider their respective obligations on human rights, the right to health, the rights of indigenous peoples, local communities, migrants, children, persons with disabilities and people in vulnerable situations and the right to development, as well as gender equality, empowerment of women and intergenerational equity” (unfccc, 2015). the remaining principles the inter-constituency alliance advocated for, i.e. just transition of the workforce, ecosystem integrity and food security, became part of the agreement’s preamble as well. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 39 university of bamberg press tactics of the inter-constituency alliance during the paris negotiations, the super-network met once a week to refine their collective strategy and to share information across the constituencies. they exchanged insights on which state actors were receptive to human rights arguments and could introduce relevant key passages into the negotiating text or which governmental delegation was opposed to the use of rights language. the alliance then decided what network or individual cso had established good personal relations and could approach the respective state delegation to advocate for human rights in the new climate instrument. what was new in paris was the package approach. whoever engaged in any advocacy meetings with state delegates, reiterated the entire message including all seven principles. an interviewee from the women and gender constituency described this as an approach uniting civil society at the negotiations: “we were telling the parties: ‘you can’t just pick which right to push forward; it’s either all or none’. it was also a way to say: ‘don’t try to divide us, because we are united’. (interview care, women and gender, cop 23 in bonn, 8 november 2017). besides these weekly meetings with all participating constituencies, the alliance heavily relied on the advocacy activities of each individual network or cso and their relationships with key governmental actors, which they had established over a period of many years. national ties also played a role here: french observers had often established meaningful personal relationships with french negotiators or observers from chad had good chances to address representatives from the republic of chad. awareness-raising in relation to human rights and bringing local concerns regarding climate change and climate policies to the international negotiation table constituted an important part of the lobbying activities. this proved to be crucial because climate negotiators often were not aware of the fact that many climate policies have adverse consequences for human rights and indigenous peoples’ rights at the local level. and as the negotiators are not necessarily trained in the field of human rights, they are also often uninformed about the rights commitments their own government has made. “a negotiator on cdm [clean development mechanism] is very likely to have absolutely no idea what we’re speaking about when you raise the human rights issue. it is just not his field, he might come from the wrong department. […] people have no idea what the human rights obligations of their countries are. […]” (interview ciel, cop 19 in warsaw, 17 november 2013). more awareness on adverse human rights impacts of climate policies has been created over the last years, and especially in paris, but highlighting the link between both policy fields remains one of the most important advocacy activities. csos and networks frequently work with case studies emphasizing the adverse effects of climate change and climate policies on local people from asia, latin america, and africa (interview environmental think tank, cop 19 in warsaw, 16 november 2013). at the negotiations, representatives from civil society present case studies during so-called side events (interview indigenous rights organization, cop 19 in warsaw, 16 november 2013) that run parallel to the closed meetings of state parties and are accessible to all governmental delegations and non-state observers. locally affected people themselves sometimes present cases on problematic rights situations. international civil society partners sponsor some of them so that they are able to join international meetings and articulate their experiences and corresponding demands. in this way, relevant networks in cooperation with their partners – affected civil society – feed local claims into the international negotiation process. a human rights watch representative summarizes the objective of transporting local claims to the international negotiation table: “[…] these won’t be the most disadvantaged people that will make it to these international negotiations. so i think this is also what we are trying to do with our work generally but also in complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 40 university of bamberg press this context is bringing the voices of those that are not usually being heard to the international negotiations” (interview human rights watch, cop 21 in paris, 8 december 2015). to understand adverse local rights impacts of climate policies is crucially important as it creates moral leverage for those countries that have ratified these human rights and are now funding these policies, like european states engaged in cdm or redd+ projects (interview environmental think tank, cop 19 in warsaw, 16 november 2013). raising awareness on local human rights implications of climate change and climate policies, informing state delegates of already existing rights commitments, and exchanging information about relevant actions of state delegations and their receptiveness to rights arguments, as well as formulating one priority human rights message were the most important tactics employed by the super-network. transnational advocacy networks fostering institutional interaction the interplay between the human rights and the climate regime we can observe in this case study is institutional interaction through commitment (gehring & oberthür, 2009, 136). standards agreed to in the source institution, the human rights regime, become relevant in the target institution, the climate regime, as well. commitments made in the human rights arena led to different decisions in the climate regime and affected the outcome of the 2015 paris agreement. even though states agreed to comply with human rights standards in a different policy field, they are still obliged to adhere to these norms when it comes to establishing and implementing climate policies as highlighted by an ohchr representative: “we think this is an issue of consistency and policy coherence that it is important that these two legal frameworks are brought together and in fact should complement each other. so the international human rights framework is a legally binding commitment made by the states and we think that commitment should be recognized in the context of environmental laws and we’re pushing hard to see that be the case.” (interview ohchr, cop 21 in paris, 10 december 2015). keck and sikkink’s typology of tactics helps us to understand how institutional interaction between the human rights and the climate regime was initiated even though many state representatives and climate negotiators were not aware of the link and/or opposed to incorporating human rights in a climate instrument. the tactics of transnational advocacy networks, in particular of the super-network, help explain the politics behind this regime interaction: at the climate negotiations, different tans use information on adverse local rights impacts of climate change and climate policies to criticize existing state practices and to highlight the need to institutionalize human rights in climate instruments. the inter-constituency alliance exchanged information on states that are receptive to rights arguments to refine their advocacy and lobbying strategy (information politics). at the side events, tans encouraged local actors to share their cases and stories from home countries to raise awareness of the situation of the target group of climate instruments (symbolic politics). these cases are presented as instances of climate injustice in which local population groups who have contributed little to greenhouse gas emissions and have few resources to adapt cannot fully enjoy their human rights due to climate impacts or experience rights infringements as a consequence of climate policy implementation. this creates moral leverage over states that have historically contributed to emissions and that are financing climate policies in developing countries (leverage politics). moreover, tans persuade states to vote for an incorporation of human rights into climate agreements and procedural rights into climate policies. mechanisms of persuasion (and discourse) function according to a logic of appropriateness (march & olsen, 1998, 951) (or a logic of arguing) and are particularly successful with (often liberal democratic) state governments (risse & ropp, 2013, 16) that have already legally committed to human rights, understanding them as part of their state identity, e.g. european states like france, sweden, and ireland (accountability politics). actively engaged and in favor of rights institutionalization are also those states that are pressured from above through tans and from below through domestic civil society organizations. these often are latin american countries with strong cso movements representing complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 41 university of bamberg press local communities’ and indigenous peoples’ concerns. among them are mexico, peru, costa rica, guatemala, and uruguay. also in favor of rights institutionalization are small island states, such as the maldives, kiribati, or samoa, that fear severe climate change consequences for the citizens living in their territory. some states (together with csos, ios and other actors of the human rights regime) also try to pressure less democratic states to vote in favor of rights institutionalization. another mechanism applied is naming and shaming, e.g. through a joint amnesty international and human rights press release widely circulated on social media labeling saudi arabia, the united states, and norway “human rights deniers” in addressing climate change (amnesty international & human rights watch, 2015). another example is annex i countries claiming that they will not fund climate policies with adverse right affects anymore, such as redd+ and cdm programs. thus, they use negative incentives or sanction mechanisms that function according to a logic of consequences (march & olsen, 1998, 949; risse & ropp, 2013, 14). the package approach used by the inter-constituency alliance has not yet been discussed in ir scholarship and can enhance our current knowledge of transnational advocacy. employing a broad human rights framework, the alliance managed to integrate seven different networks with seven corresponding principles. trade union networks, for instance, in the past often voiced concerns about jobs being at risk in the face of green economy transition and therefore demanded just transition processes. indigenous peoples, in contrast, prioritized the protection of natural resources and heritage sites over the creation of new jobs. this new form of cooperation across constituencies empowered civil society towards state governments: “[…] the fact that a trade unionist who is coming from illinois to a climate conference and suddenly hearing the indigenous peoples’ representative saying: ‘oh remember, indigenous peoples’ rights only make sense in the context of just transition for workers’. that is something that was incredibly empowering and motivating and feeling that no matter what the unfccc does, we have already reached a degree of solidarity that is unusual and is powerful.” (interview initiator inter-constituency alliance, 26 march 2020, via skype.) by repeating the entire message package consistently throughout the negotiations and as a united super-network, civil society gained strengths, changed decision-making in the context of the paris agreement and fostered institutional interaction between the human rights and the climate regime (package approach politics). figure two displays tan’s tactics in fostering institutional interaction between the human rights and the climate regime. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 42 university of bamberg press figure 2 institutional interaction between the human rights and the climate regime (source: own compilation) what is new – and why is this relevant for the study of institutional complexity? this case study demonstrates how transnational advocacy networks built and mobilized the inter-constituency alliance as a super-network with a view to incorporate various principles under the umbrella of a human rights framework into the new 2015 climate instrument. this leads to the creation of a new regime complex, in which the norms of one source regime (human rights) become relevant for a target regime (climate policy). the super-network described in the empirical case study is unique and different from keck and sikkink’s (1998) definition of tans as it works across a number of issue areas, including organizations concerned with intergenerational equity, land issues, environmental problems, the situation of indigenous peoples, workers’ concerns, gender aspects, and human rights. in contrast to keck and sikkink’s original work on tans, where networks center their work around either values, e.g. human rights, women’s rights, or environmental protection (1998), the super-network operates across constituencies and integrates a number of issues. although objectives and methods of the individual tans participating in the 2015 interconstituency alliance are usually quite diverse, and sometimes even contradictory, network members understood the context of the paris negotiations as an opportunity to advocate for common core principles they had agreed on. the unique mobilization structure of seven tans in the super-network meant that different networks could combine their strengths and expertise, exchange valuable information, agree on overlapping objectives and refine their tactics. then they could go back into their individual networks to lobby and interact with the respective state representatives they had built relationships with over many years. as the entire alliance spoke with one voice and promoted common rights principles, their influence was much stronger than in previous negotiations and changed the decision-making of states, in particular of those who were opposed to include rights in the agreement, like the usa, saudi arabia, or several african states (human rights watch, 2015). rights protection was highlighted as important by raising awareness to the adverse local impacts and unintended consequences that international governmental climate policies, like the cdm human rights regime transnational advocacy networks human rights agreements climate regime state parties climate agreements incl. human rights interaction through commitment information symbols/stories leverage accountability package approach unanimously adopt complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 43 university of bamberg press or redd+, can have. at the same time, the alliance made governmental negotiators who are usually only trained in their own field aware of already existing human rights obligations of the international community that respective countries had ratified. by doing this, and by making concrete suggestions for the negotiation text, including references to existing human rights commitments, the supernetwork fostered institutional interaction through commitment. by advocating for an incorporation of human rights in the preamble of the paris agreement, the alliance also advanced further institutionalization processes between the human rights and the climate change regime, creating a new regime complex. they urged states to take human rights, as a part of the new climate instrument, seriously in the following negotiations of the implementation guidelines, the so-called paris rulebook (ciel et al., 2017). with the postponed cop 26 in glasgow to be held only at the end of 2021, there will be delayed negotiations on the san jose principles, also known as article 6 principles, defining standards for the creation of carbon markets. after cop 25, human rights have not yet been included in article 6, but amendments are still to be negotiated and the alliance is trying to influence governments to incorporate human rights safeguards into these principles (hrccwg, 2018). developing a better understanding of these new forms of tans helps to explain the dynamics behind institutional interaction. this means there are not only causal mechanisms of interaction between a source and a target institution, such as interaction through commitment1, as elaborated bv gehring and oberthür (2006, 2009), but there are also politics (i.e. information, symbols/stories, accountability, leverage, and package approach) behind the interaction link. this adds to the study of institutional complexity as it contributes a multi-level perspective on adverse local rights impacts of climate policies and how they are used by transnational actors in the global climate arena to advocate for change. the unique interaction patterns between civil society has fostered institutional interaction between two different regimes. this brings a micro-perspective of actors (and local impacts) into play that is yet missing in many studies on institutional interaction and connects it with the macroperspective of structures (and regime overlap). it also links rational choice-oriented regime studies with social constructivist work on norm-guided actors, actor constellations, and networks. this study contributes to orsini’s (2013, 2015) work on “multi-forum” non-state actors and helps close a research gap by illuminating the role of non-state actors and transnational advocacy networks in the statefocused study of institutional interaction. the inter-constituency alliance investigated for this paper is not the only “super-network” active in international negotiations. other examples are the convention on biological diversity (cbd) alliance bringing together different tans active from various policy fields for common objectives in the cbd negotiations or the international campaign against nuclear weapons (ican) uniting youth groups, peace organizations, trade unions, women’s networks, health advocates, and others behind the common aim of eliminating nuclear weapons. their role in institutionalization processes and institutional interaction needs to be further investigated. critical discussion and conclusion in this paper, i have argued that we learn more about the dynamics of institutional interaction by understanding new forms of transnational advocacy networks that foster interaction (through commitment) and further institutionalization processes. empirically, this paper focuses on examining interactions between the human rights and the climate change regime, and the activities of a new form of alliance, a so-called “super-network”, for incorporating rights principles in the 2015 paris agreement. making use of the political context and opportunities to participate in the negotiations of a new climate instrument, various tans, including indigenous, environmental, youth, religious, trade union, land rights, and human rights groups, organized themselves in the inter-constituency alliance. combining strengths and expertise and sharing valuable information, the alliance agreed on a package 1 in this case study, we mainly learn about interaction through commitment (as the inter-constituency alliance refers to human rights agreements that have already been ratified by member stated of the unfccc). behavioral and impact-level interaction will only become relevant after the paris agreement is implemented. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 44 university of bamberg press approach with core principles to advocate for and influenced governmental decision-making. by making use of tactics, such as information politics, symbolic politics, leverage politics, and accountability politics, the inter-constituency alliance persuaded and pressured states to incorporate human rights in the 2015 paris agreement. this case demonstrates how local impacts and experiences play a role in institutional interaction and thus adds a multi-level actor perspective to the study of institutional complexity. it also contributes to understanding the role of non-state actors and actor constellations in influencing governmental decision-making to advance further institutionalization processes at the intersection of two regimes. two aspects concerning the formation of tans at the cop in general, and the interconstituency alliance in specific, have to be viewed critically. first, the hybridity of tans makes them dynamic and flexible but also comes with a major disadvantage. many csos are crucial in one negotiation but cannot participate in the next one. this means, in every new negotiation the respective network may miss out on expertise or relationships with state representatives that individual csos had established during previous years (interview representative of the hrccwg, cop 19 in warsaw, 17 november 2013). lacking continuity and consistency hamper the success of these alliances. second, for the formation of the inter-constituency alliance, tans have used the unique political context around the 2015 paris negotiations, in particular the opportunity to contribute to a new climate instrument, to advocate for commonly agreed principles. they were able to mobilize and successfully build a strong alliance among very diverse csos that usually pursue different, sometimes even competing, objectives. different network representatives emphasize that with the paris negotiations, unprecedented forms of civil society cooperation have been initiated that remain to be relevant in the future (interview representative of indigenous peoples’ caucus, saami council, 12 march 2020 via skype). but even though the different constituencies continue a dialogue, the supernetwork lost its momentum, unique mobilization structure, and unity behind a common aim after the paris negotiations. thus, the text for the implementation guidelines, which was subject to the following negotiations, has so far remained relatively weak from a human rights perspective – although negotiations on amendments are still ongoing (interview initiator inter-constituency alliance, 26 march 2020, via skype.) this demonstrates that strong mobilization and successful network-building to foster institutional interaction, as we could observe in the case of the inter-constituency alliance, only work under unique political opportunity structures. references abbott, k. (2014). strengthening the transnational regime complex for climate change. transnational environmental law, 3(1), 57-88. adelman, s. (2020). human rights in pursuit of climate justice. wisconsin international law journal (forthcoming). amnesty international & human rights watch (2015). human rights in climate pact under fire. retrieved june 17, 2016, from: https://www.hrw.org/news/2015/12/07/human-rights-climate-pact-under-fire. atapattu, s. & schapper, a. (2019). human rights and the environment: key issues. new york: routledge. bäckstrand, k. (2012). democracy and global environmental politics. in. p. dauvergne (ed.) (pp.507-519). handbook of global environmental politics. cheltenham: edward elgar. buzan, b., jones, c. & little, r. (1993). the logic of anarchy: from neorealism to structural realism. new york: columbia university press. carazo, m. (2017). contextual provisions (preamble and article 1). in. j. bulmer, m. carazo, m. doelle, a. higham, & d. klein (eds.) (pp. 107–122). the paris climate agreement. analysis and commentary. oxford: oxford university press. ciel et al. (2017). delivering on the paris promises: combating climate change while protecting rights. recommendations for the negotiations of the paris rule book. retrieved april 17, 2020, from: https://www.deliveringonparis.com/wp-content/uploads/2017/06/delivering-on-paris-web.pdf. climate case (2020). armando ferrão carvalho and others v. the european parliament and the council, “the people’s climate case”. retrieved august 25, 2020, from: http://climatecasechart.com/non-us-case/armando-ferrao-carvalho-andothers-v-the-european-parliament-and-the-council/. gehring, t. & oberthür, s. (2006). introduction. in. oberthür, s. & gehring, t. (eds.) (pp. 1-18). institutional interaction in global environmental governance: synergy and conflict among international and eu policies. cambridge: mit press. gehring, t. & oberthür, s. (2009). the causal mechanisms of interaction between international institutions. european journal of international relations, 15(1), 125–156. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 https://www.hrw.org/news/2015/12/07/human-rights-climate-pact-under-fire http://climatecasechart.com/non-us-case/armando-ferrao-carvalho-and-others-v-the-european-parliament-and-the-council/ http://climatecasechart.com/non-us-case/armando-ferrao-carvalho-and-others-v-the-european-parliament-and-the-council/ 45 university of bamberg press hrccwg (2018). briefing note: may 2018 bonn climate negotiations. retrieved august 26, 2020, from: https://www.ohchr.org/documents/issues/climatechange/bn_may2018_bonnclimatenegotiations.pdf. human rights watch. (2015). human rights in climate pact under fire. retrieved february 1, 2017, from: https://www.hrw.org/news/2015/12/07/human-rights-climate-pact-under-fire. keck, m., & sikkink, k. (1998). activists beyond borders. ithaca: cornell university press. keohane, r.o. (1989). international institutions and state power: essays in international relations theory. boulder: westview press. keohane, r. & victor, d. (2010). the regime complex for climate change. the harvard project on international climate agreements: discussion paper 10-33. knox, j. (2015). report of the independent expert on the issue of human rights obligations relating to the enjoyment of a safe, clean, healthy and sustainable environment, a/hrc/22/43. geneva: human rights council. kuchler, m. (2017). the human rights turn: engos’ changing tactics in the quest for a more transparent, participatory and accountable cdm. environmental politics, 26, 648–668. march, j., and olsen, j. (1998). the institutional dynamics of international political orders. international organization, 52:4, 943–969. mayring, p. (2014). qualitative content analysis: theoretical foundation, basic procedures and software solution. klagenfurt. retrieved december 17, 2019, from: http://nbn-resolving.de/urn:nbn:de:0168-ssoar-395173. oberthür, s. & gehring, t. eds. (2006). institutional interaction in global environmental governance: synergy and conflict among international and eu policies. cambridge: mit press. oberthür, s. & stokke, o. (eds.). (2011). managing institutional complexity: regime interplay and global environmental change. cambridge: mit press. oberthür, s. & pozarowska, j. (2013). managing institutional complexity and fragmentation: the nagoya protocol and the global governance of genetic resources. global environmental politics, 13, 100-118. ohchr. (2018). framework principles on human rights and the environment. retrieved december 5, 2019, from: https://www.ohchr.org/documents/issues/environment/srenvironment/frameworkprinciplesuserfriendlyv ersion.pdf. orsini, a. (2013). multi-forum non-state actors: navigating the regime complexes for forestry and genetic resources. global environmental politics, 13, 34-55. orsini, a. (2016). the negotiation burden of institutional interactions: non-state organizations and the international negotiations on forests. cambridge review of international affairs, 29, 1421–1440. raustiala, k. & victor, d. (2004). the regime complex for plant genetic resources. international organization, 58, 277-309. risse, t., ropp, s. & sikkink, k. (eds.). (1999). the power of human rights. international norms and domestic change. cambridge: cambridge university press. risse, t. & ropp, s. (2013). introduction and overview. in t. risse, s. ropp & k. sikkink (eds.) the persistent power of human rights: from commitment to compliance (pp. 3-25). cambridge: cambridge university press. schapper, a. & lederer, m. (2014). introduction. climate change and human rights: mapping institutional inter-linkages. cambridge review of international affairs, 27, 666–679. unfccc. (2010). the cancun agreements: outcome of the work of the ad hoc working group on long-term cooperative action under the convention. cancun: unfccc. unfccc. (2014). nongovernmental organization constituencies. retrieved june 24, 2016, from: https://unfccc.int/files/parties_and_observers/ngo/application/pdf/constituency_2011_english.pdf. unfccc (2015). paris agreement, adopted at the 21st conference of the parties (cop). 12 december 2015, paris. wegener, lennart (2020). can the paris agreement help climate litigation and vice versa? transnational environmental law, 9, 17-36. witzel, a. & reiter, h. (2012). the problem-centered interview. london: sage. young, o.r. (2002). the institutional dimensions of environmental change: fit, interplay, and scale. cambridge: mit press. zelli, f. & van asselt, h. (2013). the institutional fragmentation of global environmental governance: causes, consequences and responses. global environmental politics, 13, 1-13. zürn, m. & faude, b. (2013). on fragmentation, differentiation, and coordination. global environmental politics, 13, 119130. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, doi: http://dx.doi.org/10.20377/cgn-102 p. 32-45 governing complexity in world politics complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 55 university of bamberg press governing complexity in world politics authors: peter m. haas and jon western university of massachusetts, usa; mt. holyoke college, usa email: haas@polsci.umass.edu; jwestern@mtholyoke.edu complexity is the new global ontology for world politics. this article summarizes the characteristics of complexity and its implications for informed us state policy making. we conclude with some suggestions about administrative reforms to improve us policy making to address global complexity. keywords: hegemony, us leadership, institutional reform, institutional design, systemic complexity introduction the lack of coherent, organized, and effective international responses to climate change and more recently to the covid-19 pandemic reflect the growing challenge for addressing highly complex global problems. it is increasingly clear that we are living in a highly globalized world in which complex and inter-related dynamic problems quickly transcend national boundaries, overwhelm local, national, regional, and global institutional and political capacities. while history has long shown that international problems have always been complex, the analytic implication for world politics today, with the decline of us hegemony and the absence of concerted leadership with coherent problem-solving institutions and strategies, is that today’s problems have demonstrated to be increasingly illusive. it is long understood that globalization has been a generative force for complexity. international relations are today, differently complicated not just because of more actors, institutions and issues, but also because power is increasingly diffused throughout the system and the decision-makers tasked to address the problems and the institutions designed to manage them are outmoded and increasingly confounded by the impacts of decisions in one area that produce multiple and intersecting first, second, and thirdorder impacts on other areas they are trying to govern producing a whole range of technical, political, economic, and social challenges. (slaughter, 2020) problems defy simple, linear or issue-based solutions. systemic complexity generates a profound challenge: under conditions of complexity it is perhaps more necessary than ever to coordinate across issues, but the complexity, itself, coupled with the decline of american power and leadership, at a time when no other has taken the mantle, creates more challenges to organization and coordination in decision-making and effective problem-solving. the result is increasing failures to address global challenges in a coordinated, coherent, and effective manner. this article considers the ontological character of global complexity, its governance, and the policy and analytic implications of this shift amid the decline of american hegemony. we begin with a review of the nature and meaning of complexity followed by a discussion of the relationship to global governance and power and the historic significance of american global leadership. we conclude with some illustrative examples of this complexity and present a brief set of u.s. administrative and institutional strategies which can help enhance prospective national planning and global governance for a globalized world. mailto:haas@polsci.umass.edu complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 56 university of bamberg press complexity complexity became widely recognized in the 1960s. (orsini et al., 2019) cybernetics and open systems theory helped focus attention on intertwined problems, and the complex arrays of social forces by which they are recognized and addressed. (deutsch, 1963) the 1972 apollo 17 photograph of the earth as “the blue marble” helped visualize a single system made up of intertwined natural and social systems. global un conferences and global environmental governance had the effect of alerting decision makers to the urgency of addressing new environmental threats and how they are causally intertwined with other global issues, requiring more comprehensive governance. (p. haas, 2002; p. m. haas, 2016) the growing popularity of the concept of the anthropocene further consolidated attention on complex global open systems. (crutzen, 2003) this broad awareness and investigation has extensively been applied to the study of international relations which was conceived and understood as the systemic property of world politics. (e. b. haas, 1975; laporte, 1975; perrow, 1999) herbert simon pioneered much of the literature on complexity and social action and noted: “everything is connected. some things more than others.”(simon, 1981, p. ch 7) robert jervis’s landmark study on systems effects demonstrated the wide range of unanticipated and unintended consequences, and often perverse, localized effects of action filtered through systemic level variables. (jervis, 1997) we now understand that the challenge with many of today’s problems is that they are complex across multiple domains: time horizons vary both within and across issue areas; many variables interact in non-linear and often unsuspecting ways. (homer-dixon, 2015) the world is full of issues interacting with other problems creating a system of problems – or, similar to what russel ackoff simply called “messes” (ackoff, 1974) and the urban planners horst rittel and melvin webber called “wicked problems.” (rittel & webber, 1973) more recently these approaches have been absorbed under general attention to complexity and uncertainty in world politics. (urry, 2003; urry 2005) our current complex system of world politics rests on several elements that are continually evolving. first, and perhaps most significantly, the 20th century saw a geometric increase in the number of new actors in world politics, and in the volume of flows between countries. while post war settlements established uniform codes and institutions for governing international behavior, these institutional arrangements have not adequately responded to new actors, voices, and issues on the agenda, while underlying power shifts in specific issues have led to a dizzying quilt of rules, powerful parties, and practices across functional areas of world politics as influence within existing regimes has decayed. (raustiala and victor, 2004; alter and muenier, 2009; and, alter and raustiala, 2018) the growth of the number of states in the international system over the past six decades has been coupled with a dramatic rise in a whole new range of non-state and domestic actors co-acting with greater input into state-level policy, international interactions, and transnational events. democratic developments have fueled this growth. in the early 1970s, there were just over 30 democratic states in the international system. today, more than 120 have various forms of democratic institutional frameworks. (freedom house, 2020) and, within this states, we have seen a dramatic rise in domestic and transnational actors, including ngos that have grown geometrically since the beginning of the 20th century. (jonsson & tallberg, 2010) transnational scientific networks, including epistemic communities, have also flourished. (drori, meyer, ramirez, & schofer, 2003; p. haas, 1992) this dramatic proliferation of democratic institutions has opened the political space for the proliferation of a whole new generation of multinational business and financial enterprises, transnational monitoring and advocacy networks, civil society actors and networks, as well as transnational criminal actors – terrorist organizations, human traffickers, narcotics traders, and others. these actors have a broad complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 57 university of bamberg press range of information and communication tools to collect, analyze, and disseminate information horizontally across organizations and sectors. and they can communicate, transmit information, and often act with a speed and volume on par with state actors. the proliferation of legitimate actors engaged in diplomacy has led to the emergence of new voices, new issues and problems, and new methods and considerations to address the broad range of global problems. but this proliferation has also presented new challenges: which actors should be invited to participate in discussions on particular topics? who sets the agendas? what are the relative weights and dimensions of issue, interests, and information areas under discussion? what forums and forms of power and influence are legitimate and appropriate? with more actors intersecting with each other, issues also become more difficult to govern with the simultaneity of decisions and synchronicity of more forms of interdependence that correspond with the greater political, economic, and social involvement of non-state actors. second, the international agenda has also expanded dramatically since world war ii. more voices reflect more and varied concerns, many often longignored or long-dismissed. in addition to the traditional agenda items of trade, finance and security, a whole new set of issues with their own technical challenges, policy networks, and social, cultural, political, and economic contexts have emerged. issues today include public health; intellectual property rights; the environment; migration; human rights and human security; transnational crime; nuclear proliferation and weapons of mass destruction; telecommunications; management of the internet, cybercrime and cyber security, and energy security. a comparison of un general assembly agendas from 1970 to 2020 clearly demonstrates this burgeoning agenda, covering climate change, demographic stress, environmental degradation, resource scarcity, pervasive inequality, the host of challenges with international trade and financial regulatory systems, as well as expanding security concerns associated with failed states, proliferation, terrorism, migration, energy security, global public health, and a variety of forms of trafficking can all produce rapid and significant transnational effects and externalities across other issue areas requiring significant levels of coordinated and integrated international responses. third, related to the rise in actors and the number of issues is that many of the issues and problems are causally coupled, due to the long causal-chains and complex political, social, economic, and cultural tentacles connecting within and between issues. (beck and willms, 2003, and urry, 2003) as a result, there are frequently unanticipated consequences and surprises from policy choices and practices in general, policy choices involve trade-offs and choosing equilibria, and there are unavoidable normative choices involved in choosing policies. moreover, complex systems express emergent properties that cannot be understood or anticipated from a reductionist focus on the component parts, but rather requires a systemic and interconnected level (or scale) or analysis. for nearly a decade, government and private studies of contemporary global trends have increasingly reflected these new complexities on the international agenda. the 2019 worldwide threat assessment of the us intelligence community describes a broad range to traditional challenges – competition from and between russia and china – intersecting with new forms of threat ranging from on-line election interference, to hacktivist activities, to human and traditional security impacts of migration, rising nationalisms and xenophobia, and threats associated with potential pandemics and crises in global public health. (coats, 2019; see also, international risk governance council, 2013; uk ministry of defense, 2014; us national intelligence council, 2012; world economic forum, 2016) along with the recognition of the increased number of actors, new issues, and causal coupling, we have also observed the profound challenge with addressing them. the proliferation of actors means that there are multiple and interacting parties with conflicting interpretations of data and differing goals, analyses, and prescriptions. these actors have widely varying capabilities across and within issues, with complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 58 university of bamberg press which they undertake action. as a result, it is often difficult, if not impossible, to identify the key actors or the range of influential actors on a given issue. furthermore, there is extensive uncertainty on multiple paths because they are causally interconnected. uncertainty has long been a defining feature of global complexity: actors cannot assign confident probabilities to the likely outcomes of policy interventions, or the likelihood of any occurring at all. (hayek, 1945; iida, 1993; knight, 1921) moreover, efforts to govern specific issues in either domain end up with unanticipated consequences for other interconnected issues. today, the uncertainty is simply more prevalent across so many actors and intersecting issue areas. uncertainty may stem from lack of data, biased data, or ambiguous data – or some combination of all three simultaneously. in a related manner, without a clear framework to impart meaning, too much data also swamps decision making capacity. it also may be, as james march wrote in 1976, that many problems face “intrinsic uncertainty” where “[t]he hard reality that the world in which we must act is often beyond our understanding. in part because, each action that we take is partly an instrumental step and partly a learning experience.” (axelrod & cohen, 1999) typically, new issues have cumulative effects, are beset with uncertainty, and have interlocking moving parts. consequently, it is difficult to anticipate policy consequences. the components to each issue requires multidisciplinary understanding and the different issues may each require its own disciplinary lenses. because discrete problems have long causal chains and are causally interconnected with other issues, governance is challenging. in addition, today’s challenges often require much more extensive and specialized technical and scientific knowledge to understand, diagnose, and analyze them. for example, climate change and proliferation – to name just two – require highly specialized communities of experts to collect, analyze, and interpret that nature and meaning of the challenges. these challenges do not fit easily into the alltoo-frequent superficial public debate that privileges entertainment over public enlightenment. finally, today’s challenges must be addressed in a much more dynamic and fluid political environment. domestic and international politics around the globe increasingly have become more contested – and dysfunctional – especially in an environment of 24-hour news cycles that have changed the speed, content, and depth of information and analysis presented through the media and challenged basic conceptions of scientific understanding and reasoning in profound ways. this has been particularly true in the united states over the past decade, and more specifically, in the past four years under president trump where there has been a proliferation of splintered voices in american politics, hyper-partisanship, open challenge, disdain and mockery of scientific evidence and reasoning and of administrative and bureaucratic procedures to analyze and respond to complex issues. add to that the known and unknown influences of external actors infiltrating american political discourse and amplifying discord through social media, hacking, and other forms of cyberattacks. the result is that many of today’s problems are connected in diverse and convoluted ways. some issues have easily identifiable first and second order effects – protracted ethno-national conflicts produce migration, disrupt economies, and produce downstream temporal and spatial effects for the country and the region. but, add acute stress in global public health, climate change, chronic inequality, resource competition and the range of externalities found in the initial conditions of each and the connections grow exponentially. we can imagine a variety of difficult effects to understand on causally connected issues from taking action or avoiding action (inaction). in short, while we may be sensitive to the concept of system effects and emergent properties, under our current institutional and organizational designs, we simply often do not have the diagnostic or analytical capabilities to detect or understand them. as the ordered effects grow more complicated the unanticipated/anticipated distinction frays. 59 university of bamberg press complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 limits of rational problem-solving approaches the cumulative effect of this complexity is that today’s problems transgress the parameters of traditional public planning that focused primarily on efficiency and resolution. for more than five decades, public planners have sought to identify the core logics of problems -to find “where in the complex causal network the trouble really lies.” (rittel & webber, 1973, p. 159) the standard task has been to understand the interplay between cause and effect and to identify and isolate the variables to manipulate in order to produce optimal policy responses to problem. and while, we have long realized that most problems defy easy diagnosis analyses and we frequently witness cumbersome and convoluted responses, the effort has nonetheless been to search for optimal policy responses within a set of given constraints. the emergence of many new issues on the international agenda offers more technical, organizational, and political challenges for policy makers. the increasing degree of causal interdependencies between issues generates a newfound need for more comprehensive and integrated policy making. these problems – and the relationships between them -are increasingly more difficult to identify, to diagnose and to mobilize the political, institutional, and economic capabilities to manage them the traditional state-based power alignments and governance structures – both internally inside of states and transnationally across states -are no longer reliable guides to managing these global. (slaughter, 2020) first, there is significant variation in governance both within and across issue areas. different states and actors contribute in new and different ways to each of these problems. some issues and challenges are more salient to specific actors within countries and across countries than others. because solutions cannot be obtained with the support and coordination of multiple states and actors, it is not simply a matter of gaining better diagnostic capabilities, but also the ability to understand the goals and incentives of the multiple-interacting parts of the problems. whereas the still high levels of international interdependence command some form of multilateralism to ensure mutual gains from international commerce and to mitigate their externalities, the foundations for such order necessitate more scrutiny than commonly made. global governance, however desirable, cannot simply be attained through traditional forms of state-based governance focusing on de-coupled issue areas. the proliferation of political actors and political issue spaces makes international cooperation logically more difficult but also offers the promise of more diverse involvement in collective decision making by more stake holders and perspectives, (ashby, 1960) including such networked approaches as polycentric governance, private governance, expert led governance, and orchestration. american hegemony and global governance global governance since world war ii has rested principally on an american-led liberal world order. (ikenberry, 2001; mazower, 2012) this combined widespread ideational commitment to liberal norms of human rights, the rule of law, free trade, and multilateralism. it was steered by the united states, which generally espoused these beliefs and deployed its power resources – material capabilities and soft poweron promoting and enforcing those norms, albeit selectively in many instances. while the us was often viewed, justifiably, with suspicion, the global governance institutions it helped set in place and supported contributed to the development of institutions, policies, and broader governance that improved life expectancy, reductions in infant mortality, access to potable water, nuclear non-proliferation, selective environmental improvements, economic growth outside the west, macroeconomic stability, inflows of foreign investment, and the absence of open, great power interstate wars. despite benefits in many areas, global governance however, was, and remains disjointed. regimes were originally designed and developed to deal with isolated problems. each regime was distinct, complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 60 university of bamberg press reflecting different arrays of actors, beliefs, capabilities, and procedures. the world health organization, for example, focused on international disease classification and focused its early efforts on mitigating or eradicating specific diseases – yaws, small pox, polio. today, the who’s work now focuses, not on single diseases and direct health delivery services, but on broader issues such as chronic disease, emergency preparedness, poverty, environmental stress, and access to public health. the who regime, like many others had, and has, different emergent political forces and now fall under a category that alter and raustiala call “regime complexes,” in which issue areas are deeply connected across multiple dimensions and institutions. whereas issues have traditionally been addressed discretely through the creation of distinct international regimes, these issues now refuse to stay in their boxes. the associated bureaucracies and even regime complexes lack the frames to capture the spill over between issues and inhibit their ability to coordinate and govern across the issue areas. (alter and raustiala, 2018) careful leadership focused around new multilateral approaches are necessary to capture the interconnections between issues. yet, the decline of us hegemony has demonstrated that there is no clear state leader able to direct world affairs. furthermore, emerging global and regional powers have less interest and/or investment in the existing order. hence, the diminished consensus around the core elements of the liberal world order means that it is more difficult to guide collective around clear and shared goals. today we are experiencing twin challenges associated with confronting these global complexities – not only the decline of hegemony and the loss of its political capacity and legitimacy to lead, but also the loss of the us’s own internal ability to diagnose and address the complexities of problems internal to the united states and those that are transnational. the models for problem solving will require change both within the united states and within the global governance structures. it is well established that while the united states boasted (and continues to do so) a position of military primacy, it has lost its hegemonic position in a material sense. (keohane, 1984; strange, mahbubani.) the us remains a “normal” great power, with its vast military capabilities, but, especially under the past four years of the trump administration, it is no longer capable or willing to lead the international community as it had through most of the cold war and early post-cold war era. power in the contemporary system is a multiplex concept. (barnett & duvall, 2005; hollis & lukes, 1982; lukes, 1974) it entails coercive power via the use of material capabilities to induce others’ behavior, (baldwin, 1985) framing power via shaping others beliefs via a persuasive presentation of options, (adler & haas, 1992/1997) leadership on behalf of collective beliefs and shared goals; (j. nye, 1990) and productive power to regenerate material and ideational capabilities. us power is diminished in all these aspects – and all of these declines have an impact on the existing global governance structures to effectively address and manage the complexity in the system today. as significant to its relative material decline, the united states has lost its legitimacy as the stalwart promoter of shared liberal values, such as free trade and democracy. even before the cataclysm of the trump administration, faith in democracy and free trade as universal norms have been increasingly contested by the emerging powers. (acharya, 2014; mahbubani, 2008) although seventy-four percent of countries are full or partial democracies (according to minimalist criterion for democracy), only 40% of the world’s population lives in a democracy and we have witnessed 14 years of democratic decline since 2005. (freedom house, 2014, pp. 3-5; freedom house, 2020) few substantive norms currently exist around which the us can seek to lead or command voluntary support. individual norms such as free trade, democracy, human rights, the prohibition of mercenaries, and even the protection of whales are contested. the only major procedural norm that seems to be holding is the broad commitment to multilateralism, (pew, 2020 https://www.pewresearch.org/global/2020/09/21/international-cooperation-welcomed-across-14complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 61 university of bamberg press advanced-economies/; better world campaign, 2020 https://betterworldcampaign.org/wpcontent/uploads/2020/09/executive-summary-september-u.s.-un-poll.pdfnewman, 2006; ruggie, 1993) as states continue to realize the need for collective action to provide international public goods and to cope with transboundary and global risks. on the other hand, no other world order aspirants appear to have collective revolutionary values which would undermine global multilateralism as well as the attendant technical efforts to address complexities of global problems. the rise of china, india, the eu, brazil, south africa and russia collectively provide counterweights to us primacy and leadership, yet, none of these countries have yet developed coherent global visions of how to deploy their newfound influence. a central challenge is that the united states has lost many of its domestic material and political capabilities for exercising global leadership and its ability to acknowledge, diagnose, and respond to the levels of complexity. (goldgeier and jentleson, 2020, haass, 2013; mandelbaum & friedman, 2011; nasr, 2013; parent & macdonald, 2011) the united states which held 51% of global gdp in 1951 is now at 15% with chronic budget and trade deficits. nearly two decades ago, at the outset of the us unilateral decision to invade iraq, the central question appeared to be how american power and primacy could be tamed and contained.(brooks & wolforth, 2007; hoffmann, 2004; walt, 2004) yet, a decade later the wars in iraq and afghanistan coupled with the lingering effects of the 2008 global financial crisis exacerbated a breakdown of the us domestic social contract over american international leadership. the cost of the wars alone are estimated to be in excess of $6 trillion with little enhanced security to show for it. the overall return on international leadership investment was generally seen to be the provision of secure borders and relatively stable international security environment in which american interests were generally safe – including a stable and functioning global economic system that provided opportunities for foreign investment and markets (employment) and imports for american consumers (and a hedge against inflation). despite spending nearly $3 trillion over the past decade to restore some sense of global order, ongoing challenges demonstrate the limits of american power from climate change to covid-19 to unchallenged russian interference in democracies in the united states and across europe, china’s, rising nationalisms and nativism, and the chronic global economic malaise have all undermined american confidence in its leadership role. while a majority of americans still believe that the united states should take an active role in global affairs, the largest percentage ever – including a majority of rank and file self-identified republicans – believe that the united states is doing too much and should “stay out” of international meddling (chicago council on foreign affairs, 2020). this domestic fragmentation has also reduced the public’s commitment and interest in international commitments and leadership of global institutions. while a majority of americans continue to express interest in the united states being engaged in world events, under the trump administration over the past four years, we have witnessed a rise in nativism and white supremacy that includes various forms of neo-isolationism and calls for international retrenchment that has found a home in both the right. the partisan differences over immigration are stark with 81% of republicans expressing support for the use of us troops to block immigration into the us from mexico as opposed to just 19% of democrats (ccfr: report 2019) while 78% of democrats see climate change as a critical threat as opposed to only 23% of republicans. and, finally, the threat to basic conceptions of scientific evidence and reasoning and overall to a shared understanding of america’s role and responsibilities in both give further rise to the failure to acknowledge and address complexity. improving global governance begins with changes inside the united states. without common set of accepted norms and principles or conceptual guidance about which issues are intertwined and how to address them, policy making continues to rest on outmoded reductionist complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 62 university of bamberg press approaches to both at home and abroad. within the united states, the bureaucratic planning and strategy structures are functionally differentiated, and thus lack the capability for horizontal planning across issues. the congress as well as the public remains disinterested in complexity and traditional habits of thought in the government continue to stress a small number of key issues, defined narrowly, relating to national economic growth and the defense of us national security from terrorism and nuclear attack. for instance, while a host of recent policy briefs have attempted to chart an american foreign policy for a new world in which the us lacks overarching capabilities and faces contested norms these approaches do not take account of the fundamental change in the context in which us understands its interests and pursues its foreign policies. (us quadrennial diplomacy and development review, 2015, summary of the 2018 national defense strategy of the united states, 2018 ) historically, us strategic planning has been flawed, often because it projects a unifying framework onto a world that is much more nuanced. (goldgeier & suri, 2016) this traditional planning does not work for today’s challenges: in short, the issues are connected but governance structures, internally in the united states and externally in global governance institutions, are not. this means that in addition to changes within the us foreign policymaking structures, the foreign policy establishment needs to establish stronger links and work more closely with relevant policy communities across and outside of government. the united states has long endured problems of group think and bureaucratic politics – all of which will exacerbate the dimensions of the current challenges. given the type of complexities, the planning and policy system needs greater participation of state and non-state actors drawn from diverse policy networks and institutions and interdisciplinary teams that are capable of identifying externalities. within this context, we need to design new strategies and institutions capable of developing outcomes that are normatively and technically warranted. institutions must be adaptable to complex circumstances. administratively organizations are too often siloed – unable to see the forest for the trees at the international and within the us decision-making apparatus. currently most us and international institutions (including ios and regimes) remain blinkered by their traditional narrow functional focus, combined with policy networks that are primarily experienced with individual issues and often narrowly disciplinarily trained. organizational theory and experiences make it clear that complex issues are understood in light of the existing issue frames of the existing bodies. to this end, we need better monitoring of the policy environment in a variety of areas that are more reflective to new information and can explore the interactive effects. effective institutions capable of tracking and mapping a complex policy environment must enjoy autonomy, access to high levels of decision making, a high quality staff with relevant skill sets, a steady budget, and with porous organizational boundaries so that it can reach outside the formal bureaucracy for new ideas and relevant advice. (e. b. haas, 1990; p. haas & haas, 1995) while the us government has taken strides in recognizing the urgency and magnitude of climate change threats to national security, the organizational foundations of the assessments have seriously biased the nature of the assessments. even though the pentagon has been active in integrating climate change into its planning, the issue is framed in light of traditional dod responsibilities. thus, the multiple climate change documents from dod emphasize tactical gains from reducing supply chains and costs through the adoption of solar power in the field, cleaning up military bases, preparing for operational challenges associated with climate change, and focusing on climate change effects with impacts on us geopolitical assets. it does not address the sustainability, human security, or environmental implications of climate change for economic sustainability or human quality of life. (the cna corporation, 2007; "usa national security strategy," 2015) there have been a number of efforts to improve comprehensive planning and to infuse the process with specialized expertise. senator henry m. “scoop” jackson called for organizational redesign of the complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 63 university of bamberg press foreign policy establishment to better anticipate and respond to complex challenges. (jackson, 1960) the commission on the organization of the government for the conduct of foreign policy released a sevenvolume study in 1975, although it ultimately failed to yield any administrative reforms. the national security council and homeland security are efforts to provide multiple perspectives on security threats. the nsc may be too small and rely too extensively on military secondments to achieve as broad and independent a view as would be liked. the white house national economic council is an effort to provide a broad perspective on economic policy issues. anticipatory governance, now an eight-year-old effort, was designed to provide the procedural and institutional parameters for dealing with complex foreign policy issues. (l. fuerth; l. s. fuerth & faber, 2012) an early example of integrated governance was the formation of the office of technology assessment (ota) (1972-1995) which was designed to craft policy analyses for congress of emerging technologies. it was widely respected for producing high quality studies produced by a strong professional interdisciplinary staff that was independent from congress. ultimately it was eliminated due to cost cutting and because congress felt that many of its reports were submitted too late to be used effectively in drafting legislation. (bimber, 1996; brown) a number of more contemporary examples come from efforts to deal effectively with environmental issues. to a large extent the united states, as with other major countries relies on two institutions to develop insightful environmental policy. one body – the council on environmental quality (ceq) coordinates interagency work on environmental issues. another body – the environmental protection agency (epa) devises and enforces environmental regulations. other countries have experimented with various administrative combinations in order to achieve broad policy coordination amongst functional agencies combined with efforts to devise and enforce environmental policies. (p. m. haas, 1990, p. ch 5) president clinton created environmental hubs in state department embassies intended to impart environmental analyses into all issues which the embassy would encounter. due to lack of funding the hubs were unable to recruit sufficient staff in order to perform their tasks. similarly, other efforts on integrated and networked policy development were inspired by the foreign policy failures or mistakes since the end of the cold war. the national counterterrorism center (nctc) was created in 2004 to integrate and analyze all us intelligence on terrorism. its mission was to ensure adequate sharing of information across all us departments and agencies and ensure the “integration of all instruments of national power.” despite is ambitious structure, however, throughout much of its first decade the nctc was hamstrung by serious staffing flaws, persistent cultural conflicts among participating departments and agencies, and limited authority structures.(schmitt & shanker, 2010; "toward integrating complex national missions," 2010) today, it continues to struggle with the same problems. (author interview with senior ntct official, west point, april 24, 2016) president obama’s atrocities prevention board (apb) established in 2012 was another example of an effort to develop integrated policy coordination. (alleblas, aloyo, et al, 2017) the apb was born out of the recognition of the need to integrate functional and regional considerations into the formulation of coherent policy. the apb used a “whole of government approach” that was designed to tap existing institutional and bureaucratic capacities to prioritize surveying the whole environment, develop structures, gather intelligence, and coordinate analysis to identify and understand inter-linkages. the challenge, however, is that the bureaucracy has been slow to embrace this institution and the new practices. the approach illustrated by the apb recognized and emphasized the importance of interagency involvement to assure that stakeholders from the operational agencies contribute to a comprehensive approach to studying complex issues and implementing policies on the ground. yet, as evidenced by the experiences of the apb and other similar institutional reform efforts, training and socializing bureaucratic complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 64 university of bamberg press actors is difficult. furthermore, the trump administration’s national security approach has been to reduce and weaken interagency exchange and coordination in favor of more centralized, top-down executive directives. (bolton, 2020) the road ahead we need better instruments to improve bureaucratic and organizational buy-in into the practices of breaking down problems to their appropriate scale – to identify the first and second order effects -and, to promoting outcomes that reflect fluid conceptions of national interests in a complex environment. underlying all these efforts lies a catch-22, however: how to mobilize a network of diverse expert competencies and perspectives within the entrenched bureaucratic politics and organizational cultures of washington decision-making structures – particularly in an era of acute partisanship. optimally, it would be preferential to rely on existing organizational structures to mobilize these diverse expert competencies. the national security council staff originally was envisioned to serve as the conduit of planning and coordination. yet, although its staff has grown nearly eight-fold in the past thirty years, it too remains highly siloed and increasingly politicized (bolton, 2020). most us government agencies have developed more sophisticated forecasting, planning, and coordinating mandates. yet they remain hamstrung by their narrow organizational charges and strong bureaucratic cultures. without some formal structure and dedicated authorities and budgets, integrated whole-of-government approaches too often remain stifled by existing bureaucratic rivalries, contested staffing and missions, and limited or non-existent budgetary authority. as a result, us decision making could be improved by the creation of two new bodies: an interagency council responsible for complex global issues supported by a small stand-alone executive unit responsible for collecting and aggregating data, conducting policy analysis, and formulating alternative policy proposals. such a body would have a standing professional staff drawn from multiple disciplinary backgrounds with a regular budget appropriation with flexible authorities to directly program or reprogram a set pool of funds as needed. a narrative is necessary to justify the creation of such a new body. strobe talbott lamented the difficulty of selling “interdependence” and issue-linkages to a skeptical congress as a framing concept behind administrative reform or policy. the standard tropes of american foreign policy: liberal multilateralism, conservative unilateralism, and a conservative policy agenda lack the ability to meaningfully engage with the challenges of complexity. (kissinger, 2001; mead, 2001; nau, 2013; ruggie, 1996) the american people need not be afraid of uncertainty and the unknown. american technological prowess has been able to recognize and persevere over complex challenges: from the apollo space program to the shuttle to the internet and successfully networked energy systems. complexity need not be feared; it can be something that is recognized as a challenge and overcome through reasoned analysis. by preparing for the next crisis and then presenting it as an instance of complexity to be governed, the warrants will be laid for appropriate institutional reforms. there is a growing sense in washington and the country that us actions generate surprises and unanticipated outcomes. they may even cause backlash and instill the very conditions we are seeking to avoid. institutional reforms could be justified by recognizing that we are now in a new complex environment which requires smarter planning. the trump administration’s response and management of the covid-19 pandemic has, perhaps more than other issues caused people to realize that they are affected by long causal chains with origins elsewhere in the world and in issues which may be disconnected from their immediate experiences. (pew research center, 2020) presenting a more sophisticated map of the policy world would support institutional reforms able to capture the intricacies complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 65 university of bamberg press of that map. as captain barbossa said in pirates of the caribbean: “you’re off the edge of the map, mate. here there be monsters.” we may be encountering monsters because we have been looking at the world wrong and crafting inaccurate maps. such institutional reform will contribute to more effective and accurate american foreign policy. in addition, policy emerging from such procedures will enhance us soft power by virtue of being based on a carefully reasoned understanding of global challenges. the un system also can benefit from institutional reforms to improve coordinated management of global complex issues. there are no centralized bodies able to coordinate between the multitude of specialized agencies. to take one example, the wto environmental dispute settlement procedures are insufficient to balance trade and environmental concerns. trade and development are not adequately balanced within the wto – especially in the face of the failed doha round and lack of communication between undp, wb, and wto. correspondingly, those institutions responsible for the environment– unep and various regimes, for example – lack the resources, mandate, and power to inform policymaking in trade, finance, and other areas. similarly, human rights and security remain narrowly defined within the purviews of narrow policy networks. the sustainable development goals which serve at the core of the international development agenda through 2030 highlight the need for an administrative body to help capture the interconnections between the 17 diffuse goals, and to promote international compliance. the un secretariat needs a strengthened unit which has more disciplinary skills and competency than the current undesa in order to collect and appraise progress in achieving the sdgs. the un also needs a new expert sustainability panel capable of understanding the interconnections between the sdgs, to identify gaps in knowledge, to mobilize the networks of expertise about and between the sdgs, and to collect best practices for their attainment. conclusion new circumstances require new approaches. in this article we have suggested that the signature characteristic of the contemporary world order is its complexity. we do not make the claim that the world is more complex, or even more dangerous, than previous eras, only that the complexities are profoundly different from previous eras. the multitude of actors, issues, and the complex interactions between such issues have created a patchwork international order, rife with uncertainties about the relevant policy networks, the specific us interest, and the available assets to promote us interests. such changes require a better and broader recognition of this new world order, as well as new domestic and international institutions that are capable of parsing the uncertainties in order to generate policies that are more likely to resonate with the us national interest and others’ interests, as well as being technically feasible. institutional reforms can help the us formulate and project more effective policies that can better respond to these changed conditions. it will enable more effective us leadership and engagement through more accurate understandings of the international policy environment, and by making us decision making more transparent and legitimate in the eyes of others. complexity is surely likely to persist as a systemic property. while global governance may face a rocky spell in the foreseeable future, governments still need new lenses and approaches to collectively deal with complexity. more comprehensive decision-making processes at the national and international levels can contribute to more effective governance, and the avoidance of rude surprises. 66 university of bamberg press complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 references acharya, a. (2014). the end of american world order. cambridge: polity. ackoff, r. (1974). redesigning the future. new york: wiley. adler, e., & haas, p. m. (1992/1997). conclusion. in p. m. haas (ed.), knowledge power and international policy coordination. columbia: university of south carolina press. alleblas, tessa, $ aloyo, eamon, et.al. (2017) in the shadow of syria: assessing the obama administration's efforts on mass atrocity prevention. hague institute for global justice. the hague. alter, k., & meunier, s. (2009). the politics of international regime complexity. perspectives on politics. cambridge university press. 7:1. 13-24. alter, k., & raustiala, k. (2018). the rise of international regime complexity. annual review of law and social science. 14. 329349. ashby, w. r. (1960). design for a brain (vol. chapman and hall): london. axelrod, r., & cohen, m. d. (1999). harnessing complexity. new york: the free press. baldwin, d. a. (1985). economic statecraft. princeton: princeton university press. barnett, m., & duvall, r. (eds.). (2005). power in global governance. cambridge: cambridge university press. biersteker, t. j. (2000). globalization as a mode of thinking in major institutional actors. in n. woods (ed.), the political economy of globalization. new york: st. martin's press. bimber, b. (1996). the politics of expertise in congress. albany: state university of new york press. bolton, john. (2020). the room where it happened: a white house memoir. simon and schuster. new york. brooks, s. g., & wolforth, w. c. (2007). world out of balance. princeton: princeton. brown, g. environment. castells, m. (2000a). end of millennium (2 ed. vol. 3). malden, ma: blackwell publishing. castells, m. (2000b). the power of identity (2 ed. vol. 2). malden, ma: blackwell publishing. castells, m. (2000c). the rise of the network society (2 ed. vol. 1). malden, ma: blackwell publishing. chicago council on foreign affairs. (2014). foreign policy in an age of retrenchment. retrieved from chicago: coats, daniel. (2019). statement for the record: worldwide threat assessment of the us intelligence community. senate select committee on intelligence. 29 january 2019. cohen, b. j., & benney, t. m. (2014). what does the international currency system really look like. review of international political economy, 21(5). crutzen, p. (2003). how long have we been in the anthropocene. climatic change, 61, 251-257. deutsch, k. w. (1963). the nerves of government: models of political communication and control. new york: the free press. drori, g. s., meyer, j. w., ramirez, f., o.; , & schofer, e. (eds.). (2003). science in the modern world polity: institutionalization and globalization. stanford: stanford university press. freedom house. (2014). freedom in the world 2014. washington, dc: freedom house. freedom house. (2020). freedom in the world 2020: a leaderless struggle for democracy. washington, dc: freedom house. frieden, j. a. (2006). global capitalism. new york: norton. fuerth, l. operationalizing anticipatory governance. prism 2, 4. fuerth, l. s., & faber, e. m. h. (2012). anticipatory governance. (october). giddens, a. (1991). the consequences of modernity. cambridge: polity. giddens, a. (2000). runaway world. london: routledge. goldgeier, j., & suri, j. (2016). revitalizing the us national security strategy. the washington quarterly(winter). haas, e. b. (1975). is there a whole in the hole? international organization, 29(3), 827-876. haas, e. b. (1990). when knowledge is power. berkley: university of california press. haas, p. (1992). introduction: epistemic communities and international policy coordination. international organization, 46(1), 1-37. haas, p. (2002). un conferences and constructivist governance of the environment. global governance, 8(1), 73-91. haas, p., & haas, e. b. (1995). learning to learn: improving global governance. global governance, 1(3), 255-284. haas, p. m. (1990). saving the mediterranean. new york: columbia university press. haas, p. m. (2016). epistemic communities, constructivism and international environmental politics. london: routledge. haass, r. (2013). foreign policy begins at home. new york: basic books. hayek, f. a. (1945). the use of knowledge in society. the american economic review, 35(4). held, d., & mcgrew, a. (1999). globalization. global governance, 5(4), 483-496. held, d., mcgrew, a., goldblatt, d., & perraton, j. (1999). global transformations. stanford, ca: stanford university press. hoffmann, s. (2004). gulliver unbound. lanham, md: rowman and littlefield. hollis, m., & lukes, s. (eds.). (1982). rationality and relativism. cambridge: mit press. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 55-67 doi: http://dx.doi.org/10.20377/cgn-104 67 university of bamberg press homer-dixon, t. (2015). synchronous failure. ecology and society, 20(3). iida, k. (1993). analytic uncertainty and international cooperation. international studies quarterly, 37(4), 431-457. ikenberry, g. j. (2001). after victory: institutions, strategic restraint, and the rebuilding of order after major wars. princeton: princeton university press. international risk governance council. (2013). preparing for future catastrophes. lausanne. jackson, h. m. (1960). organizing for survival. foreign affairs(april). jervis, r. (1997). system effects. princeton: princeton university press. jonsson, c., & tallberg, j. (eds.). (2010). transnational actors in global governance. new york: palgrave macmillan. keohane, r. o. (1984). after hegemony: cooperation and discord in the world political economy. princeton, nj: princeton university press. kissinger, h. a. (2001). does america need a foreign policy? new york: simon & schuster. knight, f. (1921). risk, uncertainty and profit. chicago: university of chicago press. laporte, t. r. (ed.) (1975). organized social complexity. princeton: princeton university press. lukes, s. (1974). power. new york: nyu press. mahbubani, k. (2008). the new asian hemisphere. new york: publicaffairs. mandelbaum, m., & friedman, t. l. (2011). that used to be us. new york: farrar, straus and giroux. mazower, m. (2012). governing the world: the history of an idea. new york: penguin. mead, w. r. (2001). special providence: american foreign policy and how it changed the world. new york: alfred a. knopf. nasr, v. (2013). the dispensable nation. new york: doubleday. nau, h. (2013). conservative internationalism. princeton: princeton university press. newman, e. (2006). multilateral crises in historical perspective. in e. newman, r. thakur, & j. tirman (eds.), multilateralism under challenge (pp. 160-178). tokyo: united nations university press. nye, j. (1990). bound to lead. new york: basic books nye, j. s. (2011). nuclear lessons for cyber security. strategic studies quarterly(winter). orsini, a., le prestre, p., haas, p. m., brosig, m., pattberg, p., widerberg, o., . . . chandler, d. (2019). forum: complex systems and international governance. international studies review. doi:10.1093/isr/viz005 parent, j. m., & macdonald, p. k. (2011). the wisdom of retrenchment. foreign affairs(november/december), 30-47. perrow, c. (1999). normal accidents. princeton: princeton university press. pew research center. (2020). international cooperation welcomed across 14 advanced economies. https://www.pewresearch.org/global/2020/09/21/international-cooperation-welcomed-across-14-advancedeconomies/. accessed 10-1-2020. raustiala, k. & victor, d. the regime complex for plant genetic resources. international organization. cambridge university press. 58:2. 277-309. rittel, h. w. j., & webber, m. m. (1973). dilemmas in a general theory of planning. policy sciences, 4, 155-169. robertson, r. (1992). globalization. london: sage. ruggie, j. g. (1996). winning the peace: america and world order in the new era. new york: columbia university press. ruggie, j. g. (ed.) (1993). multilateralism matters. new york: columbia university press. schmitt, e., & shanker, t. (2010, 22 february). hurdles hinder counterterrorism center. new york times. scholte, j. a. (2000). globalization: a critical introduction. new york: st. martin's press. simon, h. (1981). the sciences of the artificial (2nd ed.). cambridge: mit press. slaughter, anne-marie. (2020). reinventing the state department. democracy. https://democracyjournal.org/magazine/reinventing-the-state-department/. accessed 10/01/2020. washington, dc. strange, s. (1996). the retreat of the state. cambridge: cambridge university press. the cna corporation. (2007). national secuirty and the threat of climate change. toward integrating complex national missions. (2010, february). new york times. urry, j. (2003). global complexity. polity press. london. urry, j. (2005). the complexity turn. theory, culture and society. sage. london. 22:5, 1-14. uk ministry of defense. (2014). global strategic trends. london. us national intelligence council. (2012). global trends 2030. washington, dc. us quadrennial diplomacy and development review. (2015). enduring leadership in a dynamic world. washington, dc. usa national security strategy. (2015). walt, s. m. (2004). taming american power. new york: w.w. norton. weinstein, m. m. (ed.) (2005). globalization: what's new? new york: columbia university press. world economic forum. (2016). the global risks report 2016 11th edition. zurich: wef. shapeshifting to address complexity: advancing a typology of network evolution and transformation 25 complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 university of bamberg press shapeshifting to address complexity: advancing a typology of network evolution and transformation author: kate albrecht, ph.d. university of illinois-chicago, united states email: kalbrech@uic.edu in practice, health and social services are delivered through purpose-oriented networks (pons) that are often favored by government and philanthropic investment as an effective means for collectively solving complex social problems. current theories examine the evolution of these groups by resting on the traditional organizational forms of market, hierarchy, and network, without a consideration of trajectories that show movement between organizational forms over time. this article utilizes the network itself as the unit of analysis within a larger network domain to examine the most common trajectories and changes in organizational forms over time. to date, little theory has been developed or applied to account for both endogenous characteristics and exogenous system-wide dynamics and their longitudinal effects on networks. as is appropriate in the early stages of developing new theories, this article addresses the foundational steps of first clarifying the phenomenon of interest with the creation of a typology of the ways in which networks evolve as organizational forms and suggesting future network-level and network domain research agendas. keywords: purpose-oriented networks; evolution; network forms introduction in practice, health and social services are commonly delivered through purpose-oriented networks (pons) (nowell & kenis, 2019) comprised of public agencies, nonprofit organizations, for-profit businesses, and consumer associations to address complex problems and implement public policy. these networks are often favored by government and philanthropic investment as an effective means for collectively solving complex social problems (head & alford, 2015; morçöl, 2014; weber & khademian, 2008). pons are multi-organizational governance arrangements that can form and evolve over time, yet the nature of these changes has only been considered in the current literature under the assumption that the network itself maintains its organizational form as a network. to date, there has been little consideration of how and why a pon may potentially move between organizational forms. overall, research and scholarly advancements have focused on differentiating the dynamics of organizational interactions, as seen in traditions like transaction cost economics (williamson, 1975, 1981), development of clans (ouchi, 1980), and collaborative governance regimes (ansell & gash, 2007; emerson et al., 2012), rather than considering how and why a network may evolve out of, and into, other organizational forms. this article seeks to address the concerns of a recent article taking stock of network and collaborative governance literature in public management that notes that there is a “lack of research on network evolution and tie dissolution” ( siciliano, wang, & medin, 2020, p. 14). past studies have overlooked the possibility for system-wide dynamics that could contribute to a network’s changes over time in network domains with shared members (one exception is nowell, hano, and yang, 2019). this static view has considered networks as insular and only understood through process and structure theories that are driven by the assumption that the network organizational form itself is stable over time. for example, network governance research has largely grown from provan and kenis’ (2007) explication of three progressively more centralized internal processes like shared-governance, lead organization, and network administrative organization. this internal focus overlooks how and why these groups may respond to environmental contingencies by evolving into other organizational forms as markets or hierarchies over time, rather than remaining a network at all. this article utilizes the pon itself as the unit of analysis within a larger network domain which is bound as a collection of networks sharing an environmental niche defined by a similar geographic area mailto:kalbrech@uic.edu http://dx.doi.org/10.20377/cgn-116 26 university of bamberg press and problem-area of focus (nowell et al., 2019). the specific context of this research includes a variety of health and wellness networks that are situated in a broader network domain defined by shared members who create meaningful inter-network ties. our field’s theory building and empirical research has mostly focused on the nature of changes within a network over time by most often leveraging data from case studies that focus on one or two networks alone (voets et al., 2020). while there has been some discussion and examination of the nature of the network as a whole to advance an exogenous theory of understanding performance, there is a limited body of studies that examine evolution using the network itself as the unit of analysis (isett et al., 2011; kenis & provan, 2009; raab & kenis, 2009; rethemeyer & hatmaker, 2007). this research seeks to advance our field’s body of knowledge about the nature of a network’s trajectory as an entity in-and-of-itself in a network domain over time. mechanisms of internal network characteristics are considered here, as has been a tradition in public administration scholarship (isett et al., 2011), but these forces are also examined in combination with a focus on how a pon’s larger network domain (an exogenous force) coupled with its unique characteristics (endogenous factors) may drive evolutionary changes between organizational forms. by considering exogenous system-dynamics, this research highlights issues of flexibility and adaptability, as well as the possibilities for networks to differentiate, merge, or spin-off new groups over time (nowell et al., 2019). by considering the network’s longitudinal trajectory, evolutionary outcomes can also be examined to determine what important aspects may remain after the network organizational form disbands. as is appropriate in the stages of developing new theories, this research addresses the foundational step of both clarifying the phenomenon of interest and describing its components. two research questions are addressed in this article: 1. do networks change form over time? 2. if so, what are the ways in which networks can change? this article first examines organizational and network theories regarding structure and evolution of networks over time, highlighting the current gap in scholarship regarding how and why networks may evolve out of a network organizational form. the data and methods of this research are discussed, followed by a typology of evolution of organizational forms. to conclude, discussion of the findings and avenues for future research are presented. network evolution: structure and process change theories networks as a phenomenon can be examined both as social structures and governance forms. this article focuses on networks as an organizational form of governance, or a way to collectively and strategically address issues through purpose-oriented networks (agranoff & mcguire, 2001; nowell & kenis, 2019). this view aligns with the commonly held definitions of community-based networks that engage in collaborative governance (emerson, nabatchi, & balogh, 2012) and are highly boundary spanning, and focused on complex social problems that cannot be solved by one organization or agency alone (agranoff & mcguire, 2001; bryson, crosby, & stone, 2015, morçöl, 2014). as noted by provan, fish, and sydow (2007), networks as a form of governance have focused on two levels of analysis: 1) the micro or egocentric level in which individuals and organizations are the center of attention, or 2) the macro or whole-network level. to situate this study also in the field’s traditions of collaborative governance research, it is important to note distinctions of definitions. collaborative governance is “the processes and structures of public policy decision making and management that engage people constructively across the boundaries of public agencies, levels of government, and/or the public, private and civic spheres in order to carry out a public purpose that could not otherwise be accomplished” (emerson et al., 2012, p. 2). pons are one mode through which collaborative governance may take place, and indeed they often struggle with similar issues of time, trust, and interdependence (ansell & gash, 2007; berthod & segato, 2019). complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 27 university of bamberg press this review of extant literature focuses on a whole-network, external view of pons and is summarized in table 1 below. this overview is an illustration of how far the field has come, yet also highlights gaps. as will be discussed in more detail in the next section, these past studies do not consider the more complex dynamics that are occurring within a community that houses many network efforts in overlapping issue areas that constitute a domain. indeed, prior network literature has overlooked the influence of environmental effects of a network domain, defined as a population of pons focused on the same broad mission area within a defined geographic area, by studying networks as if they are in insular (one notable exception is nowell et al., 2019). network domains are also defined as systems of resources in which networks may compete or collaborate. past literature has examined forces external to a network, like financial resources and policy changes, but research has yet to clearly define and operationalize the effects of the dynamic nature of interdependencies across networks in a system with many other networks. scholars who have looked at the whole-network level of analysis have broadly been concerned with governance, structures, and how network outcomes are generated (provan et al., 2007). whole-network studies still recognize the importance of individual organizations or agencies, but put more emphasis on the nature of the network’s behavior over time as a collective group. whole-network research is defined by three or more organizations that are connected in meaningful ways to achieve a common goal (provan et al., 2007). how networks evolve (frameworks) why networks evolve (mechanisms) what features of networks evolve (components) linear-sequential • (gray, 1985) overall processes of a network as it forms, based in domainlevel interactions • (lowndes & skelcher, 1998) consideration of partnership forms across market, hierarchy, and network; progression of forms is based on interorganizational relationships, but level of analysis is the form itself • (provan & kenis, 2007) governance focus from participant to nao, linear "maturing" of network organizing structure cyclical • (doz, 1996) learning cycles for a whole network; initial conditions enter cycle of learning, reevaluation, and revising conditions dialectical • (mcguire, 1988) focus on network construction contingencies and environment • (koka et al., 2006) changes in resource munificence and uncertainty lead to organization or agency actions, like tie creation or deletion and changes in network portfolio size and scope (egocentric dynamics that aggregate up to the network level) developmental needs • (butterfoss & kegler, 2009)network outcomes are driven by roles, relationships, decisionmaking processes, and member satisfaction/participation patterns • (feiock, 2013) focus on nature and change of institutional collective action (ica) arrangements based on collaboration risk and transaction costs path dependencies • (gray, 1985) overall processes of a network as it governance and formality • (herranz, 2009) different network coordination strategies that help balance the informal and formal aspects of ring and van de ven's (1994) egocentric framework • (feiock, 2013) focus on nature and change of institutional collective action (ica) arrangements based on collaboration risk and transaction costs • (provan & kenis, 2007) governance focus from participant to nao, linear "maturing" of network organizing structure • (saz‐carranza & vernis, 2006) based on critique of past linear process studies; view of governance and formality as a more complex process size, composition, and structure • (koka et al., 2006) changes in resource munificence and complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 28 university of bamberg press how networks evolve (frameworks) why networks evolve (mechanisms) what features of networks evolve (components) and externalization based on the social paradigms of component organizations and agencies nonlinear • (koliba et al., 2010) environmental factors (such as privatization, devolution, public-private partnerships) that give rise to network forms • (scheinert et al., 2017)complex adaptive systems view emphasizing stochastic or random nature of network evolution forms, based in domain-level interactions • (herranz, 2009)different network coordination strategies that help balance the informal and formal aspects of ring and van de ven's (1994) ego-centric framework • (moynihan, 2009) and (nowell & steelman, 2015) impact of network characteristics like diversity, shared authority, and trust/relationships; disasterresponse context uncertainty lead to organization or agency actions, like tie creation or deletion and changes in collaboration portfolio size and scope, that aggregate up to the network level • (milward et al., 2010) comparison of for-profit and nonprofit nao governed networks along the dimensions of structure, relationships/trust, and preliminary performance measures • (nowell & steelman, 2015) impact of network characteristics like trust and relational embeddedness; disasterresponse context • (provan & kenis, 2007)governance focus from participant to nao, linear "maturing" of network organizing structure group dynamics • (milward et al., 2010) comparison of for-profit and nonprofit nao governed networks along the dimensions of structure, relationships/trust, and preliminary performance measures • (provan & kenis, 2007) governance focus from participant to nao, linear "maturing" of network organizing structure • (saz‐carranza & vernis, 2006) based on critique of past linear process studies; view of complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 29 university of bamberg press how networks evolve (frameworks) why networks evolve (mechanisms) what features of networks evolve (components) governance and formality as a more complex process capacity • (milward et al., 2010) comparison of for-profit and nonprofit nao governed networks along the dimensions of structure, relationships/trust, and preliminary performance measures • (provan & kenis, 2007) governance focus from participant to nao, linear "maturing" of network organizing structure table 1: traditions of whole-network evolutionary theories trajectories and frameworks of evolution: recent whole-network literature has also examined linearsequential frameworks of evolution. these linear-sequential models have had such a strong hold on our understanding of evolution that these models have even been laid over theories of other aspects of network structures. in provan and kenis’ (2007) exploration of governance types, they argue that networks will inevitably evolve from shared governance to more formal types of lead-organization or network administrative governance (nao), with no discussion or consideration of the possibility of regressing or cycling backwards. alongside ring and van de ven’s (1994) egocentric model of cyclical evolution, doz (1996) also advanced a whole-network framework that focuses on the nature of learning and feedback loops. the dialectical model focuses on the ways in which network group members make sense of their social paradigms and power relations, with a clear drive towards creating stability so that the work of the whole group can be formalized and externalized (mcguire, 1988). within the nonlinear theories, complex adaptive system (cas) models have begun to view networks as systems themselves that are informed by path dependence (pierson, 2004). koliba et al. (2010) have suggested a framework in which evolution is a function of environmental factors like privatization, devolution, and public-private partnerships that give rise to network forms of governance. this is the one example within network evolution scholarship, and a very recent area of exploration, that has been pushing on the idea that networks can be part of larger system and susceptible to exogenous pressures and shocks (scheinert et al., 2017). mechanisms and drivers of evolution: within public management literature, a diverse collection of theories and frameworks about why networks evolve over time has emerged. while these theories do collectively focus on the drivers and conditions for evolution, the three approaches have widely different assumptions. one of the earliest explorations of drivers of network evolution is rooted in contingency theories that were previously applied to organizations (drazin & van de ven, 1985). while a focus on contingencies does take a pon’s environment into account, koka et al.’s (2006) theory of network evolution leverages only single-network dynamics, with an emphasis on egocentric dynamics that are complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 30 university of bamberg press aggregated up to the whole-network level. in this research. koka et al. (2006) do not consider that a network’s external domain may function as a network in-and-of-itself whose structure can have unique effects on evolution over time. developmental perspectives on network evolution have different theoretical traditions, largely drawing upon the concept that networks will evolve through stages that build upon each other. butterfoss and kegler’s (2009) community collective action theory (ccat) is rooted in social psychological theory and uses the assumption that networks will evolve in the same ways as other kinds of social groups. the institutional collective action (ica) framework as developed by feiock (2013) rests in the traditions of political economy and rational choice theory by focusing on intergovernmental networks. given this viewpoint, feiock’s work assumes that interdependencies will drive the formation of a network, and that as these interdependencies shift over time new drivers of evolution will emerge. some scholars have expanded on the exogenous driver category, developing propositions informed by path dependency theories. in these studies, initial starting conditions are shown to lead to different evolutionary trajectories for the whole network over time. for example, some studies have focused on the legitimacy of the network’s convener (gray, 1985), differences in the institutional logics of coordinating strategies within a network (herranz, 2009), and social capital and relational embeddedness before networks convene (moynihan, 2009; nowell & steelman, 2015). again, while pre-conditions are considered exogenous factors in these cases, less attention is given to whether other networks also in the network domain and the potential for that broader system to contribute to the dynamics of changes organizational forms features that evolve: whole-network studies have dominated research that is concerned with the components of networks that evolve. logically, the whole-network level of analysis does lend itself to more holistic examinations of what features of a network can change over time, like governance and formality, size and structure, and group dynamics. within the area of governance and formality, feiock’s (2013) ica framework embraced a whole-network approach to address issues of coordination and change as risk and transaction costs change over time. as discussed in other examples above, provan and kenis’ (2007) work made major strides in describing and defining three potential governance arrangements, creating a spectrum spanning from shared participant responsibility to a formal network administrative organization (nao). another interesting contribution to this literature is herranz’s (2009) whole-network research that leveraged ring and van de ven’s (1994) ego-centric framework by applying it to network coordination strategies of the whole group based on the balance of informal and formal needs. governance processes in practice at the whole-network level have also been cited as an example of why linear models of evolution are potentially too simplistic, given that the negotiations that result in formality and governance arrangements have many cyclical, feedback loop characteristics (saz‐carranza & vernis, 2006). network size, composition, and structure has also been widely examined at the whole-network level of analysis. koka et al.’s (2006) work examined whole-network contingencies, drawing propositions about the effects of resource munificence and environmental uncertainty on tie creation or deletion within a network. structure was also examined in milward et al.’s (2010) comparison of for-profit versus nonprofit nao-governed networks. in the disaster response context, nowell and steelman’s (2015) research considered how trust and relational embeddedness within a network lead to dynamic structures and ties between responders. group dynamics are also unique to whole-network studies given the broader focus on network outcomes and the nature of network processes themselves. again, milward et al.’s (2010) and provan and kenis’ (2007) work explores changes in whole-network group dynamics over time. closely associated with research questions regarding group dynamics have been explorations of whole-network capacities over time (milward et al., 2010; provan et al., 2007). in addition to a wide variety of potential aspects of networks that may undergo evolution, across many studies, multiple aspects of a network are theorized to have a relationship with other aspects in such a way that networks function as complex systems (nowell et al., complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 31 university of bamberg press 2016). overarchingly, this literature is concerned with components changing over time and is the most internally focused of all network evolution research. integrating organizational theories and network evolution theories while the field has examined the phenomenon of the network itself, including its formation, processes, and internal characteristics, less is known about the nature of when, why, and how a network as an entity may change organizational form over time. to begin to examine changes in networks to other organizational forms over time, there are some insights in returning to traditional organizational theory for a foundation and suggested paths forward. this research considers the dynamics of network evolution to build from traditions of scholarship that define a network as an entity in-and-of-itself with meaningful differences from markets and hierarchies (thorelli, 1986). the literature reviewed below is offered as a jumping-off point, and a means for understanding where we have come regarding our understanding of the network form and where we still need to go. in the traditions of network scholarship that have followed the foundations laid by organizational theory, network forms have been examined through the structures that support shared decision-making (bryson et al., 2015). based on provan and kenis’ (2007) discussion, three specific modes of governance were established: shared, lead organization, and network administrative organization (nao). while these designations of governance types have been widely accepted and studied in public management, (for an overview see bryson, crosby, and stone 2015) they remain bounded to the network organizational form. since the establishment of provan and kenis’ (2007) three governance types, little scholarship has moved beyond this frame to consider the potential phenomenon of a network exiting or evolving out of the network organizational form altogether. organizational theory, notably the examination of mechanistic versus organic systems of management (burns & stalker, 1961), is illuminating in this context. while burns and stalker (1961) were examining organizational structure, this research also considers networks as a possible system of management. as rational choice and contingency theories suggest, there is no one ideal type, which in the context of complex social problems suggests that perhaps there may not only be one ideal type of solution over time—such that networks are only one of many possible organizational forms. contingency theories recognize that organizations can, and should, move back and forth between mechanistic or organic management systems forms in response stability or change (burns & stalker, 1961; drazin & van de ven, 1985; gresov, 1989). mechanistic systems rest on specialization and hierarchies to maximize decision-making processes and respond to a stable external environment (burns & stalker, 1961). organic systems are more appropriate during change by harnessing an on-going commitment to adjustment and re-definition of the task (burns & stalker, 1961). as such, the fit of a network form of governance as an organic approach can be considered using the system lens within contingency theories (drazin & van de ven, 1985). networks are a response to complex problems and systems, and a way to integrate the multitude of contingencies and structures required for multiple organizations to collaborate. and, network organizational forms may adapt and evolve towards more mechanistic organizational forms when the issue at hand is more stable and has a lower task uncertainty (drazin & van de ven, 1985; ring & van de ven, 1994). although network theories have engaged with the concept of evolution, again there is limited consideration of the possibilities of the actions and activities of a network moving between organizational forms (koliba et al., 2010). as discussed below, this research recognizes that many pons cannot be considered in a vacuum because human resources are shared between groups in the form of members who engage in meetings and initiatives. in other words, the actors in a network domain are often a finite group of people who serve in multiple networks. thus, each person is an essential tie that provides structural and functional connections, as well as a potential resource that can be overtaxed by engagement in multiple networks over time (nowell et al., 2019). complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 32 university of bamberg press network evolutionary trajectories: establishing a typology to begin to examine the phenomena of networks themselves evolving within a network domain, this research aims to establish potential evolutionary trajectories of a pon over time. using a qualitative analytical inductive approach, a typology of network evolutionary trajectories is discussed below. this typology emerged from the data, being established based on the presence of forms and functions of networks changing organization forms over time across three network domains. data this research leverages a secondary longitudinal population-level dataset that includes all health‐ oriented networks in three counties in a southeastern state from 2012-2017. the data for this analysis is a subset of a population of 74 networks and their members taken over two time points. data for this article is a subset of a larger data collection that also included network data, survey data, and interview data. the data collection at the first time point was part of a state-funded initiative to establish more effective community-based health and wellness collaborations. time 1 data collection was able to achieve a 100% response rate from all identified health-oriented networks in the counties included in the study, and time 2 data collection achieved a 97% response rate. the list of networks was cross-validated by county informants as well as an exhaustive web search to ensure comprehensiveness. pons in this dataset are defined as three or more organizations that meet on a regular basis and have a health or wellness focus to their work. this definition of a network served as the inclusion criteria for data collection for the duration of the study. each participating network was asked to provide a current list of their members and associated organizational affiliations. data was collected from network coordinators who serve in a wide variety of roles from formal, paid positions, to rotating, less formal, voluntary leadership. the included pons exhibit governance approaches ranging from member-led, organization-led, and network administration organization (provan & kenis, 2007), and each coordinator serves in a network-recognized leadership position. network coordinators were asked to participate in a semi-structured phone interview. to meet the needs of this research’s inductive approach, all informants were asked to share details about the story of how their network began, with an emphasis on the motivations and impetus for the group to convene. informants were also asked to explain the process of how the network was formed. additionally, all network leaders were asked if the network was originally part of another group and could be consider as a spin-off or beginning as part of another network in the community. any coordinators of networks that were identified as no longer meeting regularly in 2017 were asked additional questions to determine the nature of the change to their network over time. coordinators of discontinued networks confirmed that the group was no longer meeting regularly in a network organizational form. next, these coordinators were asked to discuss the nature of the changes to their group, what factors may have led to this change, and if any aspects of the network were still functioning in the community in a different organizational form. methods data from the 2017 network leader interviews were analyzed using a qualitative analytical inductive approach to create a codebook of characteristics and definitions of a network’s survival, death, or transformation to a different organizational form. this codebook was established from the emergent definitions provided by network leaders across all 58 cases in this subset of the data. as a potential evolutionary trajectory was described within an interview, it was defined and added to the codebook. over the course of the interview process, the cases of evolutionary trajectories coalesced into three main categories: survival, death, and transformation. survival is defined as the presence of the network across complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 33 university of bamberg press data collections, while death is defined as the absence of the network at the second data collection. the category of transformation also included three sub-categories that are described below. within the category of transformation, possible organizational forms include: 1) merging within another existing network, 2) program, service, or hierarchy in the form of an organization, and 3) spinningoff in a new direction by redefining mission and approach while retaining many of their original members in a new pon. this codebook was then applied to all networks in the dataset to create a typology that includes definitions for survival, death, and most importantly transformation. application of the codes was verified through peer checks with peer researchers involved in other data collection from the same project. all peers agreed on the appropriateness of coding. table 2 below illustrates this study’s sample and the number of evolutionary outcomes assigned based on the codebook that was developed. outcome number of networks survival of network 43 death of network 6 transformation of organizational form merger 4 program 3 spin-off 2 table 2: network organizational form evolutionary typology findings: defining network trajectories over time network survival and death are the most intuitive evolutionary outcomes to define. for the purposes of advancing suggested definitions of organizational transformation, the definitions of network survival and death are expanded upon below with supporting examples from this study’s data. pons in this study are defined as three or more organizations that meet on a regular basis and have a health or wellness focus to their work. consistent with this inclusion criteria, if a network has continued meeting from 2012 to 2017 and the mission has remained focused on community health or wellness, the network is considered to have survived. in this study, 58% of networks fit this definition. networks that have discontinued meeting regularly are considered to have died. meeting discontinuation was verified with the most recent past coordinator or facilitator of each network group. in this study, 8% of networks fit this definition. in addition to verifying that the group has ceased its activities, the coordinators provided a detailed description of the nature of the end of the network’s meeting and activities. survival of the networks included in the sample, a little more than half survived from 2012 to 2017. survival is defined by a network continuing to meet on a regular basis to address issues of community health and wellness. while this trajectory does align with current assumptions about the momentum and presence of networks over time, survival of networks in a community-system exhibit patterns of dynamic stability regarding membership and leadership turnover. across the three counties included in this analysis, there was a variety of expansion, contraction, and maintenance of members and organizations involved in their network domains. as discussed above, human resources and the nature of shared membership can be an important frame and subsequent mechanism for understanding the nature of network survival over time. coordinators of networks that have continued from 2012 to 2017 were asked to discuss how and why their membership changed over time. one network coordinator shared, “our membership has almost completely turned over…due to aging of the workforce and changing jobs,” while another noted that, “three of our big leaders have retired. nothing is static, there's always change.” a third coordinator also complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 34 university of bamberg press described how her network has not only survived, but expanded its presence in the community by saying, “we have new engagement with increased social capacity and network…communication of successes and plans has increased reach and built more connections.” these comments from network coordinators suggest that member turnover, even when the network has survived over time, comes as both a threat and an opportunity. there can be a positive influx of new energy to sustain networks, as well as significant challenges in developing new leadership and member commitment. death among those networks that ceased meeting regularly, one subset is considered to have died. when a pon dies, it also does not retain any type of other organizational form or function in the network domain. the transformation typology in the section below describes another type of trajectory where networks may have discontinued but remained in the network domain in a different organizational form. past coordinators of the networks that were discontinued offered their narrative about the reasons for the group no longer meeting. of major consideration was the challenge of leadership, whether provided by paid staff or members. one coordinator shared that, “we asked for someone to step up [as new coordinator] and no one did.” another leader noted that, “the staff had to be pulled off project because there was no state money to support it. there were staffing resources that changed. committees were not meeting as much when staffing resources went down. when [the new governor] came in, sustainability had to be changed and taken out of project.” overall member engagement was also a challenge for networks that died. a coordinator shared that the “group was dwindling and there was push back from those in power in community.” another coordinator described engagement challenges due to the “number of people engaged was not as great. there was a lot of organizer turnover and trouble with making items actionable.” in networks that died, there was also tension with both leadership issues and member engagement together. one coordinator described this occurrence by saying, “[a new person] became chair and we lost people in the transition because there was some conflict between members. [the past coordinator who founded group left], and when you don't have a funding source and no strong leadership there was no focus. we were spinning in circles and fewer and fewer people [attended meetings] every time.” these findings themselves highlight the need for an external view of networks and consideration for the dynamics that can lead a network to survive or cease meeting. to date, case study research into the nature of network changes over time has recognized that a group may experience life-cycle changes (gray, 1985; lowndes & skelcher, 1998; provan & kenis, 2007; straub et al., 2007), but there are few studies that examine network dissolution or transformations in organizational form. two studies that do examine the dynamics of networks disbanding focus on single-sector alliances or consortiums with narrow, contextual goals (hu, guo, & bies, 2016; provan, 1984), not multi-sector networks striving to address complex, community-wide wicked problems. the presence of network death in multi-sector network domains exemplifies the need to further understand how and why a network may no longer exist in a network form. additionally, a portion of networks in this research took on a different organizational form than a pon. these organizational form transformations are discussed below. transformation if some aspects or activities of the network changed in the three ways described below, the cases were included in the transformation typology analysis. fifteen percent of networks were defined as complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 35 university of bamberg press transforming by merging with a current network, becoming a program of another organization or agency, or spinning-off a new pon. one important finding to also note is that no transformations of pons in this study show evidence of internal reorientations or recreations of their collaborative governance processes (ansell & gash, 2007; emerson et al., 2012; ulibarri et al. 2020). rather, networks studied here fundamentally changed their organizational form instead of simply re-negotiating the collaborative processes to remain a network. it is important to note again that all networks included in this study are defined as purpose-oriented with “some degree of shared purpose, some degree of membership of three or more actors, some degree and form of joint effort, and some form of governance” (carboni et al., 2019; nowell & kenis, 2019, p. 194). thus, transformations here are considered a conscious choice of the members of the network to establish a new organizational form, rather than simply adjust internal processes. table 3 below provides definitions of the organizational form of each type of transformation, as well as a visual representation. in these representations, blue squares represent networks, red circles represent organizations, and green triangles represent subcommittees or task forces of networks. transforming by merging with an existing network: the majority of transformations in this sample represent a network that discontinues meeting under its current name and function and begins meeting as part of another network that already exists in the network domain. importantly, when a network transforms by merging with another, a meaningful aspect of its former identity and mission is retained as it begins its new form as a subcommittee, working group, or task force of a pre-existing network. for example, a group focused on addressing issues of diabetes in both adults and children transformed into a support group aspect of another pon in the same network domain: “about the support group with [the other community network], we have that going and we help with that. we have also developed a diabetes food list and best things to donate to help people with diabetes and being mindful of sodium and things like that is distributed through [the other network].” in another example, a former pon that was focused on community issues of nutrition and access to opportunities for physical activity became a specific task force of another already-functioning pon. in this instance, network members were willing to merge despite losing the name and mission of the original group to gain resource efficiency: “we wanted to maximize peoples' time because that is not an unlimited resource. we had two groups that really were overlapping in a lot of ways. it was an intentional, well-thought process. could we be more efficient and effective in our use of staff time, dollars, and resources by merging these two groups? yes! and so we did.” a final example of a network transforming by merging is when a pon to address elder abuse and victimization disbanded. many members continued the work by engaging in a task force of another local pon that is more broadly focused on senior health issues: “yes, [the network] to address adult abuse statewide was competing and our members were more active there too. [network that no longer meets had] members that were state level workers and were pulled to state initiative that has a local network in our community.” transforming into a hierarchical organizational form: other networks’ main activities and mission transformed into a program of a local organization or agency. in these examples, the work of the pon moved back into the hierarchy space, sometimes retaining original network members as partners or advisors. ultimately, examples of transformations into programs are described as the collective action of a network group disbanding, but with its substantive activities becoming the responsibility of a single organization or agency. one example is of a local food advocacy pon discontinuing their meetings when the municipal government undertook efforts to support an ordinance to allow urban farms and farm complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 36 university of bamberg press stands. while the policy change was one goal of the pon, their main aim was to financially support development of urban farms and distribution of local food: “once the [new guideline] was passed, the city started its own meetings and a program [for food advocacy]. and that is why we paused meeting, and eventually disbanded…our initiative was now being handled by the city.” another transformation of a network into a program is evident in an end-of-life support coalition pon becoming a work group of the local nonprofit health care system: “the [local health system] now has an end of life work group that includes numerous people from the [transformed network] and the community. the work is continuing and it's important.” in another community, a network that was focused on offering coordinated mental health support for adults transformed into a care review program of the local managed care organization. while the pon has discontinued meeting, some previous members have been retained in the care review program: “we still continue on with care reviews. this is a venue for individuals who are experiencing challenges and barriers. they come to care review to see a team from the systems in the community…this is housed within the [local managed care organization], and it is a system of care initiative program.” transforming as a spin-off that creates a new pon: a final type of transformation is evident in a new pon, with a similar mission, spinning-off from a previous, discontinued pon. as in the example below, a spin-off is defined as a strategic and premediated action to develop a new network from some of the aspects of one that is no longer meeting. in one community, a group of physicians concerned about the opioid epidemic recognized that it was just one aspect of larger community issues surrounding those in crisis: “our group became the [community crisis group] as the opioid safety task force spin-off. the membership was very similar and the idea was to just move things forward in the [community crisis group].” discussion the typology of network trajectories advanced above raises several important questions about the nature of network evolution when network domain conditions and dynamics are considered. given that the phenomenon of changes in organizational form has been documented in this study, future research to advance theory should move beyond the internal focus of past network studies complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 37 university of bamberg press merger • form – subcommittee, task force, or work group of a former network • group is subsumed within a new advisory role with a new network in the same domain hierarchy • form – programmatic aspect or initiative of an organization or agency in the network domain • specific, bounded activity within the hierarchy of an organization or agency spin-off • form – new collective group meeting regularly as a network in the network domain • focuses on an issue that was partially addressed in a former network in the network domain table 3: typology of network transformation indeed, supported by the typology presented here, networks have transformed into other organizational forms like programs, becoming like a hierarchical organization, and networks have merged with others, remaining network-like but within the new structure and governance system of a different network. the transformations present in this data challenge our field’s current focus on evolution that takes place solely within a network. ego-centric and whole-network studies that examine evolutionary theories have proliferated, yet little attention has been paid to when, why, and how a network as an entity may change organizational form over time. indeed, until recently, the assumption across some public management research has been that once a group of organizations comes together in a network form, that group will only survive or die over time. this research highlights how the network as a governance form is more fluid, answering calls for theory building at the network level of analysis (isett et al., 2011; nowell et al., 2019). findings from the development of a typology of network evolution and subsequent organizational forms demonstrate that networks can survive, die, or transform over time. the phenomenon of transformation highlights that, even in complex problem spaces, a network organizational form may not be the most sustainable, or most appropriate, approach over time. this broader view harkens back to past advancements in organizational theory which focused on the evolution of organizational form (pfeffer & salancik, 1978; williamson, 1975, 1981). in this study, network members and the network domain shape complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 38 university of bamberg press the form of collaborative approaches to be only as complex and effective as necessary based on a variety of endogenous and exogenous pressures. across all three organizational types, mergers, hierarchies, and spin-offs, the nature of the network domain was a key component. data shows that mergers were facilitated by the presence of another network doing similar or overlapping community work, hierarchies were chosen when the network domain and problem space had adjusted such that the pon’s focus could be better addressed through a specific activity or initiative, and spin-offs were a result of a specific aspect of a network gaining more traction or resources in its network domain such that the other activities were no longer effective. the qualitative analytical inductive approach of this research is suited to theory and concept development, but is limited to transferability and not generalizability (nowell & albrecht, 2019). future research agendas should explore the mechanisms of network transformation that lead to different organizational forms, as well as develop potential propositions that can advance network evolutionary theories that focus on pons in complex network domains. conclusion this research suggests that networks nested within network domains may transform organizational form to respond to both internal and external pressures. the nature of these transformations and the associated theory building is an essential direction for future research. continuing scholarship in this area could advance our understanding of the fluidity of network organizational forms by further clarifying other types of network domains, those created by shared funding sources or shared policy mandates for example. as shown in the typology of survival, death, and transformation, networks as governance systems and structured entities themselves do undergo shifts in their organizational form over time. these changes can result in a pon’s sustainability or its discontinuation. most importantly, transformations can also take place that retain the elements of past networks into conditions more commonly defined as non-network arrangements. the nature of these shifts is nested within the context of networks not being insular or isolated, but rather meaningfully connected to other networks through shared membership in their network domain. further research may also establish other important exogenous, systemic conditions that tie networks together in such a way that the network domain becomes an essential dynamic force for understanding pon evolution over time. recent calls by network scholars have issued the challenge to begin developing theory to examine networks as the unit of analysis and an outcome of concern in-and-of themselves (carboni et al., 2019; nowell et al., 2019; raab & kenis, 2009). to answer this challenge, this article establishes the phenomenon of network trajectories encompassing survival, death, and transformation. this research also advances a typology that focuses on the changing organizational forms and functions of these networks embedded within a larger network domain. in doing so, this continues an emergent research agenda that embraces the external view of networks and moves beyond the assumptions of networks existing in isolation. references agranoff, r., & mcguire, m. (2001). big questions in public network management research. journal of public administration research and theory, 11(3), 295–326. ansell, c., & gash, a. (2007). collaborative governance in theory and practice. journal of public administration research and theory, 18(4), 543–571. https://doi.org/10.1093/jopart/mum032 berthod, o., & segato, f. (2019). developing purpose-oriented networks: a process view. perspectives on public management and governance, 2(3), 203–212. https://doi.org/10.1093/ppmgov/gvz008 bryson, j. m., crosby, b. c., & stone, m. m. (2015). designing and implementing cross-sector collaborations: needed and challenging. public administration review, 75(5), 647–663. https://doi.org/10.1111/puar.12432 burns, t., & stalker, g. (1961). mechanistic and organic systems of management. in the management of innovation (pp. 96– 1125). oxford university press. complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 https://doi.org/10.1093/jopart/mum032 https://doi.org/10.1093/ppmgov/gvz008 https://doi.org/10.1111/puar.12432 39 university of bamberg press butterfoss, f. d., & kegler, m. c. (2009). the community coalition action theory. in r. diclemente, r. crosby, & m. c. kegler (eds.), emerging theories in health promotion practice and research (2nd ed.). jossey-bass. carboni, j. l., saz-carranza, a., raab, j., & isett, k. r. (2019). taking dimensions of purpose-oriented networks seriously. perspectives on public management and governance, 2(3), 187–201. https://doi.org/10.1093/ppmgov/gvz011 doz, y. l. (1996). the evolution of cooperation in strategic alliances: initial conditions or learning processes? strategic management journal, 17(s1), 55–83. drazin, r., & van de ven, a. h. (1985). alternative forms of fit in contingency theory. administrative science quarterly, 30(4), 514. https://doi.org/10.2307/2392695 emerson, k., nabatchi, t., & balogh, s. (2012). an integrative framework for collaborative governance. journal of public administration research and theory, 22(1), 1–29. https://doi.org/10.1093/jopart/mur011 feiock, r. c. (2013). the institutional collective action framework. policy studies journal, 41(3), 397–425. gray, b. (1985). conditions facilitating interorganizational collaboration. human relations, 38(10), 911–936. gresov, c. (1989). exploring fit and misfit with multiple contingencies. administrative science quarterly, 34(3), 431. https://doi.org/10.2307/2393152 head, b. w., & alford, j. (2015). wicked problems: implications for public policy and management. administration & society, 47(6), 711–739. herranz, j. (2009). endogenous development dynamics of multisectoral network management. international public management journal, 12(3), 370–397. https://doi.org/10.1080/10967490903094145 hu, m., guo, c., & bies, a. (2016). termination of nonprofit alliances: evidence from china. voluntas: international journal of voluntary and nonprofit organizations, 27(5), 2490-2513. hu, q., khosa, s., & kapucu, n. (2016). the intellectual structure of empirical network research in public administration. journal of public administration research and theory, 26(4), 593–612. https://doi.org/10.1093/jopart/muv032 isett, k. r., mergel, i. a., leroux, k., mischen, p. a., & rethemeyer, r. k. (2011). networks in public administration scholarship: understanding where we are and where we need to go. journal of public administration research and theory, 21(supplement 1), i157–i173. https://doi.org/10.1093/jopart/muq061 kenis, p., & provan, k. g. (2009). towards an exogenous theory of public network performance. public administration, 87(3), 440–456. https://doi.org/10.1111/j.1467-9299.2009.01775.x koka, b. r., madhavan, r., & prescott, j. e. (2006). the evolution of interfirm networks: environmental effects on patterns of network change. academy of management review, 31(3), 721–737. koliba, c., meek, j. w., & zia, a. (2010). governance networks in public administration and public policy. crc press. lowndes, v., & skelcher, c. (1998). the dynamics of multi-organizational partnerships: an analysis of changing modes of governance. public administration, 76(2), 313–333. mcguire, j. b. (1988). a dialectical analysis of interorganizational networks. journal of management, 14(1), 109–124. milward, h. b., provan, k. g., fish, a., isett, k. r., & huang, k. (2010). governance and collaboration: an evolutionary study of two mental health networks. journal of public administration research and theory, 20(supplement 1), i125–i141. https://doi.org/10.1093/jopart/mup038 morçöl, g. (2014). complex governance networks: an assessment of the advances and prospects. complexity, governance & networks, 1(1), 5–16. moynihan, d. p. (2009). the network governance of crisis response: case studies of incident command systems. journal of public administration research and theory, 19(4), 895–915. https://doi.org/10.1093/jopart/mun033 nowell, b., & albrecht, k. (2019). a reviewer’s guide to qualitative rigor. journal of public administration research and theory, 29(2), 348–363. nowell, b., hano, m. c., & yang, z. (2019). networks of networks? toward an external perspective on whole networks. perspectives on public management and governance, 2(3), 213–233. https://doi.org/10.1093/ppmgov/gvz005 nowell, b., & kenis, p. (2019). purpose-oriented networks: the architecture of complexity. perspectives on public management and governance, 2(3), 169–173. https://doi.org/10.1093/ppmgov/gvz012 nowell, b., & steelman, t. (2015). communication under fire: the role of embeddedness in the emergence and efficacy of disaster response communication networks. journal of public administration research and theory, 25(3), 929–952. https://doi.org/10.1093/jopart/muu021 ouchi, w. g. (1980). markets, bureaucracies, and clans. administrative science quarterly, 25(1), 129–141. pfeffer, j., & salancik, g. (1978). the external control of organizations: a resource dependence perspective. provan, k. g. (1984). interorganizational cooperation and decision making autonomy in a consortium multihospital system. the academy of management review, 9(3), 494–504. https://doi.org/10.2307/258289 provan, k. g., fish, a., & sydow, j. (2007). interorganizational networks at the network level: a review of the empirical literature on whole networks. journal of management, 33(3), 479–516. https://doi.org/10.1177/0149206307302554 provan, k. g., & kenis, p. (2007). modes of network governance: structure, management, and effectiveness. journal of public administration research and theory, 18(2), 229–252. https://doi.org/10.1093/jopart/mum015 raab, j., & kenis, p. (2009). heading toward a society of networks: empirical developments and theoretical challenges. journal of management inquiry, 18(3), 198–210. https://doi.org/10.1177/1056492609337493 complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 https://doi.org/10.1093/ppmgov/gvz011 https://doi.org/10.2307/2392695 https://doi.org/10.1093/jopart/mur011 https://doi.org/10.2307/2393152 https://doi.org/10.1080/10967490903094145 https://doi.org/10.1093/jopart/muv032 https://doi.org/10.1093/jopart/muq061 https://doi.org/10.1111/j.1467-9299.2009.01775.x https://doi.org/10.1093/jopart/mup038 https://doi.org/10.1093/jopart/mun033 https://doi.org/10.1093/ppmgov/gvz005 https://doi.org/10.1093/ppmgov/gvz012 https://doi.org/10.1093/jopart/muu021 https://doi.org/10.2307/258289 https://doi.org/10.1177/0149206307302554 https://doi.org/10.1093/jopart/mum015 https://doi.org/10.1177/1056492609337493 40 university of bamberg press rethemeyer, r. k., & hatmaker, d. m. (2007). network management reconsidered: an inquiry into management of network structures in public sector service provision. journal of public administration research and theory, 18(4), 617–646. https://doi.org/10.1093/jopart/mum027 ring, p. s., & van de ven, a. h. (1994). developmental processes of cooperative interorganizational relationships. the academy of management review, 19(1), 90. https://doi.org/10.2307/258836 saz‐carranza, a., & vernis, a. (2006). the dynamics of public networks: a critique of linear process models. international journal of public sector management, 19(5), 416–427. https://doi.org/10.1108/09513550610677753 scheinert, s., zia, a., koliba, c., & merrill, s. (2017). growing collaborations: forecasting changes in partnership networks using a bottom-up approach. policy and complex systems, 3(1), 3–27. siciliano, m. d., wang, w., & medina, a. (2020). mechanisms of network formation in the public sector: a systematic review of the literature. perspectives on public management and governance, gvaa017. https://doi.org/10.1093/ppmgov/gvaa017 straub, d. m., deeds, b. g., willard, n., castor, j., peralta, l., francisco, v. t., & ellen, j. (2007). partnership selectio n and formation: a case study of developing adolescent health community-researcher partnerships in fifteen u.s. communities. journal of adolescent health, 40(6), 489–498. https://doi.org/10.1016/j.jadohealth.2006.11.136 thorelli, h. b. (1986). networks: between markets and hierarchies. strategic management journal, 7(1), 37–51. https://doi.org/10.1002/smj.4250070105 voets, j., keast, r., & koliba, c. (eds.). (2020). networks and collaboration in the public sector: essential research approaches, methodologies, and analytic tools. routledge press. weber, e. p., & khademian, a. m. (2008). wicked problems, knowledge challenges, and collaborative capacity builders in network settings. public administration review, 68(2), 334–349. williamson, o. e. (1975). the economics of governance: framework and implications. journal of theoretical economics, 140, 195–223. williamson, o. e. (1981). the economics of organization: the transaction cost approach. american journal of sociology, 87(3), 548–577. complexity, governance & networks – vol. 8, no. 1 (2022), p. 25-40 doi: http://dx.doi.org/10.20377/cgn-116 https://doi.org/10.1093/jopart/mum027 https://doi.org/10.2307/258836 https://doi.org/10.1108/09513550610677753 https://doi.org/10.1093/ppmgov/gvaa017 https://doi.org/10.1016/j.jadohealth.2006.11.136 https://doi.org/10.1002/smj.4250070105 what makes a “regime complex”complex? it depends complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 68 university of bamberg press what makes a “regime complex” complex? it depends author: james hollway graduate institute of international and development studies, geneva, switzerland email: james.hollway@graduateinstitute.ch what makes the collections of international institutions or regimes governing various domains—called in the literature regime, institutional, or governance complexes—“complex”? this article examines several conditions for complexity discussed in that literature and finds them necessary but not sufficient. it argues that the sufficient condition is dependence and outlines a framework of increasing levels of synchronic (social/spatial) and diachronic (temporal) dependence. putting dependence at the centre of discussions on regime complexes has four advantages: (1) it is analytically more precise a condition than proliferation or linkage; (2) it orients us toward questions of degree, ‘how complex’, instead of the binary ‘whether complex’; (3) it informs a range of research design and theoretical choices, especially highlighting extra-dyadic dependencies and an underdeveloped temporal dimension; and (4) it arguably reconciles competing uses of the term “complex” in the literature without conflating it with complexity, structure, or topology. keywords: regime complex; institutional proliferation; institutional interlinkages; network dependence; temporal dependence what is complex about a complex? the collections of international institutions or regimes governing various domains are often called “complex”. indeed, alter and raustiala (2018) maintain that “the signature feature of twentyfirst century international cooperation is arguably not the regime but the regime complex”. while some prefer “institutional complexes” (aggarwal, 1998; oberthür and stokke, 2012; zelli and asselt, 2013) or “governance complexes” (underdal, 2010; biermann et al., 2010; hollway and koskinen, 2016) to “regime complexes” (raustiala and victor, 2004; keohane and victor, 2011), all this work faces a similar question:1 what’s complex about a complex? for readers of this journal, the term “complex” may evoke special theoretical ideas such as complex adaptive systems (cas; morçöl, teisman, and gerrits, 2014; teisman and gerrits, 2014). but scholars of international cooperation seem more comfortable invoking these concepts with respect to the system to be governed (i.e. governance of complexity) than empirically applying them to the systems of governance themselves (i.e. complexity of governance) (orsini et al., 2019; biermann et al., 2020). instead, it is more common to see the term “complex” used metaphorically to indicate a set of interacting institutions that represents a particular analytic challenge and thus demands different questions or different approaches. herbert simon (1962, p. 467) once said: “metaphor and analogy can be helpful, or they can be misleading”. this paper therefore seeks to examine more closely how this term is used in the literature to identify what makes a regime complex ‘complex’.2 in this paper, i argue that what makes complexes “complex” and such profound analytic challenges is not institutional multiplicity, diversity, overlap, or even interaction, as often argued or implied in the literature—these are necessary but not sufficient—but dependence. putting dependence at the centre of discussion on regime complexes is useful for four reasons. first, it helps refine analytic tools to concentrate on those features that matter most for “complexifying” domains. for example, it cautions against equating institutional proliferation with complexity; many 1 i thus use them interchangeably here. 2 note that here i am exploring the definitional features common to all complexes, i.e. proliferation, linkages, dependence, rather than structural features that may vary across complexes, e.g. a hierarchical, nested, or clustered topology. mailto:james.hollway@graduateinstitute.ch complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 69 university of bamberg press units may make a domain complicated, but not necessarily complex. second, relating complexity to dependence helps us to think of complexity not in binary terms—is it complex or not?—but rather in terms of degree. some “complexes” (noun) will be more complex (adjective) than others, involving broader or deeper degrees of dependence. for example, are two, three, or more international institutions entangled or interdependent in their behaviour? thinking of complexity in terms of degree demands that we think of theories and methodological approaches that are commensurate to the level of complexity in the system rather than just its presence. third, dependence provokes us to consider complexity dimensionally, opening up different kinds of complexity. dependence can exist not only among units synchronically, but also diachronically through time. relating complexity to dependence rather than the existence of many (different) units emphasises temporal dependencies in a productive way. lastly, dependence can reconcile work that conceives of complexes as a noun, akin to apartment or military complexes, with those that interpret complexes with more cas-lenses. for example, it is where there is broad dependence among apartment units that we might say that there is a “complex” and where the operation of such dependencies can introduce emergence and nonlinearity. the next section outlines my argument for why other features of complexes are only necessary whereas dependence is sufficient. i then follow with a first attempt to outline some orders, levels, or degrees of spatial and temporal dependence, illustrating their relevance to research design on regime complexes with a discussion on how case selection is impacted by each level of dependence. i then elaborate some additional methodological and theoretical implications, and conclude with an outlook on the use of the term complex in the literature. from proliferation to dependence proliferation: multiplicity and diversity a common answer to what makes regime complexes “complex” stems from this literature’s original observation of increasing institutional density. for example, raustiala and victor (2004), in their seminal piece, begin from the observation that international institutions have proliferated in the postwar period and argue that this makes it difficult to “decompose” the complex and study individual institutions.3 concentration on this original observation has produced two related variables being used to operationalize complexity: unit multiplicity and diversity (biermann et al., 2020). multiplicity is implicit in many accounts. for example, orsini, morin and young (2013, p. 27) highlight the complexity borne of institutional densification “as the number of new treaties has grown at an exponential rate and existing intergovernmental organizations have crept into neighboring issue areas”. they argue that a regime complex must consist of at least three elementary institutions. asselt and zelli (2014) also explicitly consider the number of institutions a useful measure of institutional complexity. a similar argument is that it is not the number but the diversity of institutions that makes a system complex. those who focus on the appearance of private and hybrid authority often advocate such a position (e.g. gulbrandsen, 2010; kalfagianni and pattberg, 2013; widerberg, 2016), though this is also broadly consistent with applications of organizational ecology (abbott, green, and keohane, 2016; morin, 2018). however, while proliferation may be a necessary condition for complexity, it is not sufficient. a larger system may make complexity possible but not inevitable. multiple and maybe even diverse units is an important condition but, if these units can be considered to operate independently of one another, the collection of units pose no special analytic challenge. a system with many independent units may still be entirely decomposable. for example, single-celled organisms might proliferate but 3 “decomposability” here references herbert simon’s (1962) classic paper “the architecture of complexity”, in which he argued that, in contrast to the “decomposability” of many physical systems, many social systems are “nearly decomposable” in that there are weak but not negligible interactions among their subsystems. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 70 university of bamberg press even their aggregation alone may not be conceived of as particularly complex. indeed, proliferation may represent more of a methodological opportunity than a challenge. after all, more institutions, say, could simply mean more cases with which to learn more about the institutional strategies that are most effective for different purposes and under different conditions. that is, institutional proliferation is a necessary but itself insufficient condition for institutional complexity. linkages: overlap and interaction another answer stems from early definitions of regime complexes that emphasise institutional linkages (e.g. young, 1996; keohane and victor, 2011) or overlaps (e.g. raustiala and victor, 2004, p. 279). concepts of overlap and interaction both build upon proliferation but go one step further. the literature on overlap highlights how institutional complexity derives from the competing or synergistic effects of institutions that overlap in membership or mandate. alter and meunier (2009) argue that the “increasing density of international regimes has contributed to the proliferation of overlap across agreements, conflicts among international obligations, and confusion regarding what international and bilateral obligations cover an issue”. zelli, möller, and asselt (2017, p. 670) understand institutional complexity as “a diversity of international institutions that legally or functionally overlap in addressing a given issue area of global governance”. a recognition that overlap is a necessary condition for complexity has resulted in a burgeoning literature examining how different types of overlap result in different kinds of outcomes or demand different strategies (hofmann, 2011). but while overlap is a necessary condition for complexity, it is also not sufficient: overlapping institutions could simply have linear, additive effects on their mandates or members, and not interact either statistically or in terms more common to this literature. the literature on interactions goes one step further, arguing that there is not just a common referent, but that it is the institutions themselves interacting to manage overlapping mandates that makes complexes “complex”. raustiala and victor (2004, p. 278) maintain that “the rising density of the international system makes it likely that interactions among regimes will be increasingly common”. axelrod (2014, p. 987) asserts that more institutions increase the probability of “inconsistent legal commitments” that requires more interaction. as with the literature on overlap, a literature on institutional interactions (e.g. oberthür and stokke, 2011) has emerged to identify and explain different types of linkages, such as ideational, strategic, and normative. as such, this perhaps comes closest to identifying what makes a complex “complex”. however, as kim (2019) highlighted, this literature has overwhelmingly sought to disaggregate and decompose larger systems into dyads of institutions so that the linkages could be treated as analytically similar (stokke, 2012; oberthür and stokke, 2011). just as complexes are thought to be more than the sum of their units, they also ought to be considered more than the sum of their dyads, given that they are typically larger than two institutions (orsini, morin, and young, 2013). to be clear: proliferation, diversity, overlap, and interaction are all necessary conditions for a set of regimes or institutions to be complex. proliferation and diversity reference the necessary units for complexity. overlap and interaction reference the necessary linkages among those units for complexity. however, they are also insufficient and more distal criteria of complexity. these linkages still offer only opportunities for inter-institutional influence and do not characterise or describe it. it is not just that there are multiple institutions somehow connected or linked; such a link needs to induce dependence. dependence: spatial and temporal consider simon’s (1962) classic watchmaker parable. two watchmakers are tasked with each constructing a watch of 1000 elementary units in the face of disturbance. however, while one adds each unit in turn and has to contend with everything falling apart upon disturbance, the other complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 71 university of bamberg press constructs successive sets of stable subassemblies that mean their work, upon disturbance, is only destroyed to the last stable subassembly. simon argues that it is the hierarchical layering of only near decomposability in the latter that represents complexity. simon’s parable highlights why the number of units or relations between them are insufficient to characterise complexity. a watch of 1000 elementary units is not necessarily more complex than one of 500, nor is one of 1000 connections necessarily more complex than one of 500 connections. what makes a system more complex is that the units interact such that they depend on one another. changing something in one part affects other parts. for example, reversing a component cog in a watch may make it not work, or work backwards, which may then affect yet further elements. applied to institutional complexity, we might observe how an institution’s membership, policies, or effectiveness might depend on the membership, policies, or effectiveness of other related institutions. or a relationship between two institutions might depend on the presence or absence of relationships between each institution and other (third) institutions or actors. so what is the distinction between linkages and dependence? consider a wristwatch and its band.4 while the two components are linked, there is not really any dependence here. the watch would continue to tell time without the band. in other words, formal linkages can incur no real dependencies. institutionally, an example might be two institutions that observe each other’s meetings for historical reasons without any kind of coordination or adjustment. now consider the watch and a magnet. here there is no formal linkage, but the magnet’s presence can cause the hairspring component to seize, either shortening the active length of the spring and speeding up time or arresting the watch altogether. in other words, dependencies can appear even without observable linkages. institutionally, an example might be institutions that operate in “conscious parallelism” (abbott 2012, p. 583), adjusting mutually despite the absence of any formal linkage. dependence is similar to the concepts of cognitive interaction (gehring and oberthür 2009) or functional interdependence (young 2002, p. 23) raised in the literature. but dependence offers a broader conceptual base in two important regards: it can be extra-dyadic and temporal. while most notions of interaction and linkage have focused tightly on institutional dyads, dependence can encompass broader setups in both degree and kind. first, dependence can expand beyond the dyad. this is important because, if such broader degrees of dependence are present, then decomposing a complex into supposedly constituent dyads can be misleading. institutional dependencies are thus what makes governance so complex, because the presence of dependencies mean we can never do just one thing. second, dependence can be temporal too. so far, cross-sectional spatio-structural dependencies are the most obvious and studied to date in the literature on regime complexes. while temporal dependence has long been a focus in (historical) institutional theory (pierson, 2000) and policy change, this work has had greater focus on how institutional arrangements persist than how novelty arises (see hollway, morin and pauwelyn, 2020), and has not been well integrated into the study of institutional or regime complexes so far (excepting e.g. andersen, 2002; colgan, keohane and van de graaf, 2011; morin, hollway and pauwelyn, 2017). but as morin and gomez-mera argue in orsini et al. (2019, p. 20), the temporal dimension cannot be ignored. dependence is a concept as sensitive to time as (social) space. action reconfigures existing relationships, but the residue of those relationships often remain, making more things depend on one another than before (hollway, 2020). in the following section, i outline a categorisation of sociospatial dependence that incorporates unit and linkage-based dependence, but also extends it to broader dependencies, as well as outlining corollaries for temporal dependence. 4 i thank an anonymous reviewer for this example. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 72 university of bamberg press orders of dependence simon (1962, p. 481) said: “how complex or simple a structure is depends critically upon the way in which we describe it. most of the complex structures found in the world are enormously redundant, and we can use this redundancy to simplify their description. but to use it, to achieve the simplification, we must find the right representation.” various disciplines including computer science have considered ways to characterise complexity, such as by length or class. i propose to borrow from statistics—especially statistical network, time series, and sequence analysis—a formalisation of the notion of dependence to construct a scale of complexity. though there could be various types, forms, and pathways of dependence, i follow simon’s advice here to reduce down and simplify to two kinds or dimensions of dependence: synchronic or spatial dependence and diachronic or temporal dependence. both are outlined here as forms of autocorrelation, in which observations may depend on other observations of the same variable, but this can quite simply be extended to other variables without loss of generality. table 1 below outlines a first attempt at classifying orders or degrees of synchronic and diachronic (in)dependence. for the synchronic or spatial configurations, the orders of dependence are highlighted on a hypothetical network of adjacency between five nodes {i,j,k,l,m}. one might imagine these as institutions identified in a given field or with some potential salience to one another. for the diachronic or temporal configurations, the orders of dependence are illustrated with a network of relationships between time points {t,t−1,t−2,t−3}. for synchronic dependence, we are interested in how a relationship between nodes i and j might depend on the six other ties in this hypothetical network. for diachronic dependence, we are interested in how time point t might depend on earlier time points, which is why the ties can only go back in time. in each figure, dependencies are shown as solid lines, whereas inactive or irrelevant relationships for that order of dependence are still shown, though as dashed lines only. each row represents an increasing order of dependence, from a null of complete independence for institutional observations across space or through time, to higher order configurations of dependence across space or time. so for example, bernoulli independence means a relationship between institution i and institution j is independent of all other institutional linkages, including others between i and j, whereas circuit dependence suggests that that relationship depends on all other local institutional linkages. and whereas temporal noise suggests that the behaviour or effects of an institution(al complex) at time t is independent of any part of its history, juncture dependence suggests that all time points after t − 3 depend on some institutional behaviour at that point. note that synchronic dependence nests lower orders, whereas diachronic dependence involves different types. below i discuss synchronic or spatial dependence first and then temporal, but there is nothing to suggest one is more elementary than the other. together they represent a useful way to think about how complex many social systems are, irrespective of their size, and offer a starting point for measuring relative complexity, a prerequisite for empirical investigations into whether and how this matters. i have included the formal representation of these dependencies in the table for completeness, but have kept the discussion in the text conceptual. i illustrate the consequentiality of successive orders of dependence, of both types, with reference to decisions of case selection where appropriate. in the following section, i outline some further methodological and theoretical implications. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 73 university of bamberg press complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 74 university of bamberg press social, spatial, or synchronic dependence synchronic dependence references observational dependencies that are conditioned on some proximity or distance, whether in physical space (neumayer & plümper, 2010) or social space (hollway et al., 2017; knoke et al., 2021). physical or geographical space is the more commonly understood, with distance measured in meters or miles. social space is constructed by the links between social actors (or institutions), with distance measured in terms of degree (as in six degrees of separation) or path length between nodes. both allow some kind of spatial mapping on the basis of these distances though, with the implication that neighborhoods of adjacent nodes can be identified in that space. in this subsection, i treat spatial and social structural dependencies together, as their cross-sectional quality and emphasis on neighborhoods make them analytically similar for our purposes. moreover, spatial dependencies are often treated as networks to allow more flexible specification of what neighborhood or adjacency means (juhl, 2019). statistical network models have been acknowledged as an improvement over traditional statistical models where researchers recognise the need to take into account or investigate dependencies between observations (e.g. robins, lewis, & wang, 2012). the most elaborate hierarchy of structural dependence is defined in statistical network modeling for exponential random graph models (ergms) (lusher, koskinen, & robins, 2013; amati, lomi, & mira, 2018). ergms express the probability of observing a given network system as a function of subgraphs into which this network can be decomposed. an example of such a configuration is transitivity, known colloquially as “friends of my friends are likely also my friends”. researchers might include such a configuration to improve their inference on some potentially correlated mechanism, such as homophily, or where they theorise that the opportunity structure afforded by shared contacts is consequential for ties between two individuals. such subgraphs express local dependencies in the relationships or interactions between units (pattison & snijders, 2013). for our purposes, what is important is that specifications requiring more complex configurations to obtain a well-fitting model indicate broader dependencies. four orders of structural dependence are defined and employed, from structural independence to circuit dependence. first, ties that appear statistically independent of one another is typical of a bernoulli or erdosrenyi model, which operates as our order zero of dependence complexity. for example, a claim that a system composed of one institution on local mining conditions or a single river basin might rightly be challenged as no complex system at all. this is, in a sense, our null model, and so we would expect to rarely observe their proposition. if an institutional complex were truly bernoulli, then we could decompose “system” into its institutional components and analyse them separately without concern that we would be missing important dependencies. second, ties may depend (only) on other ties between that same dyad. the dyads themselves are independent, but all other interactions between that dyad must be considered and so the system is only decomposable to the dyad level. this includes reciprocating ties. if an institutional complex were “dyad independent”, this could facilitate the kind of examination oberthür and stokke (2011) outline in their work on institutional interaction. dyadic studies have the chief advantage that they can allow for more elaborate consideration of all the different types of interlinkages that are possibly present but are inappropriate if higher levels of dependence are implicated. examples include the interaction between the international labour organisation and the world trade organization (gehring and oberthür, 2009). third, ties may depend on immediately adjacent ties in a form of markovian dependency. this generally means third units are involved. for example, the interaction between two institutions may depend on how they either or both interact with third institutions. if an institutional complex is “markov dependent”, institutions should be examined as at least triads, though following this logic may quickly extend to larger groupings. that is, configurations of multiple interactions become a single dependence structure. a contemporary application of this might be the study of the biodiversity complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 75 university of bamberg press beyond areas of national jurisdiction negotiations (de santo et al. 2019; vadrot 2020). there scholars have been studying the dependencies induced by the introduction of additional institutional concerns to the negotiations: the cbd, cites, unclos, fao, a number of different regional fisheries management organizations, etc. the existing interactions between any two of these institutions will be affected by the introduction of a bbnj institution, highlighting these dependencies. lastly, ties may be partially conditional on ties that are not immediately adjacent but appear in a circuit (snijders et al., 2006). this represents the most complex substructures currently modelled in statistical networks, and yet is quite common in social networks. it reflects how ties between two units may depend on ties between two other units if those pairs of units are somehow connected. for example, the united kingdom acts in the united nations security council influenced by the relationship of the united states to the north atlantic treaty organization, since the us is also a member of the unsc and the uk is also a member of nato. such dependence is likely to be even more prevalent in the study of multilevel governance (e.g. betsill and bulkeley 2006) or when conceiving of collective action problems as not only horizontal and functional but also vertical (e.g. feiock 2013; knoke et al., 2021). this level of dependence may not result in much broader complexes than if only markov dependence is considered, but its existence does introduce further analytic challenges within the complex. temporal or diachronic dependence diachronic dependence references observational dependencies that are conditioned on time. a hierarchy of temporal dependence is less well developed in the statistical literature, partly because temporal dependence has been treated through several different methods, including time series analysis (hamilton, 1994), event history analysis (allison, 1984), and sequence analysis (a. abbott, 1995). here i propose a hierarchy of temporal dependence concerned with how past information impacts expectations for the present that roughly relates to the orders of spatial complexity outlined in the previous subsection. first, a lack of temporal dependence can be characterised as a random walk or white noise in a time series. no information about the present state or an institution or a complex can be gleaned from its past states. in an autocorrelation function, this would show as the absence of any significant correlation with past values at any step removed from the current observation. if past states of an institutional complex hold no information for its present state, then a wholly cross-sectional approach is perfectly appropriate. however, this is likely to be extraordinarily rare—it would suggest that there is absolutely no memory to institutionalised cooperation—which highlights why it is important to push the study of institutional complexity in more historical and dynamic directions. second, observations may depend on the previous observation(s). a trend takes as information the current trajectory, identified in the high autocorrelation with observations lagged by 1 and less autocorrelation with observations at increasing lags. this can also be characterised as a form of markovian dependence, as information relevant to the present is held no more than one step away. depending on how successive lags are correlated, this could also be represented as autoregressive, moving average, or exponentially smoothed. institutional complexes subject to a trend may at least partly represent the continuation of processes already underway, and so even if researchers settle upon a cross-sectional approach, they need to be conscious of which aspects are part of this trend. periodisation is difficult without at least an informative historical account preceding the cross-sectional analysis (isaac and griffin, 1989). better is a temporal analysis that identifies the slope and length of the current trend, and whether it is linear or nonlinear. a common example of this approach is to study difference or change, say in institutional effectiveness (mitchell, 2002): is chemical pollution trending down or up under a regime’s influence? third, observations may depend on previous observations that are not immediately prior but defined at some regular or seasonal interval. seasonal cycles are common in economic systems but complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 76 university of bamberg press appear in political systems too as regular periods of policymaking, chairmanship alternations or rotations, or oscillations in public attention (goetz and meyer-sahling, 2009). somewhat more complex cycles can appear too. for example, electoral cycles may be somewhat irregular or lie out of step with one another, introducing some advantages to the continuity of an institutional complex in terms of the actors they represent as well as challenges. identifying such complexity can pose an empirical challenge however, for it will require a certain maturity for the system to be able to identify such seasonality. but where seasonality does exist, say for example with the chemical pollution example above, it will be important to adjust for it. fourth, observations may depend on a previous observation having happened at some point in the past, and all later events will depend somewhat on the realisation of this juncture. historical institutionalism demands this degree of temporal dependence, as significant events in the past can switch trajectories onto different paths that then get “locked in” through increasing returns to that switching event (pierson 2000; see also rixen and viola 2014; hanrieder 2015). this might be the introduction of a new institution to the complex, or a policy change within one of the institutions (that then percolate through the system according to markovian or circuit dependencies outlined above). the introduction of bbnj will irrevocably alter other interinstitutional relationships, for example. repetition may also be part of this order of temporal dependence: it may matter if an institution has ever done something, or the sequence of events (thornton and ocasio 2008), not how long ago it was. for example, an institution that has been seen to be effective in a crisis may enjoy greater legitimacy, enabling strategies and policies that would have otherwise lain out of reach. or an institution that is strongly associated with a particular blunder, for example the league of nations, may never recover, no matter how much time elapses. if junctures are present, this can imply two things for research designs. if the goal is to identify the effectiveness of any policy associated with the juncture point, then the design will need to cover a period both prior and following the critical event to identify the effects that event had. but if the goal is simply to forecast, then only the complex’s history after the juncture point will be salient. of course, some degree of dependence will almost always be present in sociopolitical systems. because of this, asking whether a system or setting is a “complex” or not is misleading. the more consequential and productive question is which kinds and degrees of dependence are present. we have seen in this section how the presence of particular orders of dependence have implications for case selection. in the following section, i outline some further methodological and theoretical implications. methodological and theoretical implications taking a perspective on complexity that is conscious of the scope of (structural or temporal) dependency has important methodological and theoretical implications. complexity is consequential because dependence frustrates our ability to make inferences about institutional influences and impacts independently of actors and institutions related through space and time. in introducing spatial and temporal orders of complexity, i have already outlined how case selection might be impacted by the maximum level of dependence present in a system. indeed, characterising complexity as dependence highlights why the biggest challenge in studying regime complexes is where the appropriate analytic boundary lies (orsini, morin, and young, 2013). but here i also outline some other aspects of research design affected by the order of complexity present in a set of institutions: sampling, prediction, replication, and theory-building. first, a key part of traditional statistics and its assumption of independent observations is sampling. simple random sampling is the gold standard because it provides an unbiased way of surveying a population. this is why koremenos (2013) employs it to support generalisable statements about a population of international institutions. however, dependence impacts the logic of such a strategy. random sampling excises observations from their context in the population. if institutions are, say, markov dependent, then observing an institution at random from that population will mean complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 77 university of bamberg press missing other (actors and) institutions on which it is dependent, introducing bias another way. each missing observation thus represents n−1 missing dyadic observations. this drives us towards the collection and development of more purposive and comprehensive populations of international institutions in different domains. fortunately, institutional complexes are comparatively small systems and information about constitutive relations if not interactions are often of public record, encouraging at least the ambition for comprehensive datasets on international institutions. another aspect of traditional statistics impacted by dependence and complexity is generalisability. a recognition that higher orders of complexity exist raises the questions: to what we are generalising? what is the population? while explicit transposition of insights from institutional complex to institutional complex is rare in contemporary scholarship, scholars often implicitly suggest that their findings in one domain are transportable to others. higher orders of spatial dependency highlight how it is not simply the parts that must be commensurate for translation, but that there may be different types of markovian or circuit dependencies in play without correspondence in other settings. moreover, higher orders of temporal dependency highlight how baseline trends may be stronger in one regime than another, or that particular junctures in one regime may be absent or framed in a different way in another. one option forward is to avoid generalising variables and instead seek lessons on the level of mechanisms (tilly, 2001; hedström & ylikoski, 2010). these mechanisms can concatenate in complex ways to generate quite different, even idiosyncratic systems in time, but also provide portable insights into complex phenomena (epstein, 1999). i would argue that particularly fruitful “mechanisms” for the sociopolitical world where we are ever concerned with agency in the face of complexity are those that might explain actors’ choices (hollway, 2020; see also block et al., 2018), for they offer us a way of understanding and acting in a social world. prediction, often touted as the gold standard of statistical models, is another aspect dependence touches upon. epstein (2008) enumerates sixteen reasons other than prediction to build models. these include explanation and theory-testing (quite distinct from prediction), but also the illumination of micro-macro linkages, the suggestion of dynamical analogies, prompting new questions and guiding data collection or case selection, for pedagogical and public education purposes, and to discipline the policy dialogue by identifying risks, efficiencies, and trade-offs. we can and should build statistical models of institutional complexes for all these purposes, to build scenarios and inform policy decisions, to understand the past and our future options. but prediction is not just only one among many purposes of statistical models, it can sometimes be an unhelpful metric in the context of complexity (orsini et al., 2019). after all, spatial and temporal dependencies may be such that prediction is either trivial or impossible. for example, we might predict an institutions policy from what it had implemented yesterday (trend dependence), but this does not help us understand or act in the world. but predicting an institution’s membership or policy in just ten years nonetheless demands courageous assumptions about (a lack of) spatial and temporal dependence (cederman and weidmann 2017). instead, we might think about fit in terms of whether simulations from our models replicate core dependencies (block et al., 2018). lastly, a concentration on dependence also has theoretical implications. it highlights the importance of studying the conditions under which proliferation leads to linkage (e.g. hofmann, 2019), and linkage leads to different degrees of dependence (e.g. oberthür & gehring, 2006). but it also highlights how we also need to further theorise how institutions and regimes depend on one another through time. the literature to date has been overwhelmingly cross-sectional, and yet temporal dependencies of the sort outlined here are surely as prevalent. this means bringing the sort of work prevalent in regime complexity and historical institutionalism closer together. thinking through how spatially and temporally dependent a particular governance domain is points us not only to methodological adaptations that must be made, but also (as it always has for the regime complexity literature) where the interesting questions are. introducing more temporal and spatial dimensions is both data intense and analytically intensive. on the data side, accepting the presence of (higher levels of) dependence means thinking complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 78 university of bamberg press carefully about the boundaries of regime complexes, what data is available, and what further data would need to be collected so as not to miss important dependencies thought to be present. on the analysis side, dependencies demand the use of methods, for example network and sequence analysis, that can not only account for dependencies in their evaluation of uncertainty surrounding estimates, but present such dependencies in the estimates themselves as representative of mechanisms of theoretical interest. the implications of recognising dependence in regime complexes is thus demanding. the alternative would be to act as if such dependencies were not present or were irrelevant. but their presence and relevance, i have argued, is the premise of regime complexity. outlook i have argued here that the feature of institutional complexes that makes them complex in the sense of analytically challenging is not the number or diversity of institutions or linkages between them, though these are necessary conditions, but rather that these linkages induce dependencies that make it difficult to analytically decompose the institutional complex or act within it. i then elaborated this notion of dependence into classes or orders of spatial and temporal dependence and related these to particular demands in terms of the chief research design question of institutional complexes: boundary specification. i have concluded by discussing some further theoretical and methodological implications of a concentration on dependence, including brief considerations for what this means for sampling international cooperation, the possibilities of replication, generalization, and prediction, as well as how this pushes the literature to consider temporal dependencies in institutional complexes as much as spatial dependencies. characterising complexity as dependence is consistent with much of the current literature. for example, orsini, morin, and young (2013, p. 27) contrast current trends with “earlier times” in which “most intergovernmental organizations (igos) and multilateral treaties were relatively independent from one another”. moreover, identifying higher-order dependence as the sufficient condition for a complex is consistent both with those that use the term complex as noun (e.g. military industrial complex) and those who use it as adjective (e.g. complex systems). for example, eisenhower’s 1951 speech highlighted how legislators, the military, and the defence industry were becoming increasingly dependent upon one another and can therefore represent some identifiable aggregate that would lead to bias if we were to analytically decompose. moreover, features of complex adaptive systems such as feedbacks, nonlinearity, and learning all represent dependence. a concentration on which orders of dependence are prevalent could therefore offer a way to square the circle, as it were, between these literatures. is complexity and dependence the same thing then? i think it behooves us to keep these conceptually distinct. this means we can highlight their relationship and identify how institutional dependence is a sufficient condition for but also a response to complexity (see also le prestre, 2017). only by incorporating temporal dependence more explicitly can we parse out when and how dependence affects complexity and vice versa (pantzerhielm et al. 2020). the dependence hierarchy outlined here allows us to recognise that higher-order dependencies may be more prevalent in some domains than others, and reflect on the research design and theoretical implications thereof. likewise, this discussion highlights how questions of complexity and structure or topology differ and should not be treated as interchangeable, at least if we aim for more precise terminology in this field. lastly, a conceptual toolkit such as that outlined here allows us to formalise the relationship between dependence and complexity, supporting more explicit research designs that correspond to the theoretical assumptions of the literature, and in turn promote the identification of further research frontiers and their properties. most prominently, identifying dependence at the root of this literature highlights the relative paucity of work on temporal dependence and regime complex dynamics. the challenge will be to develop this theoretically: yes it depends, but on what? complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 79 university of bamberg press acknowledgements i would like to thank karsten donnay, stephanie hofmann, lauren hollway, johan koskinen, jean-frédéric morin, and the participants of the workshop on “theory and methods on institutional complexity in global governance” at the käte hamburger kolleg / centre for global cooperation research for comments and suggestions on earlier versions of this piece. i also thank two anonymous reviewers and the editors of the special issue for their valuable comments and suggestions. references abbott, andrew (1995). “sequence analysis: new methods for old ideas”. in: annual review of sociology 21, pp. 93-113. abbott, kenneth w (2012). “the transnational regime complex for climate change.” in: environment and planning c 30.4, pp. 571–90. abbott, kenneth w, jessica f green, and robert o keohane (2016). “organizational ecology and institutional change in global governance”. in: international organization 70.2, pp. 247–277. aggarwal, vinod k, ed. (1998). institutional designs for a complex world: bargaining, linkages & nesting. ithaca, nj: cornell university press. allison, paul d (1984). event history analysis. regression for longitudinal event data. london: sage. alter, karen j and sophie meunier (2009). “the politics of international regime complexity”. in: perspectives on politics 7.1, pp. 13–24. alter, karen j and kal raustiala (2018). “the rise of international regime complexity”. in: annual review of law and social science 14.1, pp. 329–349. amati, viviana, alessandro lomi, and antonietta mira (2018). “social network modeling”. in: annual review of statistics and its application 5.1, pp. 343–369. andersen, regine (2002). “the time dimension in international regime interplay.” in: global environmental politics 2.3, pp. 98–117. asselt, harro van and fariborz zelli (2014). “connect the dots: managing the fragmentation of global climate governance”. in: environmental economics and policy studies 16.2, pp. 137– 155. axelrod, mark (2014). “clash of the treaties: responding to institutional interplay in european community-chile swordfish negotiations”. in: european journal of international relations 20.4, pp. 987–1013. betsill, michele m, and harriet bulkeley (2006). “cities and the multilevel governance of global climate change.” in: global governance 12, pp. 141–59. biermann, frank, fariborz zelli, philipp h pattberg, and harro van asselt (2010). “the architecture of global climate governance: setting the stage”. in: global climate governance beyond 2012: architecture, agency and adaptation. cambridge: cambridge university press, pp. 15–24. bierman frank, rakhyun e kim, kenneth w abbott, james hollway, ronald b mitchell, and michelle scobie. (2020). “ taking stock and moving forward.” pp. 299–321 in architectures of earth system governance, edited by f. biermann and r. e. kim. cambridge: cambridge university press. block, per, johan h. koskinen, james hollway, christian e. g. steglich, and christoph stadtfeld (2018). “change we can believe in: comparing longitudinal network models on consistency, interpretability and predictive power.” social networks 52: 180–91. cederman, lars-erik and nils b. weidmann (2017). “predicting armed conflict: time to adjust our expectations?” science 355: 474–76. colgan, jeff d, robert owen keohane, and thijs van de graaf (2011). “punctuated equilibrium in the energy regime complex”. in: the review of international organizations 7.2, pp. 117– 143. de santo, e, a asgeirsdottir , a barros-platiau, f biermann, j dryzek, lr gonçalves, re kim, e mendenhall, r mitchell, e nyman, m scobie, k sun, r tiller, dg webster, and o young (2019). “protecting biodiversity in areas beyond national jurisdiction: an earth system governance perspective.” in: earth system governance 2:100029. epstein, joshua m (1999). “agent-based computational models and generative social science”. in: complexity 4.5, pp.41–60. – (2008). “why model?” in: journal of artificial societies and social simulation 11.4, p. 12. feiock, richard c (2013). “the institutional collective action framework.” in: policy studies journal 41.3, pp. 397–425. gehring, thomas and sebastian oberthür (2009). “the causal mechanisms of interaction between international institutions”. in: european journal of international relations 15.1, pp. 125–156. goetz, klaus h, and jan-hinrik meyer-sahling (2009). “political time in the eu: dimensions, perspectives, theories”. in: journal of european public policy 16.2, pp. 180–201. gulbrandsen, lars h (2010). transnational environmental governance. the emergence and effects of the certification of forests and fisheries. edward elgar publishing. haftel, yoram z and stephanie c hofmann (2019). “rivalry and overlap: why regional economic organizations encroach on security organizations”. in: journal of conflict resolution 63.9, pp. 2180–2206. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 80 university of bamberg press hamilton, james d (1994). time series analysis. princeton: princeton university press. hanrieder, tine (2015). “the path-dependent design of international organizations: federalism in the world health organization.” in: european journal of international relations 21.1, pp. 215–39. hedström, peter and petri ylikoski (2010). “causal mechanisms in the social sciences”. in: annual review of sociology 36.1, pp. 49–67. hofmann, stephanie c (2011). “why institutional overlap matters: csdp in the european security architecture”. in: jcms: journal of common market studies 49.1, pp. 101–120. – (2019). “the politics of overlapping organizations: hostage-taking, forum-shopping and brokering”. in: journal of european public policy 26.6, pp. 883–905. hollway, james (2020). “network embeddedness and the rate of water cooperation and conflict.” pp. 87–113 in networks in water governance, edited by manuel fischer and karin ingold. cham: palgrave. hollway, james and johan h koskinen (2016). “multilevel embeddedness: the case of the global fisheries governance complex.” social networks 44, pp. 281–94. hollway, james, alessandro lomi, francesca pallotti, and christoph stadtfeld (2017). “multilevel social spaces: the network dynamics of organizational fields.” network science 5.2, pp. 187–212. hollway, james, jean-frédéric morin, and joost pauwelyn (2020). “structural conditions for novelty: the introduction of new environmental clauses to the trade regime complex.” international environmental agreements: politics, law and economics 20.1, pp. 61–83. isaac, larry w. and larry j. griffin (1989). “ahistoricism in time-series analyses of historical process: critique, redirection, and illustrations from u.s. labor history.” american sociological review 54.6, pp. 873–90. juhl, sebastian (2019). “the sensitivity of spatial regression models to network misspecification”. political analysis 28(1): 1–19. kalfagianni, agni and philipp h pattberg (2013). “fishing in muddy waters: exploring the conditions for effective governance of fisheries and aquaculture”. in: marine policy 38, pp. 124– 132. keohane, robert owen and david g victor (2011). “the regime complex for climate change”. in: perspectives on politics 9.1, pp. 7–23. kim, rakhyun e (2019). “is global governance fragmented, polycentric, or complex? the state of the art of the network approach”. in: international studies review 30.4, pp. 1–29. knoke, david, mario diani, james hollway, and dimitris c christopoulos (2021). multimodal political networks. cambridge: cambridge university press. koremenos, barbara (2013). “the continent of international law”. in: journal of conflict resolution 57.4, pp. 653–681. le prestre, philippe g (2017). global ecopolitics revisited: towards a complex governance of global environmental problems. london: routledge. lusher, dean, johan h koskinen, and garry l robins, eds. (2013). exponential random graph models for social networks: theory, methods and applications. cambridge, uk: cambridge university press. mitchell, ronald b (2002). “a quantitative approach to evaluating international environmental regimes”. in: global environmental politics 2.4, pp. 58–83. morçöl, göktug, geert teisman, and lasse gerrits (2014). “complexity, governance, and networks: introduction to the inaugural issue”. in: complexity, governance & networks 1.1, pp. 1–4. morin, jean-frédéric, joost pauwelyn, and james hollway (2017). “the trade regime as a complex adaptive system: exploration nd exploitation of environmental norms in trade agreements.” journal of international economic law 20.2, pp. 365–90. morin, jean-frédéric (2018). “concentration despite competition: the organizational ecology of technical assistance providers”. in: the review of international organizations 70.2, pp. 1– 33. neumayer, eric and thomas plümper (2010). “spatial effects in dyadic data”. in: international organization 64.1, pp. 145– 166. oberthür, sebastian and thomas gehring (2006). “institutional interaction in global environmental governance: the case of the cartagena protocol and the world trade organization”. in: global environmental politics 6.2, pp. 1–32. oberthür, sebastian and olav schram stokke, eds. (2011). managing institutional complexity: regime interplay and global environmental change. cambridge, ma: the mit press. orsini, amandine, philippe g le prestre, peter m haas, malte brosig, philipp h pattberg, oscar widerberg, laura gomez mera, jean-frédéric morin, neil e harrison, robert geyer, david chandler, and chandl (2019). “forum: complex systems and international governance”. in: international studies review, pp. 1–31. orsini, amandine, jean-frédéric morin, and oran r young (2013). “regime complexes: a buzz, a boom, or a boost for global governance?” in: global governance 19, pp. 27–39. laura, pantzerhielm, holzscheiter anna, and bahr thurid (2020). “governing effectively in a complex world? how metagovernance norms and changing repertoires of knowledge shape international organization discourses on institutional order in global health.” in: cambridge review of international affairs. pattison, philippa e and tom a b snijders (2013). “modeling social networks: next steps”. in: exponential random graph models for social networks: theory, methods, and applications. ed. by dean lusher, johan h koskinen, and garry l robins. cambridge, uk: cambridge university press, pp. 287–301. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 68-81 doi: http://dx.doi.org/10.20377/cgn-100 81 university of bamberg press pierson, paul (2000). “the limits of design: explaining institutional origins and change”. in: governance: an international journal of policy, administration, and institutions 13.4, pp. 475– 499. raustiala, kal and david g victor (2004). “the regime complex for plant genetic resources”. in: international organization 58, pp. 277–309. rixen, thomas, and lora anne viola (2014). “putting path dependence in its place: toward a taxonomy of institutional change.” in: journal of theoretical politics 27.2, pp. 301–23. robins, garry l, jenny m lewis, and peng wang (2012). “statistical network analysis for analyzing policy networks”. in: policy studies journal 40.3, pp. 375–401. simon, herbert alexander (1962). “the architecture of complexity”. in: proceedings of the american philosophical society 106.6, pp. 467–482. snijders, tom a b, philippa e pattison, garry l robins, and mark s handcock (2006). “new specifications for exponential random graph models”. in: sociological methodology 36.1, pp. 99–153. stokke, olav schram (2012). disaggregating international regimes: a new approach to evaluation and comparison. cambridge, ma: mit press. teisman, geert and lasse gerrits (2014). “the emergence of complexity in the art and science of governance”. in: complexity, governance & networks, pp. 17–28. 24 thornton, patricia h, and william ocasio (2008). “institutional logics.” in: the sage handbook of organizational institutionalism, edited by royston greenwood, christine oliver, thomas b lawrence, and renate e meyer, pp. 99– 128. london: sage. tilly, charles (2001). “mechanisms in political processes”. in: annual review of political science 4.1, pp. 21–41. underdal, arild (2010). “complexity and challenges of long-term environmental governance”. in: global environmental change 20.3, pp. 386–393. vadrot, alice b. m. 2020. “multilateralism as a ‘site’ of struggle over environmental knowledge: the north-south divide.” critical policy studies 14(2): 233–45. widerberg, oscar (2016). “mapping institutions and actors in global climate governance: a network approach”. in: environmental politics and governance in the anthropocene. ed. by philipp h pattberg and fariborz zelli, pp. 95–116. young, oran r (1996). “institutional linkages in international society: polar perspectives”. in: global governance 2, pp. 1– 24. young, oran r (2002). the institutional dimensions of environmental change: fit, interplay, and scale. cambridge, ma: mit press. zelli, fariborz and harro van asselt (2013). “introduction: the institutional fragmentation of global environmental governance: causes, consequences, and responses”. in: global environmental politics 13.3, pp. 1–13. zelli, fariborz, ina möller, and harro van asselt (2017). “institutional complexity and private authority in global climate governance: the cases of climate engineering, redd+ and short-lived climate pollutants”. in: environmental politics 26.4, pp. 669–693. global governance in complex times: exploring new concepts and theories on institutional complexity complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: http://dx.doi.org/10.20377/cgn-110 1 university of bamberg press global governance in complex times: exploring new concepts and theories on institutional complexity authors: fariborz zelli, lasse gerrits, ina möller lund university, sweden; erasmus university rotterdam, institute for housing and urban development studies, netherlands; wageningen university, netherlands email: fariborz.zelli@svet.lu.se this article, and the special issue it introduces, seek to contribute to the emerging and much-needed dialogue between the study of global governance and the study of social complexity. we hold that, while there is wide acceptance that global governance is becoming increasingly complex, studying this complexity still faces significant challenges in terms of concepts, theory, and methodology. the article outlines why that dialogue is needed, and how the complexity sciences can help us address some of these challenges. it then introduces key questions central to such an integrated research programme, for instance: under what conditions can a global governance system be regarded as complex? which methods can help us recognize and assess patterns of stability, iteration, and change in global governance? how can a theory-driven analysis take into account that complexity may influence spaces for political agency, i.e. that it may alter key aspects of legitimacy, accountability, transparency, technocracy, and power and ultimately the strategical options of certain actors? finally, the article looks ahead to the special issue and summarizes how the authors contribute crucial conceptual, theoretical, and methodical ideas for addressing these and other questions. keywords: institutional complexity; complexity theory; global governance; interdisciplinarity; research agenda introduction global governance can be defined as “all coexisting forms of collective steering of social affairs, by public and private actors, that directly or in their repercussions, transcend national frontiers” (zelli, 2018). following this broad understanding, global governance today is increasingly characterized by a proliferation of actors and relationships, including intergovernmental and transnational institutions, technological artefacts, information flows, intergovernmental treaties, and trade/financial lending dependencies. a dense network of connections between these elements across scales and spheres of authority has turned global governance increasingly complex. although the importance of this development is widely recognized, studying this complexity still poses significant challenges. different scholarships, e.g. on international relations, international law, and institutional economics, have addressed the phenomenon over the past decades, starting from work on interlocking organizations and polycentricity in the 1960s (e.g. galaskiewicz, 1979; v. ostrom & e. ostrom, 1965; cf. aligica & tarko, 2012) and has since led to a series of research programmes around overlapping concepts like institutional interlinkages, fragmentation, or regime complexes (e.g. oberthür & gehring, 2006; biermann et al., 2009; raustiala & victor, 2004). these different literatures have broken important ground on conceptualizing and theorizing the complexity of governing global affairs. yet, as we argue in this introduction to our special issue, they have almost exclusively done so by revisiting and adapting theories from their own turfs, such as organizational and institutional theories from law and the social sciences. while these important literatures set out to better understand complexity in global governance, they emphasized existing governance theories while not (yet) engaging with theories dedicated to complexity itself. as a consequence, while there seems to be a wide acceptance that global governance is becoming increasingly complex, studying this complexity still faces significant challenges in terms of concepts, theory and methodology. it is against this backdrop that this special issue seeks to contribute to the emerging and muchneeded dialogue between the study of global governance and the study of social complexity. to this mailto:fariborz.zelli@svet.lu.se complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: 2 university of bamberg press end, the special issue brings together scholars from research traditions as different as international relations, complexity sciences, institutional economics, and network analysis. they contribute short think-pieces that point towards new approaches in research as well as full-fledged research articles that use ideas from social complexity research to answer questions about global governance. this introduction will first outline why that dialogue is needed, and how the complexity sciences speak to conceptual, theoretical, and methodical endeavours that may be employed for studying the complexity of global governance. next, we introduce the key questions central to such an integrated research programme. in a third step, we summarize how the authors in this special issue contribute crucial conceptual, theoretical, and methodical ideas for addressing some of these questions. rationale for a stronger cross-disciplinary collaboration considerable conceptual and theoretical work by scholars of international relations, international law, and institutional economics has been dedicated to institutional complexity. “international institutions,” be they intergovernmental, non-governmental, or a mix thereof, are understood as “persistent and connected sets of rules (formal and informal) that prescribe behavioural roles, constrain activity, and shape expectations” (keohane, 1989, p. 3). in this interpretation, the “international institution” is a generic concept which covers both international organizations – that have the capacity to act , e.g. to assess the compliance with rules – and international regimes – that represent treaties or sets of treaties (ibid.). the themes of the aforementioned early literatures on polycentricity of institutions and interlocking organizations (aldrich, 1979; galaskiewicz, 1979; guetzkow, 1966; pfeffer & salancik, 1978; s. schmidt & kochan, 1977; polanyi & prosch, 1975; v. ostrom & e. ostrom, 1965; cf. mcginnis, 1999; aligica & tarko, 2012) demonstrate awareness of complexity in the global realm. the link between institutional research and complexity spurred some of the foremost advances in institutional research, especially in the field of international relations. in their pioneering work on neoliberal institutionalism, robert keohane and joseph nye (1977) argued that complex interdependence, both in terms of content (complexity of the problem) and structure (complexity of the institutional landscape), transforms international politics and chances for cooperation. since the 1990s the notion of “global governance” has become linked to the ascent of new, transnational types of institutions, which, in turn, represent a large diversity of norms and interests and add a further dimension of complexity (rosenau & czempiel, 1992). over the last twenty years then, scholars directed more attention toward coordination problems between institutions, and their competition for regulatory primacy. complexity in this research programme was not so much seen as a condition for a thriving and organically growing landscape of global governance, but as a property or consequence of, as well as a challenge to, global governance. one research tradition scrutinized dyadic overlaps between two or more individual institutions, e.g. between the unfccc and the wto on questions of emissions trading and border adjustments (oberthür & gehring, 2006; zelli, 2010). this research perspective includes literatures on interorganizationalism (cf. r. biermann & koops, 2017) regime interaction and global constitutionalism (since the 1990s, esp. herr & chia, 1995; chambers, 1998, 2001, 2008; young, 1996, 2002; oberthür & gehring, 2006; stokke, 2012; oberthür & stokke, 2011; selin & vandeveer, 2003; faude & fuss, 2020). other research programmes looked at complexity among institutions of an entire policy field. arguably the most influential of these programmes addresses “regime complexes” for particular issue areas like climate change or plant genetic resources (keohane & victor, 2011; orsini et al. 2013; raustiala & victor, 2004), with similar studies conducted under labels such as institutional fragmentation, polyarchy, contested multilateralism, or the nexus approach (cf. faude & parizek, 2020; morse & keohane, 2014; zelli, 2011a,b; zelli et al., 2010, 2020). to explain the degree and shape of institutional complexity in several global governance domains, scholars adapted various existing http://dx.doi.org/10.20377/cgn-110 complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: 3 university of bamberg press theories of international relations, such as different strands of institutionalism (e.g. stokke, 2012; van de graaf, 2013; zelli et al., 2013), sociological differentiation theory (zürn & faude, 2013), and problem-structuralism (zelli et al., 2017). the growing research efforts around such questions led to today’s consensus that a thorough understanding of an intergovernmental or transnational institution is not possible without taking its wider governance environment into account. notwithstanding these achievements, the more scholars brought complexity into their institutional research, the more they came to recognize the considerable need for stronger theoretical foundation, conceptual clarity, cross-disciplinary fertilization, and systematic empirical analyses. recent developments in global governance and the need for their more comprehensive explanation fuelled such calls, from the resistance to scientific knowledge on environmental issues and the reignition of trade wars to brexit and the deterioration of established multilateral institutions (geyer, 2003; haynes, 2015; jervis, 1997; kavalski, 2007; keohane & victor, 2011; room, 2015; zürn & faude, 2013). to address these gaps in a comprehensive manner, scholars on complex global governance today should, we argue, make further efforts to take on board complexity-related tenets and work with scholars from other disciplines to advance their theoretical thinking and methodical repertoire – in particular for embracing major properties of complexity such as uncertainty and unpredictability (gerrits, 2012; jervis, 1997; reed & harvey, 1992). the benefits of integrating such concepts and methods in research on international institutions have been demonstrated by pioneering studies such as robert axelrod's (1986) evolutionary model of norms dynamics or thomas schelling's (1971) agent-based model of segregation mechanisms. these are early examples of how social scientists have tried to address the complex nature of social interactions in theory and method. more recently, abbott et al. (2016) showed this for organizational ecology theories, in their focus on density, resources and niches to hypothesize trajectories of institutional populations. likewise, social network analysis (sna) has become popular among global governance scholars, offering transparent and replicable measures to classify network structures and to identify structurally privileged actors within them (knoke et al., 2021). closely related to sna, statistical network modelling is increasingly used to investigate how changes in complex multi-level governance systems are produced by concatenations of social mechanisms (hollway et al., 2020; milewicz et al., 2018). in a similar vein, actor-network theory was brought into global governance research to examine the mutual constitution of networks and technological developments in the international society (spaargaren, 2011). evolutionary psychology, finally, has been employed in several analyses to understand the heuristics through which political actors navigate their complex environment (gerrits, 2012; gerrits & marks, 2017; wagenaar, 2007). central to these examples of complexity-embracing studies in politics and governance is the recognition across disciplines that the conventional thinking that discerns between dependent and independent variables, and that builds on a linear understanding of causation, is considerably limited in its explanatory scope (mcevoy & richards, 2006). in turn, a non-linear approach that embraces the complex ways in which variables relate in order to explain change and stasis is more fitting for the study of the types of problems that governance scholars are facing today. building on these efforts, we believe it is time to integrate insights from the complexity sciences much more strongly in global governance. “complexity sciences” is an umbrella term for a whole set of theories and methods that starts from the idea that interactions between actors (of any type) can create very complex patterns, such as institutions. as a suite of ideas rather than a unified theory, the term covers approaches as different as, e.g. system dynamics modelling, cluster analysis, fractal analysis, or dynamic pattern synthesis (mitchell, 2009; reed & harvey, 1992). as jervis (1997) suggested two decades ago, complexity sciences could help global governance research to tackle some of the fundamental conceptual, methodological, and theoretical challenges of the discipline. most of this great potential, however, is still unexploited. http://dx.doi.org/10.20377/cgn-110 4 university of bamberg press complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: questions and challenges for a research programme on global complexity this special issue follows from a three-day interdisciplinary workshop on complexity in global governance, held at the centre for global cooperation research / käte hamburger kolleg in duisburg, germany, in late 2017. we are highly indebted to the centre for their logistical, financial, and intellectual support in setting up that workshop. the approximately 30 participants, including many prominent researchers from the fields of international relations, public policy, and political sociology, identified a series of key questions and challenges that they encounter in their own day-to-day research. in response to these questions and challenges, the participants outlined the elements of a comprehensive and interdisciplinary research programme for studying complexity. the purpose of this research agenda is thus not to re-invent the wheel, but to begin a dialogue among disciplines, in which the pressing issues of contemporary politics meets the analytical tools offered by the complexity sciences. we summarize this research agenda along the workshop’s original themes of concepts, methods and theories, with the triple purpose to, first, further elaborate on our rationale for a stronger cross-disciplinary dialogue and collaboration on the complexity of global governance; second, to pave the way for the contributions to this special issue and show which part of this research agenda they help addressing; and third, to invite further collaborative endeavours that pursue this research agenda and its vital and pressing questions. conceptual challenges the conceptual part of the research agenda comprises crucial questions not only in terms of which phenomena need definitions and operationalizing, but also from which angles such efforts could be taken. which disciplines could collaborate in a research agenda on complexity in global governance? political science and international relations scholars certainly need to keep playing a key role as their research questions – and related concepts, methods, and theories – are at the heart of this research agenda. they include the forms and processes of governance complexity, the roles of power, norms, discourses, or knowledge in causing and proliferating this complexity, the implications of complexity for questions of democracy, transparency, legitimacy, accountability, or effectiveness, and, ultimately, the political responses that are either needed or feasible (zelli & van asselt, 2013). yet, research efforts on these topics need crucial input from other disciplinary angles. in addition to the complexity approaches mentioned in the previous section, this includes – without being exhaustive – other social and behavioural disciplines like sociology, psychology, social anthropology, human geography, and history. these disciplines provide invaluable perspectives and concepts to describe and analyze the evolution, shape and consequences of social networks, perceptions and behavioural patterns in complex contexts, and the impact of social relationships across levels and spaces. likewise, a series of natural sciences can make vital contributions to the research on complexity in global governance, such as biology, ecology, or physics, e.g. on the implications of complex systems, cybernetics, keystone actors, steering under conditions of complexity, and scenario-building for the development of complex systems. learning from these disciplines and their concepts, we argue, implies taking into account crucial properties of complex systems that these disciplines established. this fundamental aspect is still considerably understudied in key global governance disciplines like international relations. complex global governance systems have been mostly researched as complicated systems, by decomposing them into individual institutions, processes, or interactions (cudworth & hobden, 2011). this reductionist perspective has missed fundamental properties of complexity, such as embeddedness and emergence, non-linearity and feedback loops, unpredictability and uncertainty, and self-organization and adaptability. in this special issue, and for a wider research agenda, we stress the need to analyze global governance as a complex system that is more than the sum of its parts. this appeal echoes earlier calls for a fifth debate on complexity in international relations research (geyer, 2003; kavalski, 2007), and http://dx.doi.org/10.20377/cgn-110 complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: 5 university of bamberg press parallels contemporary efforts to integrate complexity science thinking into global governance research (le prestre, 2017, orsini et al., 2019). the contributors to this issue challenge the critique that these calls and efforts have met and demonstrate that a complexity lens does not ontologically exclude midrange theories for specific and timely research questions (gunitsky, 2013; pollitt, 2009). based on such an overarching understanding, a series of conceptual questions on how to identify and assess complexity needs to be addressed from different disciplinary angles, such as: − under what conditions can a global governance system be regarded as complex, and what are its characteristics? − what constitutes the system’s boundaries, and who gets to draw them? − what new ways are there to assess processes in and of complex systems, such as their performances, transformations, self-organization and adaptability? to be clear, these questions have already been addressed in the complexity sciences, but in generic terms that require translation into the realm of global governance. a common vocabulary is needed to avoid a proliferation of parallel terms and a conceptual and theoretical lock-in that characterized some of the abovementioned research traditions on polycentricity, interlinkages, fragmentation, and regime complexes. methodological challenges one core question from our workshop is whether complex systems need different research methods than those that are common in global governance research. put differently: does the notion of complex global governance, that is seeing a governance system as more than the sum of its parts, generate research questions that require different methodological approaches? related to this are questions about limitations in studying such systems. compared to complicated systems, efforts towards modelling complex systems, for instance, require a stronger focus on dynamics and uncertainties, and ultimately may have to accept that predictions are much harder to come by. by the same token, a particular challenge for global governance scholars that choose to engage with the complexity sciences is the steep learning curve needed to grasp their methods and the seeming inaccessibility of their analyses. we hold that it is worth working on this bridge from both ends. a creative and pragmatic approach to meet this challenge is for both camps to consider and deploy methods that match the complex nature of global governance systems in various regards: − we need methods that help determine analytical boundaries of complex governance systems, based on the criteria for system borders, properties, and interdependences that novel concepts of complexity provide. which methods can identify which institutions, actors, materials, and themes are endogenous or exogenous to a complex governance system? − we need methods that help in understanding complexity from the perspective of particular actors and that help in explaining their behaviour. how can we grasp the perspectives and roles of these actors without oversimplifying or disaggregating the system in which they are embedded? − we need methods to recognize and assess patterns of stability, iteration and change in global governance. instead of predicting the most probable outcomes, we may do better in using scenarios to outline different possible outcomes, along with tools to identify the underlying conditions for the various patterns and scenarios to materialize. − we need methods that are accessible to practitioners and policy makers. as scholars of global governance, our aim is not only to understand and explain a phenomenon in the most precise way possible, but also to communicate our findings to political and administrative decision makers. studying global governance as a complex system should not imply delivering overly http://dx.doi.org/10.20377/cgn-110 complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: 6 university of bamberg press complex results, but rather provide new insights and perspectives that can contribute to political change. to address such challenges in a manageable and accessible way, research on governance complexity needs to rely on a mix of methods – one that brings together tools from complexity sciences, e.g. machine learning or network analysis, with more qualitative approaches used in conventional governance research, such as institutional and discourse analysis. such a sensible combination is a tall order and requires creative combinations of skills and expertise. respective combinations are challenging but feasible, as innovative methods like qualitative and discursive network analysis (bellotti, 2014; fuhse & mützel, 2011; leifeld, 2017) or the integration of qualitative scenarios and quantitative simulations (symstad et al., 2017) show. theoretical challenges scholars have already developed theories about how individual actors react to institutional complexity. this includes the use of specific strategies such as the prioritizing of one institution over others, so-called forum-shopping or -shifting (cf. alter & meunier, 2009; orsini, 2013); the creation of competitive organizations or regimes (cf. morse & keohane, 2014); and non-hierarchical and indirect forms of governance of one institution over others such as orchestration (cf. abbott et al., 2015). global governance scholars have also addressed questions at the institutional systems level, expecting, for example, new equilibria and institutional divisions of labour to evolve in the face of complexity (e.g. abbott and faude, 2020, abbott et al., 2016; gehring & faude, 2013; oberthür & pożarowska, 2013; tosun et al., 2016; zürn & faude, 2013). others rather anticipate iterations of governance processes across scales or “fractals” (chettiparamb, 2014; bernstein & hoffmann, 2018), while highlighting a steady increase in governance experimentation and innovation (hoffmann, 2011). notwithstanding these crucial achievements, we hold that these and other theoretical efforts need to be expanded in three ways to address major research gaps about the complex world order: − we should not only focus on the complexity of a single institutional system, but also seek to explain its interaction with networks of actors across scales and levels (national, international, and transnational) as well as constellations of materials, norms, knowledges or discourses. − we should go beyond identifying the specific strategies of actors and engage more with the wider implications of complexity when researching political agency (room, 2015; haynes, 2015; möller, 2020). concretely, a theory-driven analysis should take into account that complexity may influence spaces for political agency. as such, it may alter key aspects of legitimacy, accountability, transparency, technocracy and power, – which in turn may impede or facilitate the possibility to employ certain strategies, or to even perceive the need for such responses (cf. bäckstrand et al., 2018; faude & große-keul, 2020). we need to know more about the conditions that impact these (de)politicising effects in complexity, e.g. how certain sources of power matter more than others and how certain forms of governance draw attention while others remain largely unperceived (gupta & möller, 2019). − we need to move beyond a simple disaggregation of complex systems if we want to appropriately grasp the dynamics between agency and structure. this poses the challenge of not just studying particular actors and institutions, or even the structure and behaviour of a certain governance system as a whole. in addition, we need to scrutinize how the relationship between agents and structures causes conditions of stability and change in global governance, and how these conditions in turn affect the emergence and reproduction of certain patterns across governance systems and scales (cf. young, 2017). focusing on the relations between components of a governance system, and bridging governance scholarship on the behavior of both actors and systems can already provide crucial insights to these three research gaps. at the same time such a research focus would acknowledge the non-linearity and interdependence inherent to a complex governance system. http://dx.doi.org/10.20377/cgn-110 complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: 7 university of bamberg press in addition, there are several possible routes to integrate complexity perspectives from other disciplines into theories of global governance. on the one hand, governance scholars can learn from different literatures, e.g. the one on ecological management which offers important examples about how to account for uncertainties when theorizing about dynamics, transformations and responses in complex governance systems (levin et al., 2012). these ways of theory-building are compatible with a series of subjectivist and interpretive approaches in political science and other social sciences that seek to understand, rather than to explain, important contingencies in political systems. on the other hand, learning from formal approaches like computational modeling can help us to identify which aspects of complex governance systems are still predictable and hence can be theorized by explanatory or objectivist approaches, e.g. certain types of behaviour of actors or conditions for systemic stability (symons & boschetti, 2013). new concepts, methods, and theories on institutional complexity: contributions from the special issue in response to the conceptual, methodological, and theoretical questions raised above, some of the workshop participants have offered their own ideas, from their perspective as global governance scholars, in the form of short think-pieces or fully-fledged research articles. in this spirit, some contributions represent conceptual endeavours that evaluate the applicability of ideas from the complexity sciences to a global governance setting. others engage with methodological questions, relating to fundamental issues such as the determination of analytical boundaries or the contextualization of interdependent cases. still others engage with theoretical aspects of the research agenda, exploring how complexity-based theories can be useful for answering global governance questions, or putting forward own theories of complex dynamics within a given system of global governance. the result is a smörgåsbord of ideas on researching complexity in global governance. it represents a cross-section of the creativity and engagement we experienced during the three-day workshop, where participants felt free to experiment with ideas that were new and exciting to them. it is also meant as a way to reach out to scholars who engage with complexity in other disciplines and fields, laying out questions that governance researchers engage with and inviting response, collaboration, and involvement from interested readers. concepts fundamental to the workshop discussions was an endeavour to clarify the difference between a governance system that is complex versus a governance system that is complicated. as we argued above, much of the literature on institutional complexity has focused on the proliferation of institutions and fields, often disaggregating institutional complexes and their relations into dyadic institutional overlap or conflict (cf. oberthür & gehring, 2006; stokke, 2012). by contrast, the notion that the complexity of a system is more than the number of its components and bilateral connections, but rather depends on the manner and results of their overarching set of relations, caused extensive debate during the workshop and the ensuing work on this special issue. is it at all possible to study complexity in a governance system if one cannot disaggregate its components? against this backdrop, several contributors to this special issue focus explicitly on the fundamental question of what complexity in global governance means, and what kinds of properties or characteristics one can study without falling into the disaggregation trap. to begin with, james hollway asks how and in what ways a given regime complex is “complex”. he argues that (regime) complexes consist of both spatial and temporal dependence, and that conceptual tools of network, time-series and sequence analysis can help us diagnose how complex a system is on each of these dimensions. benjamin faude addresses the question of how to conceptually determine system boundaries and points at the trigger function of events or shocks that are external to these boundaries. moreover, david alemna with colleagues, benjamin faude, peter haas and jon western, james hollway, marielle papin, and andrea schapper all engage with what we call ‘input descriptors’, http://dx.doi.org/10.20377/cgn-110 complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: 8 university of bamberg press meaning characteristics of a governance system that make this system complex. these complexity markers differ from author to author in terms of both terminology and substance, but all point at important relational qualities that represent more than the sum of a system’s components. they include, among others, the notions of diversity and differentiation (faude), dependency (hollway), openness (papin), interventions and patterns (alemna et al.), and a new quality of complexity as an overarching systemic property with severe implications for political decision-making (haas & western). moreover, schapper elaborates on “super-networks” as new forms of transnational advocacy that link institutions from different policy fields. marielle papin and benjamin faude also address what we call ‘output descriptors’, meaning the resulting properties of a complex system once this system is running. papin stresses in particular the three aspects of self-organization, emergence, and adaptability and how these may impact functional distributions, authority and cooperation patterns and the balance of power in a complex governance system. faude equally stresses the importance of adaptability that comes with the further institutional differentiation of a governance system in reaction to external events. methods some of the methodological discussions in the workshop revolved around the fit of conventional methods with complexity. much scholarship on institutional complexity is qualitative and has traditionally relied on literature analysis, interviews, and observations of single or a small amount of case studies, with few pioneering studies addressing complexity in larger systems (cf. zelli, 2015). quantitative scholarship on the complexity of global governance originated in the statistical approach used by many political scientists, but has meanwhile ventured somewhat into the realms of network analysis and agent-based modelling. despite some advances into new terrain, many global governance scholars face the challenge of learning complexity-compatible methods while remaining true to the research questions that motivate their engagement. going beyond these conventional approaches, the methodological contributions that this special issue makes reflect on issues such as what new approaches one could take, what kind of additional data one would need and what sorts of alternative assumptions one might need to develop. the first question is, again, a question of analytical boundary. where should complexity research start, and where should it end? james hollway argues that decomposing a system into individual institutions, or pairs thereof, or studying them cross-sectionally, is only warranted for those (increasingly rare) settings that are marked very low dependence. he takes aim at common research design considerations—sampling, prediction, and replication—and points out what non-linearity and interdependence imply for the use of statistical methods. taking a more empirical angle, david alemna and colleagues, andrea schapper, as well as peter haas and jon western, provide us with examples of how a complexity perspective can be used to study specific developments of concern to global governance scholars. alemna et al. analyse the structural effects of interventions by the international monetary fund in latin america using longitudinal data and a quantitative method called dynamic pattern synthesis, based on qualitative comparative analysis and configurational modelling. with this methodology they seek to defy reductionist explanations. they manage to identify evolving policy patterns that are far from the original intentions of the international monetary fund whose interventions they analyse in their case study. while alemna and colleagues thereby take a birds-eye perspective on the complexity patterns in a governance system, schapper seeks to understand complexity from the perspective of actors that are navigating therein. she analyses the collective strategy that a network of actors adopted when bridging the interface between human rights and climate change in international negotiations, using qualitative interviews and observations to capture their perspectives. she can show how creative agency could make use of governance complexity, with transnational advocacy networks building a supernetwork that managed to incorporate rights principles into the 2015 paris agreement. on the other http://dx.doi.org/10.20377/cgn-110 complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: 9 university of bamberg press hand, the complexity contributed to the eventual loss of momentum of that super-network since it put continuous demands on each actor for keeping up expertise and relationships. haas and western, finally, contribute an historical analysis of the system of international institutions, contrasting the relatively “simple” mandates on which these were built with the change in perceived complexity amongst decision makers and the recent backlash against norms of international diplomacy. with this angle they can identify potential institutional reforms that can help a key actor like the us government to do better justice to the new quality that marks the complex world order of today. with these either more structural or agent-based angles and methods, each contribution found its unique positioning within the agency-structure dynamics of complex governance research. while being mindful of not repeating a disaggregation of the systems they put under scrutiny, they provide different methodological answers to the question of how much political agency and power actors may hold and act upon in an ever more intricate system. theories strongly connected to this last point, one key theoretical question at the core of governance scholarship is the form and effect of power, and so much of the discussion around complexity in global governance linked back to this concept. connecting power to the theoretical challenges we discussedabove has led us and the contributors to this special issue to questions like: is power somehow different in a complex system? do actors in a complex system wield other kinds of power? does the system itself embody some kind of power and can thereby inhibit or enable the actors it contains to realize their goals? the contributions to this special issue address power in complex systems both directly and indirectly. in line with their methodical approaches, alemna et al. take a structuralist theoretical approach, showing how the complex institutional environment qualified the agency of the imf and facilitated a series of non-intended outcomes. andrea schapper, on the other hand, takes a more agentoriented theoretical perspective, showing how non-governmental organizations may navigate and adjust regime complexes to achieve institutional linkage and thereby work towards their political goals. haas and western found some middle ground between these two perspectives by identifying certain spaces where domestic governmental agency is still possible within wider complex governance systems. another set of core questions particular to the literature on institutional complexity revolves around stability and change. why do institutional landscapes evolve in the way they do? what causes differences in their structure and manifestation? how can their trajectories of development be explained? and what effects do such changes have on political and social behaviour? some of the contributors to this special issue offer their own answers to such questions. benjamin faude identifies exogenous shocks as triggers for structural change and further differentiation in the institutional system. building on sociological differentiation theory, he argues that such shocks may lead to adaptation through increased complexity, meaning that institutional proliferation may be a strategy to improve the overarching resilience of the affected governance system. dennis schmidt looks at the further consequences of such structural changes and differentiation in the institutional system. he establishes the notion of complexity as law-governed emergence, based on his adaptation of english school tenets to the study of complex global governance. schmidt holds that this emergence has two major consequences for institutional settings in global governance, fragmentation, and clustering. the resulting dynamics may, in turn, reinforce changes in overarching normative structures in which this institutional complexity is embedded. conclusions and outlook our reflections about the need for a complexity science perspective in global governance research, as well as the contributions by the authors in this special issue, have largely focused on different challenges of studying institutional complexity as a complex system. on the one hand, we http://dx.doi.org/10.20377/cgn-110 complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: 10 university of bamberg press embrace the challenge of acknowledging the characteristics inherent to complex systems and avoiding the disaggregation trap. on the other hand, we also heed the challenge of staying true to core questions of political and social science (power, legitimacy, justice) while learning to employ tools and perspectives of the complexity sciences. a subject that has not been as prominent in these reflections is the role of policy responses that could be stimulated with such an interdisciplinary perspective on institutional complexity. what could research with a complexity science lens contribute in the “real” world? what is its societal impact? some important ideas around such questions were also gathered at our workshop in duisburg, and we briefly reflect upon them in this outlook on the next research frontiers. in general, we feel that adopting the meta-level approach that a complexity science lens offers can provide important insights for decision makers and practitioners about why certain things happen in the way they do, including a diversity of non-intended results. researching global governance dynamics from a complexity angle can raise awareness about the drivers behind the actions and responses of different actors in a complex system as well as about the dynamics that lead to intentional and non-intentional consequences of these actions. it can uncover underlying goals and motivations of responses, including the navigation of complex systems through forum-shaping or shifting, the changing of a complex institutional system through orchestration and management, or the harnessing of a complex governance system through simplification and the use of certain heuristics (zelli et al., forthcoming). in uncovering these dynamics, such a lens can also highlight the kinds of capacities needed to attain agency in a global governance setting that is increasingly defined by complexity. understanding the dynamics at work will help different actors to both tailor and time their resources and efforts in more efficient ways, e.g. by identifying important brokers in a network that they can target for their purposes. it can also help these actors see wider connections between different governance systems, as well as the emergent effects and consequences that certain types of actions and behaviours can have. gaining insight into these dynamics can help decision makers and practitioners identify important levers of change that may otherwise stay hidden or unrecognized. in the face of proliferating global crises, we think that this kind of knowledge is urgently needed. the biggest challenge regarding responses and societal impacts remains, then, to communicate the results of research on institutional complexity in a way that is accessible and useful to a wider audience. taken together with the aforementioned challenges, we conclude that scholars of institutional complexity today need to navigate a three-dimensional plane of challenges: first of all, the challenge of truly acknowledging and integrating the particular characteristics of complex systems into their research; secondly, doing this without compromising on key research foci of political and social sciences, while being open for new perspectives and questions on how to approach these foci; and thirdly, making sure the results are available and understandable for a wider audience. to meet these challenges, and to move the research agenda on institutional complexity forward calls for a kind of creativity and collaboration that has so far been lacking in this research field. creativity is called for in terms of asking new kinds of questions, exploring unfamiliar theories, and combining as yet un-combined methods. collaboration is called for in terms of multi-, inter-, and transdisciplinarity, combining the methodological expertise of different disciplines, attaining a holistic theoretical perspective on the subject matter, and integrating the concerns and struggles of decision makers and practitioners into the heart of the institutional complexity research agenda. it also requires carrying the ensuing knowledge out into the world, beyond the paywalls and jargon of academic journals and publications. we believe that creativity and collaboration, as experienced during our duisburg workshop, provide the key to achieving this tall, but necessary, order. likewise, we hope that this special issue can make a contribution to stimulate further creativity and collaboration across our different disciplines on the complexity of global governance. http://dx.doi.org/10.20377/cgn-110 11 university of bamberg press complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: acknowledgments we are highly indebted to the swedish research council formas and the centre for global cooperation research / käte hamburger kolleg in duisburg, germany. formas financed fariborz zelli’s and ina möller’s research project on ‘navigating institutional complexity in global climate governance’ at lund university and co-financed the project’s final workshop in duisburg, germany in late 2017. the käte hamburger kolleg provided generous logistical and intellectual support and the core funding for that workshop. the contributions and discussions at the workshop marked the beginning of a fruitful collaboration that led to this special issue. moreover, fariborz zelli is particularly grateful to the centre for generously hosting him as a senior fellow. references abbott, k. w., & faude, b. (2020). choosing low-cost institutions in global governance. international theory, first view, 1-30, doi 10.1017/s17529-719-2000020-2. abbott, k. w., genschel, p., snidal, d., & zangl b. (2015). two logics of indirect governance: delegation and orchestration. british journal of political science, 46(4), 719-729. abbott, k. w., green, j. f., & keohane, r. o. (2016). organizational ecology and institutional change in global governance. international organization, 70(2), 247-277. aldrich, h. e. (1979). organizations and environments. englewood cliffs, nj: prentice-hall. aligica, p. d., & tarko, v. (2011). polycentricity: from polanyi to ostrom, and beyond. governance, 25(2), 237-262. alter, k. j., & meunier, s. (2009). the politics of international regime complexity. perspectives on politics, 7, 13–24. axelrod, r. (1986). an evolutionary approach to norms. american political science review, 80, 1095-1111. bäckstrand, k., zelli, f., & schleifer, p. (2018). legitimacy and accountability in polycentric climate governance. in a. jordan, d. huitema, h. van asselt, & j. forster (eds.), governing climate change: polycentricity in action (pp. 338 356). cambridge, uk: cambridge university press. bellotti, e. (2014). qualitative networks: mixed methods in sociological research. london: routledge. bernstein, s., & hoffmann, m. (2018). the politics of decarbonization and the catalytic impact of subnational climate experiments. policy sciences, 51, 189-211. biermann, f., pattberg, p., van asselt, h., & zelli, f. (2009). the fragmentation of global governance architectures: a framework for analysis. global environmental politics, 9(4), 14-40. biermann, r., & koops, j. a. (2017). palgrave handbook of inter-organizational relations in world politics. london: palgrave macmillan. chambers, w. b. (ed.). (1998). global climate governance: inter-linkages between the kyoto protocol and other multilateral regimes. tokyo, japan: united nations university press. chambers, w. b. (ed.). (2001). inter-linkages. the kyoto protocol and the international trade and investment regimes. tokyo, japan: united nations university press. chambers, w. b. (2008). interlinkages and the effectiveness of multilateral environmental agreements. tokyo, japan: united nations university press. chettiparamb, a. (2014). complexity theory and planning: examining 'fractals' for organising policy domains in planning practice. planning theory, 13(1), 5-25. cudworth, e., & hobden s. (2011). posthuman international relations: complexity, ecologism and global politics. london: zed. faude, b., & fuss, j. (2020). coordination or conflict? the causes and consequences of institutional overlap in a disaggregated world order. global constitutionalism, 9(2), 268-289. faude, b., & große-keul, f. (2020). let’s justify! how regime complexes enhance the normative legitimacy of global governance. international studies quarterly, 64(2), 431-439. faude, b., & parizek, m. (2020). contested multilateralism as credible signaling: how strategic inconsistency can induce cooperation among states. review of international organizations, 15, doi 10.1007/s11558-020-09398-7. fuhse, j., & mützel, s. (2011). tackling connections, structure, and meaning in networks: quantitative and qualitative methods in sociological network research. quality and quantity, 45(5), 1067-1089. galaskiewicz, j. (1979). interorganizational relations. annual review of sociology, 11, 281-304. gehring, t., & faude, b. (2013). the dynamics of regime complexes: microfoundations and systemic effects. global governance, 19(1), 119-130. gerrits, l. (2012). punching clouds: an introduction to the complexity of public decision-making. litchfield: emergent publications. gerrits, l., & marks, p. (2017). understanding collective decision making: a fitness landscape model approach. cheltenham: edward elgar. geyer, r. (2003). beyond the third way: the science of complexity and the politics of choice. the british journal of politics and international relations, 5(2), 237-257. guetzkow, h. (1966). relations among organizations. in raymond v. bowers (ed.), studies on behavior in organizations (pp. http://dx.doi.org/10.20377/cgn-110 https://doi.org/10.1017/s1752971920000202 complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: 12 university of bamberg press 13-44). athens, ga: university of georgia press. gunitsky, s. (2013). complexity and theories of change in international politics. international theory, 5(1), 35-63. gupta, a., & möller, i. (2019). de facto governance: how authoritative assessments construct climate engineering as an object of governance. environmental politics, 28(3), 480-501. haynes, p. (2015). the international financial crisis: the failure of a complex system. in r. geyer & p. cairney (eds.), handbook on complexity and public policy (p. 432-456). cheltenham: edward elgar. herr, r. a., & chia, e. (1995). the concept of regime overlap. toward identification and assessment. in b. davis (ed.), overlapping maritime regimes. an initial reconnaissance (pp. 11-26). hobart, australia: antarctic climate and ecosystems cooperative research centre and institute of antarctic and southern ocean studies. hoffmann, m. j. (2011). climate governance at the crossroads. experimenting with a global response after kyoto. oxford: oxford university press. hollway, j., morin, j.-f., & pauwelyn, j. (2020). structural conditions for novelty: the introduction of new environmental clauses to the trade regime complex. international environmental agreements, 20, s61-83. jervis, r. (1997). system effects: complexity in political and social life. princeton, nj: princeton university press. le prestre, p. (2017). global ecopolitics revisited: towards a complex governance of global environmental problems. london; new york: routledge. mcevoy, p., & richards, d. a. (2006). a critical realist rational for using a combination of quantitative and qualitative methods. journal of research in nursing, 11(6), 66-78. kavalski, e. (2007). the fifth debate and the emergence of complex international relations theory: notes on the application of complexity theory to the study of international life. cambridge review of international affairs, 20(3), 435-454. keohane, r. o., & nye, jr, j. s. (1977). power and interdependence: world politics in transition. boston, ma: little, brown & co. keohane, r. o., & victor, d. g. (2011). the regime complex for climate change, perspectives on politics, 9(1), 7-23. knoke, d., diani, m., hollway, j., & christopoulos, d. (2021). multimodal political networks. cambridge, uk: cambridge university press. leifeld, p. (2017). discourse network analysis: policy debates as dynamic networks. in j. n. victor, m. n. lubell, & a. h. montgomery (eds.), the oxford handbook of political networks (pp. 301-326). oxford, uk: oxford university press. levin, s., xepapadeas, t., crépin, a.-s., norberg, j., de zeeuw, a., folke, c., hughes, t., arrow, k., barrett, s., daily, g., ehrlich, p., kautsky, n., mäler, k.-g., polasky, s., troell, m., vincent, j. r., & walker, b. (2013). social-ecological systems as complex adaptive systems: modeling and policy implications. environment and development economics, 18(2), 111-32, doi 10.1017/s13557-70x-12000460. mcginnis, m. d. (ed.). (1999). polycentricity and local public economies: readings from the workshop in political theory and policy analysis. ann arbor, mi: the university of michigan press. milewicz, k., hollway, j., peacock, c., & snidal, d. (2018). beyond trade: the expanding scope of the nontrade agenda in trade agreements. journal of conflict resolution, 62(4), 743-773. mitchell, m. (2009). complexity: a guided tour. new york: oxford university press. möller, i. (2020). political perspectives on geoengineering: navigating problem definition and institutional fit. global environmental politics, 20(2), 57-82. morse, j. c., & keohane, r. o. (2014). contested multilateralism. the review of international organizations, 9(4), 385-412. oberthür, s., & gehring t. (eds.). (2006). institutional interaction in global environmental governance: synergy and conflict among international and eu policies. cambridge, ma: the mit press. oberthür, s., & stokke, o. s. (eds.). (2011). managing institutional complexity: regime interplay and global environmental change. cambridge, ma: the mit press. oberthür, s., & pożarowska, j. (2013). managing institutional complexity and fragmentation: the nagoya protocol and the global governance of genetic resources. global environmental politics, 13(3), 100–118. orsini, a. (2013). multi-forum non-state actors: navigating the regime complexes for forestry and genetic resources. global environmental politics, 13(3), 34-55. orsini, a, morin, j.-f., & young, o. r. (2013). regime complexes: a buzz, a boom, or a boost for global governance? global governance, 19(1), 27-39. orsini, a., le prestre, p., haas, p. m., brosig, m., pattberg, p., widerberg, o., gomez-mera, l., morin, j.-f., harrison, n. e., geyer, r., & chandler, d. (2019). complex systems and international governance. forum introduction. international studies review, viz005, doi 10-1093/isr/viz005. ostrom, v., & ostrom, e. (1965). a behavioral approach to the study of intergovernmental relations. the annals of the american academy of political and social science, 359(1), 137-146. pfeffer, j. s. g., & salancik, g. r. (1978). the external control of organizations: a resource dependence perspective. new york, ny: harper & row. polanyi, m., & prosch, h. (1975). meaning. chicago, il: university of chicago press. pollitt, c. (2009). complexity theory and evolutionary public administration: a sceptical afterword. in g. r. teisman, a. van buuren, & l. gerrits (eds.), managing complex governance systems (pp. 37-55). new york: routledge. raustiala, k., & victor, d. g. (2004). the regime complex for plant genetic resources. international organization, 58(2), 277-309. reed, m., & harvey, d. l. (1992). the new science and the old: complexity and realism in the social sciences. journal for the theory of social behaviour, 22(4), 353-380. http://dx.doi.org/10.20377/cgn-110 complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 1-13 doi: 13 university of bamberg press room, g. (2015). complexity, power and policy. in r. geyer, & p. cairney (eds.), handbook on complexity and public policy (pp. 19-31). cheltenham: edward elgar. rosenau, j. n., & czempiel, e.-o. (1992). governance without government: order and change in world politics. cambridge, uk: cambridge university press. schelling, t. c. (1971). dynamic models of segregation. journal of mathematical society, 1, 143-186. schmidt, s. m., & kochan, t. a. (1977). interorganizational relationships: patterns and motivations. administrative science quarterly 22(2), 220-234. selin, h., & vandeveer, s. d. (2003). mapping institutional linkages in european air pollution politics. global environmental politics 3(3), 14-46. spaargaren, g. (2011). theories of practices: agency, technology, and culture: exploring the relevance of practice theories for the governance of sustainable consumption practices in the new world-order. global environmental change, 21(3), 813-822. stokke, o. s. (2012). disaggregating international regimes: a new approach to evaluation and comparison. cambridge, ma: the mit press.
 symons, j., & boschetti, f. (2013). how computational models predict the behavior of complex systems. foundations of science, 18, 809-21. doi 10.1007/s10699-012-9307-6. symstad, a. j., fisichelli, n. a., miller, b. w., rowland, e., schuurman, g. w. (2017). multiple methods for multiple futures: integrating qualitative scenario planning and quantitative simulation modeling for natural resource decision making. climate risk management, 17, 78-91, doi 10.1016/j.crm.2017.07.002. tosun, j., koos, s., & shore, j. (2016). co-governing common goods: interaction patterns of private and public actors. policy and society, 35, 1-12. van de graaf, t. (2013). fragmentation in global energy governance: explaining the creation of irena. global environmental politics, 13(3), 14-33. wagenaar, h. (2007). governance, complexity and democratic participation. the american review of public administration, 37(1), 17-50. young, o. r. (1996). institutional linkages in international society. polar perspectives. global governance, 2(1), 1-24. young, o. r. (2002). the institutional dimensions of environmental change. fit, interplay, and scale. cambridge, ma: the mit press. young, o. r. (2017). governing complex systems: social capital for the anthropocene. cambridge: the mit press. zelli, f. (2010). conflicts among international regimes on environmental issues. a theory-driven analysis. tübingen, germany: eberhard-karls university. zelli, f. (2011a). the fragmentation of the climate governance architecture. wiley interdisciplinary reviews: climate change, 2(2), 255-270. zelli, f. (2011b). regime conflict and interplay management in global environmental governance. in s. oberthür, & o. s. stokke (eds.), managing institutional complexity: regime interplay and global environmental change (pp. 199-226). cambridge, ma: the mit press. zelli, f. (2015). institutional fragmentation. in p. pattberg, & f. zelli (eds.), encyclopedia of global environmental governance and politics (pp. 469-477). cheltenham, uk: edward elgar. zelli, f., biermann, f., pattberg, p., & van asselt, h. (2010). the consequences of a fragmented climate governance architecture: a policy appraisal. in f. biermann, p. pattberg, & f. zelli (eds.), global climate governance beyond 2012. architecture, agency and adaptation (pp. 25-34). cambridge, uk: cambridge university press. zelli, f., gerrits, l., möller, i., & widerberg, o. (forthcoming). institutional complexity and political agency in polycentric governance. in f. gadinger, & j. a. scholte (eds.), polycentrism: how governing works today. zelli, f., gupta, a., & van asselt, h. (2013). institutional interactions at the crossroads of trade and environment: the dominance of liberal environmentalism? global governance 19(1), 105-118. zelli, f., möller, i., & van asselt, h. (2017). institutional complexity and private authority in global climate governance. environmental politics, 26(4), 669-693. zelli, f., bäckstrand, k., skovgaard, j., nasiritousi, n., & sanderink, l. (eds.). (2020). governing the climate-energy nexus: challenges to coherence, effectiveness, and legitimacy. cambridge, uk: cambridge university press. zürn, m., & faude, b. (2013). commentary: on fragmentation, differentiation, and coordination. global environmental politics,13(3), 119-130. http://dx.doi.org/10.20377/cgn-110 https://doi.org/10.1016/j.crm.2017.07.002 intersections in delta development; analyzing actors for complexity-sensitive spatial concepts complexity, governance & networks (2014) 79–98 79 doi: 10.7564/14-cgn18 intersections in delta development; analyzing actors for complexity-sensitive spatial concepts bonno pela,*, jitske van popering-verkerkb, arwin van buurenb, and jurian edelenbosb auniversité libre de bruxelles e-mail: bonno.pel@ulb.ac.be berasmus university rotterdam e-mail: verkerk@fsw.eur.nl e-mail: vanbuuren@fsw.eur.nl e-mail: edelenbos@fsw.eur.nl delta areas can be considered complex adaptive socio-ecological systems. the dutch southwest delta, facing serious flood risks, vulnerability to ecological decline, and various challenging issues of agriculture, industry, harbor development and energy provision, is a case in point. still, many institutional barriers exist towards governing and planning this complex whole as such. in this article we therefore develop and test a method for the development of integrative, complexity-sensitive spatial concepts: first, stakeholder analysis techniques are used to disclose the diversity of system understandings amongst the actors involved. moreover, the method mobilizes these constructivist techniques to gain insight into the cas property of co-evolving subsystems. through the subsequent inventory, classification and synthesis of such ‘intersections’ between subsystems, the method helps identify the delta’s crucial clusters of interdependent subsystems, or ‘configurations’. we present three of such configurations, to illustrate how this method informs the step from systems analysis to spatial design. keywords: deltas, co-evolution, synchronization, actor analysis, spatial design, intersections 1. introduction: delta complexity and the need for integrative spatial concepts delta areas can be considered complex adaptive socio-ecological systems. these areas are characterized by both morphologic and hydrologic dynamics, and by social complexity: many actors (governments, non-governmental organizations, private organizations, etc.) at different scales (local, regional, national) have interdependent relationships (van leeuwen & van buuren, 2013). the dutch southwest delta, where the rhine, * corresponding author. article_14-18.indd 79 03/03/15 3:06 pm 80 b. pel et al. / intersections in delta development meuse and scheldt rivers meet with the north sea (see figure 1 below), is a case in point: it is faced by serious flood risks due to climate change and soil subsidence, is vulnerable to ecological decline, and also involves challenging issues of agriculture, industrial and harbor development, and restructuration of the energy sector (meyer, 2005). actors in the delta area tend to cope with these issues in different and often isolated ways, however, and coalitions tend to focus on sub-topics: at dutch national scale a delta program has been installed which is primarily concerned with anticipating flood risks (deltaprogram, 2011), the rotterdam harbor authorities’ developed a long-term strategy for port development (havenbedrijf rotterdam, 2011; see also vanelslander, kuipers, hintjens, & van der horst, 2011), and the environmental organization wwf develops a future vision of an open and ecologically vitalized delta (wereld natuurfonds, 2010). besides these largescale visioning and planning programs, a host of local, regional and interregional initiatives can be witnessed with regard to tourism, housing and small water-related companies (e.g. dienst landelijk gebied, 2011; intergemeentelijke samenwerking voorne putten, 2011; samenwerkingsorgaan hoeksche waard, 2009). this collection of planning initiatives constitutes a considerable scope for meeting the delta’s challenges. however the main problem is that (groups of ) actors uphold their particular focus and scope, and accordingly develop their own partial solutions. complex governance systems are often characterized by fragmentation, meaning that every actor (government, citizen, ngo, or private organization) has its own main perceptions and ambitions and approaches in analyzing issues and finding solutions (folke et al., 2007; edelenbos & teisman, 2010; termeer, dewulf, & van lieshout, 2010). figure 1. southwest delta article_14-18.indd 80 03/03/15 3:06 pm b. pel et al. / intersections in delta development 81 the visions of involved actors and their underlying system understandings prove insufficiently compatible in terms of geographical scale, temporal orientation, and sector ambitions. involved actors are likely to diverge in their system understandings and subsequent problem definitions (luhmann, 1995; van meerkerk, van buuren, & edelenbos, 2013; ulrich, 1983). this poses challenges to finding integrative concepts that ‘synchronize’ multiple issues in the area in such a way that these strengthen each other, rather than cancel out (rammert, 2000; edelenbos, bressers, & scholten, 2013; pel, 2012; teisman & edelenbos, 2011). the main challenge for deltas can thus be defined as arriving at integrative spatial concepts through which perceptions, interests and ambitions can become aligned. this article is aimed to develop and test a method for analyzing delta areas as complex adaptive systems (cas), and for developing integrative spatial concepts (zonneveld & verwest, 2005) that are accordingly ‘complexity-sensitive’. this brings us to formulate our main question: how does a method for analyzing delta areas as a cas look like and how can this inform the development of integrative spatial concepts for these areas? we have answered this question by developing and conducting a particular kind of method of stakeholder analysis, which is aimed to disclose the diversity of perceived issues, problems and opportunities that arises at the intersections between the delta’s subsystems. based on interviews and document analysis, we have drawn conclusions on the frictions and convergences (interferential and symbiotic intersections) between different system understandings. synthesizing these typically multiple subsystem interactions, our intersections method has allowed us to establish three systemic configurations to inform subsequent development of integrative spatial concepts for the southwest delta. the article proceeds as follows: first, we outline our understanding of the dutch southwest delta area as a complex adaptive system (section 2). subsequently we develop our methodology of intersections analysis. combining cas thinking with actor/ stakeholder analysis, it helps identify the salient systemic configurations, as footholds for the development of integrative spatial concepts (section 3). next, as a proof-of-principle, we describe three of those configurations (section 4). based on a comparison of these configurations, we discuss the scope and limitations of intersections analysis as inputs to complexity-sensitive spatial design (section 5). 2. delta areas as complex adaptive systems (cas) in the project ‘integral planning and design in the southwest delta’ (integral planning and design in the southwest delta, 2013), the delta is conceptualized as a complex adaptive system (cas). it is treated as a dynamic and interconnected whole, with several physical and social subsystems as its co-evolving components. crucially, these subsystems develop according to different development speeds and subject to distinct logics, which gives rise to the emergence of complex patterns and dynamic interactions among article_14-18.indd 81 03/03/15 3:06 pm 82 b. pel et al. / intersections in delta development them (uhl-bien et al., 2007; dammers et al., 2014). this cas-conceptualization, and the associated decomposition into spatial layers and functional subsystems, is depicted below: this systems model has been developed to foster interdisciplinary collaboration between researchers and practitioners in public administration, spatial design & planning, and geographical information systems. it has therefore been crafted to accommodate different professional requirements and applications, and to articulate different aspects of delta complexity. in particular, the model allows to grasp delta complexity both as the interplay between spatial layers, and as the interplay between social/governmental subsystems. as regards the first, the model is based on the so-called ‘dutch layers’ approach’ (priemus, 2004; van schaick & klaasen, 2011; döpp, hooimeijer, & maas, 2011). the southwest delta is conceived of as a complex formation that is built up from upward conditioning spatial structures. the three layers are distinguished by their different transformation rates: the physical substrate develops slowly, and as a relatively solid structure it sets the conditions for infrastructure networks and occupations (settlement patterns). of the latter two, the layer of infrastructures typically conditions the development of various occupations and land-uses: without draining, dredging and dike construction, for example, the southwest delta would hardly be habitable at all (van de ven, 2004). other than the natural forces of the substrate layer however, this infrastructural layer is man-made. as also theorized in large technical systems literature, these sedimentations of human agency constitute path-dependencies but not deterministic forces (hughes, 1994; disco & van der vleuten, 2002). as a spatial model, the layers’ approach remains rather ill-equipped to articulate social and governance complexity, however. the model should differentiate further, at least reflecting that current governance is shaped by functional specialization, i.e. by administrative sectors. the layers approach has therefore been adapted to include a differentiation between societal subsystems. hydro-infra, energy and transport constitute not only physically separated infrastructure networks, for example, they are also governed by different constellations of actors. these subsystem divisions, reminiscent of the societal subsystems distinguished in studies of socio-technical transitions (grin, rotmans, & schot, 2010), bring the layers’ model in closer correspondence with the complexity of networked governance and fragmented sector policies (teisman, van buuren, & gerrits, 2009). finally, apart from the layers and their subdivisions, the above systems conceptualization theorizes general trends of both human and physical nature, including technological, cultural and economic changes as well as physical processes such as climate change. these general trends indicate how the southwest delta evolves not through the interactions between its layers and subsystems only, but also through its adaptation to a dynamic system environment: for example, current base layer challenges cannot be understood without considering climate change, and the infrastructure layer was crucially influenced by the industrial revolution and global trade (cf. dammers et al., 2014). because of these article_14-18.indd 82 03/03/15 3:06 pm b. pel et al. / intersections in delta development 83 transformative trends in its system environment, the southwest delta can be considered a complex adaptive system. and as these adaptations are not necessarily desirable, delta actors devise plans and spatial concepts for purposive intervention. the above conceptualization thus articulates the apparent cas properties of the southwest delta in a way that expresses both physical-spatial and social-governance complexity. through this conceptualization we can also reformulate the above challenge of ‘integrative’ concepts: the integration of spatial functions and the synchronization between actors are to be understood as ways to respond to delta complexity. 3. actor analysis in a cas approach: building the method of intersections analysis the search for integrative concepts can be supported through an understanding that accounts for southwest delta complexity. arguably, the layered systems model (figure 2) can inform the search for spatial synergies, usefully distinguishing spatial processes and also articulating the associated governance complexity. in the following we outline our method of intersections: first, it is positioned as a method for disclosing system diversity figure 2. southwest delta as a complex adaptive system article_14-18.indd 83 03/03/15 3:06 pm 84 b. pel et al. / intersections in delta development (3.1). next, we describe how this involves a stepwise combination of stakeholder analysis techniques (3.2) and subsequent classification and synthesis of these data into systemic ‘configurations’ (3.3). 3.1. intersections analysis: disclosing delta system diversity considering the southwest delta as a complex adaptive system and seeking to develop accordingly complexity-sensitive spatial concepts, it is imperative to gain understanding of the relations between its subsystems (i.e. the arrows in figure 2). the latter ‘intersections’ indicate the subsystem interactions that are so crucial for cas development (holland, 1995). still, cas conceptualizations of the social world should account for the fact that this generally involves interactions between essentially diverse system ‘components’. due to this diversity, cas conceptualizations should not be deceiving us into believing that some ‘hidden order’ can be relied on for policy advice rather, it should sensitize us to the idea that interacting system components give rise to alternative trajectories (byrne, 2005). we follow byrne’s plea for a non-reductionist understanding of emergence, considering that the reduction or articulation of diversity has profound implications for policy advice, and for spatial design. our method of intersections analysis is likewise informed by the ‘transformative diversity’ stressed by stirling (2011). this notion captures his critique on the prevailing reductionism in transitions studies (grin et al., 2010), a cas-based approach to sustainability governance. transitions studies is focused on structural, long-term transformations in societal systems, seeking to uncover the system feedback patterns that can account for these system shifts (see also scheffer, 2009). its normative orientation towards sustainable development tends to invite a certain preoccupation with particular system transitions however, and with particular directions for change – which do not necessarily reflect the system understandings of the actors involved. instead, stirling (2011) argued convincingly, it is crucial to observe the multiple transformations that take place in a system, the different normative directions they take, and the different ambitions, perceptions and resources of the actors who shape these transformations. other transitions scholars have argued similarly for more fine-grained approaches that are sensitive to competing programs for change (hodson & marvin, 2010), intertwined decision-making arenas (jørgensen, 2012; späth & rohracher, 2010), and different scales and areas of transformation (raven, schot, & berkhout, 2012; coenen & truffer, 2012). in line with these calls to respect transformative diversity, we seek to develop a fine-grained understanding of the delta subsystems’ intersections. the delta system model should not become a reductionist straightjacket. it should rather serve as a heuristic through which to grasp the heterogeneity of the southwest delta system, and the diversity of its governance constellation in particular. this means concretely that the intersections between subsystems are not reduced into abstract ‘mechanisms’ of co-evolution, but rather as processes that are essentially constituted by the idiosyncrasies and interactions article_14-18.indd 84 03/03/15 3:06 pm b. pel et al. / intersections in delta development 85 of situated actors. likewise, in line with walby (2007) and jørgensen (2012), we treat our subsystems as strongly intertwined ‘arenas of development’, in which these situated and more and less well-positioned actors negotiate programs for action. the systems model, with its graphical suggestion of closed and monolithic subsystems (allen, 2011), is thus brought into accordance with the ontology of ill-bounded and overlapping governance networks (law, 1992; kingdon, 1995; kickert, klijn, & koppenjan, 1997; koppenjan & klijn, 2004). the earlier differentiation within layers is thus taken a step further, also differentiating within subsystems. finally, our fine-grained type of analysis should not completely forego the advantages of aggregation. however intricate and complex the processes at the delta system intersections may be, the 9 subsystems distinguished yield a multitude of those intersections. and as the analysis of intersections should ultimately inform spatial concepts that integrate, the disclosure of delta diversity should be followed by some sort of condensation. to guide such synthesis we can rely on existing classifications of interaction modes. a well-known distinction from game theory is the one between cooperation and defection, for example. this basic distinction resurfaces in classifications of co-evolution mechanisms: sandén and hillman (2011) (see also odum & barrett, 1985) distinguish between symbiosis, neutralism and competition, for example, thus adding neutral interactions. furthermore they also distinguish asymmetrical interactions (amensalism, commensalism and parasitism), which helps articulate issues of dominance. yet it remains important for our purposes to realize that interaction modes may combine, change over time or remain uncertain (axelrod, 2006; sandén & hillman, 2011; pel, 2014). moreover, the formal elegance of these classifications should not obscure the games real actors play (scharpf, 1997). we therefore stick to the basic differentiation between ‘interfering’ (competing), and ‘symbiotic’ (mutually reinforcing) subsystems, keeping the scope for refined applications in mind. having outlined how we seek to disclose system diversity, figure 3 displays how intersections analysis zooms in on our systems model: the delta’s subsystems are broken down into checkered, diverse networks of actors. as will be discussed in more detail in the next subsection, these actors uphold different system understandings. this internal diversity of subsystems implies that their intersections can be observed in similarly differentiated fashion: within the subsystem ‘agriculture’, for example, we find both established actors associated with current modes of production and upcoming innovators, and within the ‘ecosystems’ subsystem we find both advocates of nature conservation and champions of rather dynamic environmental management: mainstream actors of subsystem 1 may then uphold symbiotic relationships with innovative strands in subsystem 2, whilst experiencing interference with its mainstream actors. combined land-use between agriculture and ecological functions, to name a traditionally contested intersection, can thus be analyzed as an ambiguous cluster of interactions, shaped by the diversity of system understandings present in the subsystems involved. article_14-18.indd 85 03/03/15 3:06 pm 86 b. pel et al. / intersections in delta development 3.2. inventory: systems-oriented stakeholder analysis arguably, the concept of intersections brings cas-thinking into closer correspondence with the synchronization challenges signaled upfront: it grounds these subsystem interactions in the interactions between situated actors. these situated actors are likely to diverge in their system understandings, identifying different systems, problems and solutions through their particular ‘boundary judgments’ (ulrich, 1983; luhmann, 1995; van meerkerk et al., 2013). an important first step in the intersection analysis is therefore to gain insight in actors’ understandings of the delta system, and their position within it. their system understandings, as diverse angles on the delta system, are reconstructed on four key dimensions (see figure 3): their ambitions, their identification of relevant past and future developments, their perceived interdependencies with other actors (i.e. their shared aspirations, their recurring interferences with others) and their framings of time and scale. in this first step of stock-taking, intersections analysis thus amounts to a kind of stakeholder analysis. stakeholder analysis methods have been developed in conflict resolution (susskind & cruikshank, 1987), project management (edelenbos & klijn, 2009), business administration (ackoff, 1974; mitroff, 1983), management studies (savage, nix, whitehead, & blair, 1991; mitchell, agle, & wood, 1997) and public administration (de bruin, ten heuvelhof, & in ’t veld, 1998; edelenbos, 2000; koppenjan & klijn, 2004). conventional stakeholder analysis can be considered the process of identifying the individuals or groups figure 3. intersections article_14-18.indd 86 03/03/15 3:06 pm b. pel et al. / intersections in delta development 87 that are likely to be affected by a proposed action, and sorting them according to their impact on the action and the impact the action will have on them (ackoff, 1974). this information is used to assess how the interests of those stakeholders should be addressed. stakeholder analysis is nowadays treated as key part of stakeholder management (edelenbos, 2000) and network management (koppenjan & klijn, 2004). in this development, stakeholder analysis has evolved beyond the identification of key interests and values, but also includes analysis of network interdependencies and interactions. similar to these actor analyses, intersection analysis is aimed to reconstruct the main actors’ perceptions of problems and solutions, as well as their network interactions and interdependencies. yet instead of charting stakeholders in regard to a particular project or issue (reed et al., 2009), intersections analysis uses the associated techniques of stakeholder interviews and document analysis to disclose the delta’s diversity of system understandings. this systems-oriented stakeholder analysis calls for some specific methodological measures, however. first of all, the time-consuming stakeholder interviews crucially require a selection of actors, representing the vast number of actors that tend to be present in delta systems. we have done so in line with our general orientation towards systemic diversity: we have selected actors who 1. are evenly distributed over the 9 subsystems distinguished; 2. occupy both dominant and subaltern positions within those, and 3. operate at different scales (local/regional/national/international). we have also distinguished between actors holding actual stakes, and experts. arguably, these dimensions of diversity help to bring out the synchronization challenges that typically arise with regard to time, policy sector and scale (edelenbos et al., 2013). as has been described more extensively in van buuren, pel, verkerk, and edelenbos (2014), this selection procedure resulted in 18  interviews, held in may-june 2012. the selection of documents to be analyzed was guided likewise by this diversity orientation. next to this issue of actor selection and system representation, a second methodological measure was to secure data gathering consistent with our research purposes. this is non-trivial, as our type of stakeholder analysis does not pertain to a specific issue, but is meant to disclose diversity. data gathering was therefore structured by the four dimensions of system understandings distinguished earlier: ambitions, interpretations of past events, perceptions of interdependencies, and frames of temporal and spatial scales. these dimensions formed the basis for an interview topic list, and also served as guideline for document analysis (see van buuren et al., 2014). third, it was important to secure geographical concreteness, and stimulate respondents towards systems thinking. after all, our stakeholder analysis was not to serve policy advice on a particular governance issue, but should inform the development of integrative, complexity-sensitive spatial concepts. the reconstruction of interviewees’ system understandings has therefore been facilitated by confronting them with our system’s model (figure 2) and a map of the delta area (figure 1). this invited them to expose which subsystems they deemed relevant to their ambitions, and to mark present and future ‘hotspot’ article_14-18.indd 87 03/03/15 3:06 pm 88 b. pel et al. / intersections in delta development locations where system developments manifest themselves. moreover, the actors were asked to relate their main ambitions, projects, interferences and symbioses to the subsystems presented in the diagram. this helped us to chart the level and extent of divergence between system understandings, and to identify symbiotic and interferential subsystem interactions. 3.3. condensing variety: from intersections to configurations after the above inventory through stakeholder analysis, the second step is to condense the uncovered variety into appropriate building blocks for spatial design. after initial emphasis on variation, intersections analysis crucially involves selection upon the bulk of intersections, and the identification of systemic configurations to play into with spatial design. this is a matter of stepwise narrowing down: first, the results from interviews and document analyses have been merged into pre-structured ‘factsheets’ on separate subsystems (van buuren et al., 2014). next, these factsheets were screened for salient intersections. this allowed us to establish a long-list of intersections, covering all nine subsystems. this first condensation already yielded valuable insights: our systems model comprises 9 × 8 = 72 intersections, discounting the interactions a subsystem may have with itself. our empirical data brought out a clear difference between theory and practice: some intersections recurred throughout the various accounts, whilst other subsystem interactions remained conspicuously absent. in other words, actors proved selective in their observation, focusing on some interdependencies and considering others as hardly relevant. having mapped the intersections long-list onto our system diagram, the subsystem relations most salient to the actors involved came out. as this subset still comprised a multitude of interactions however, further condensations was needed for subsequent identification of configurations. such configuration, i.e. clusters of more than two subsystems that display interactions salient to actors involved, have been developed as follows: first, some of the intersections proved more prominent than others: the ‘hydro-infrastructure’ and ‘agriculture’ subsystems are notorious for their mutual interferences, for example. gradually eliminating the relatively weaker intersections and lifting out the prominent ones, the crucial linkages in the southwest delta system became apparent. second, the empirical data already provided useful indications of mutually connected intersections: in their accounts of the aforementioned ‘hydro-infrastructure’/‘agriculture’ intersection, for example, actors often mentioned the ‘ecosystems’ subsystem to be an integral part of the issue. similarly, actors brought forward such clusters of intersections under general themes like urban development, economic growth, or sustainable development. third, the condensation process was conducted in iterative fashion: the configurations were to be limited in number (3 to 5), yet together they should cover the 9 subsystems. going backand-forth between the early outlines of configurations and the intersections listed, our clustering into configurations could be refined. article_14-18.indd 88 03/03/15 3:06 pm b. pel et al. / intersections in delta development 89 finally, it needs to be acknowledged that the above condensation process cannot be formalized exhaustively. the selection of salient intersections can be methodologically channeled, as described, but unavoidably contains an element of intuitive, creative synthesis. this may detract from the explanatory power of intersections analysis, and indeed there may be public administration frameworks that do better in this respect, but it also reminds of the particular purpose of our method: intersections analysis is meant to inform the development of integrative spatial concepts, and thus gradually moves from analysis to design, and from science to ‘craft’ (bertolini et al., 2012). a practical advantage is then that this method, or elements from it, can be easily deployed in participative fashion. this has been described elsewhere, in pel, duijn, janssen, and edelenbos (2013). in the next section we briefly describe three configurations, as outcomes of intersections analysis. the nature and characteristics of these configurations differs; some configurations mainly indicate clusters of ‘interference’ to be overcome, others mainly indicate ‘symbioses’ to seize and to cultivate further. these configurations are starting points for design processes that take delta diversity into account. 4. disclosing delta diversity: 3 configurations intersections analysis charts the space for spatial concepts in tune with ever-dynamic governance networks. our intersections analysis brought forward a couple of configurations that arguably capture the main bones of contestation in the southwest delta: these sets of intersections are described recurrently by respondents and also feature prominently in various policy documents and expert studies. we have labeled these configurations as follows: ‘settlement & employment’ (4.1), ‘living with nature’ (4.2), and ‘contested waters’ (4.3). 4.1. disclosing delta diversity (i): ‘settlement and employment’ the southwest delta is enclosed by the two major harbor-industrial complexes of rotterdam (netherlands) and antwerp (belgium) (see figure 1). more generally, it is a relatively sparsely populated area amidst the otherwise strongly urbanized areas of the dutch ‘rim city’ conurbation and flanders. not surprisingly therefore, an important cluster of intersections could be identified and elaborated into a configuration on ‘settlement and employment’. this cluster involves largely symbiotic intersections between the subsystems harbors & industries, transport, energy and urban structure. the key ambitions involved are the enhancement of economic development on the one hand, and the safeguarding of attractive living conditions on the other. interferences come up once economic growth is perceived to go at the expense of the environment, but several actors overcome these by arriving at common, symbiotic ambitions. the harbor-related industries around vlissingen, for example, see their envisioned expansion article_14-18.indd 89 03/03/15 3:06 pm 90 b. pel et al. / intersections in delta development of activities curbed by citizens’ attempts to prevent the concomitant disturbance to living conditions. the industries and the housing corporation have therefore jointly initiated a project to improve the isolation of these houses. as regards the identification of relevant past and future developments, most actors acknowledge this configuration to have resulted from a historic development path towards a ‘red circle’ around a ‘blue-green heart’. the first is densely populated, with all kind of large-scale economic activities and facilities. the second exists of predominantly rural islands and regional water-related economic activities. especially the actors from the urbanized rim around the delta seek to develop this blue-green heart primarily as an attractive residential area, thus serving the economic development of the delta area at large. this perspective interferes with local ambitions towards economic development, however. especially harbor-industrial development and the associated transport investments should not be dedicated to the existing ‘mainports’ only. local actors are suspicious about plans that mainly ‘exploit’ their area, for the resulting spatial-economic unevenness. whether endorsed or not however, actors agree that this structure will continue to dominate delta evolution: in this regard harbor-industrial and transport subsystems are widely considered to be leading; actors from these subsystems also tend to uphold broad and influential strategic outlooks, considering the delta’s prospects in a long-term, international context. 4.2. disclosing delta diversity (ii): ‘living with nature’ the second configuration pertains directly to the delta’s characteristic natural processes. it involves the ever-tense intersections between subsystems that rely on natural processes for their production, and the substrate-related subsystems themselves: the configuration thus comprises agriculture, ecosystems, soil & water, and water construction works. as regards ambitions, the configuration can be considered ridden with interferences between economical ambitions, and concerns for the integrity of the substrate. to a considerable extent actors have found ways to overcome those: farmers, for instance, engage in explorations of alternative ground water arrangements, together with groundwater officials and construction firms. yet the robustness of the natural system remains an important ambition that often interferes with established rights and claims to eco-services. farmers may find symbiosis with environmental conservation by allowing nature development at the edges of their land, which actually renders their production systems less vulnerable to insects. the scope for such symbiosis only goes so far however. strong environmentalist pleas for a restoration of estuarine dynamics run up against an agricultural lobby that stresses the negative consequences for a freshwater supply that has been in place for several decades. more generally, this configuration is firmly rooted in the delta’s historical development: with its alluvial and reclaimed land, and the combination of fresh river water and saline sea water, the area has fairly unique geo-physical characteristics that are of importance for agriculture, ecosystems, water construction works and soil & water. actors in favor of ecological preservation consider the large-scale engineering interventions of article_14-18.indd 90 03/03/15 3:06 pm b. pel et al. / intersections in delta development 91 the last decades as threats to the robustness and quality of the geo-physical and ecological processes. their reference to historical qualities implying that these be restored, their assessment of past development has direct consequences for future developments. as this reading of history is not generally shared however, interferential relationships arise. with regard to future developments, two distinct strands can be noticed throughout these involved subsystems. against the actors favoring all kinds of engineering measures to control and exploit the natural processes, there are actors who primarily seek to maintain ecological integrity, only allowing for mildly interferential exploitation. the ‘living with nature’ configuration is intriguing for its complex actor configuration. in the first place, involved actors display mainly diverging frames of time and scale: in agriculture, actors tend to focus on the local level, and their planning horizon is strongly based on the 7-year harvest cycle. by contrast, actors in ‘hydro-infrastructures’ are focused on nationally organized flood protection, and their investments in construction works entail planning horizons up to fifty or hundred years. opportunities for symbioses arise when these different frames in time and scale synchronize nevertheless, for example when the use of agricultural land for dike reinforcement is combined with arrangements for land reallocation. second, there is the fundamental difference between conservational and developmental attitudes: actors in the ecosystems and soil & water subsystems are focused on protecting specific ecological qualities, and rely strongly on the associated regulatory frameworks. this leads to interferences, as actors from other subsystems are constrained. the very juridical approach tends to invite zero-sum framings of the intersections at hand, doing little towards the creative resolution of interferences. finally, it is striking how the delta’s actor constellation remains at odds with the functional logic of the layers’ model: whereas this model highlights the foundational importance (and political priority) of the substrate, it seems that the associated actors remain fairly weakly positioned vis-à-vis other subsystems. a further complication here is that ‘soil & water’ and ‘ecosystems’ are governed by hardly overlapping actor constellations. 4.3. disclosing delta diversity (iii): contested waters the third configuration concerns a cluster of intersections most specific to delta areas, namely those related to their marine and fluvial waters. it is characterized by its multitude of interferences, i.e. the conflicts of interests emerging around different uses of these waters. the configuration mainly involves actors from the subsystems transport, agriculture & fishery, recreation and ecosystems, and primarily those who are directly related to the water (shipping, fishery, water recreation, wet nature). even when surrounded with interferences, this configuration offers ample opportunities for the development of integrative spatial concepts. actors’ ambitions are not interferential by themselves, but mainly arise from competing demands on scarce water surface. the delta’s basins are needed for expansion or intensification of shipping and for similar growth ambitions in fishery and water recreation, whilst ‘ecosystems’ actors rather seek to protect some parts of the water. altogether these spatial claims surpass the water surface article_14-18.indd 91 03/03/15 3:06 pm 92 b. pel et al. / intersections in delta development available, which leads to interferences: the increasing number of sailboats causes not only more damage to the fishery installations, but also leads to frictions between recreational sailboats and inland navigation ships. it is interesting to observe how actors contextualize these interferences within a historical interpretation that is widely shared: all of these competing activities are acknowledged to have been economically crucial for a very long time. their coexistence is seen to have become problematic over (approximately) the last twenty years only, however, due to european commission directives on natural protection that decreased the (commercially) available water surface. as they develop relatively autonomously, the interdependencies between the transport, ecosystems, recreation and fisheries subsystems are in general not very strong. interferences tend to arise around specific waters or specific issues, however: nature conservation organizations seek to foreclose economic activities around ecological valuable sandbanks for example, and this yields conflicts with the many fishermen who engage in mussels cultures. the place-specific nature of these interferences is reinforced by the generally regional or local orientation of involved actors. moreover, apart from these localregional orientations, most actors involved in this configuration are focused on the short or midterm. as these time frames tend to be based upon the continuation of the economic activities, interferences are not easily resolved: concessions in this struggle over water resources have fairly immediate financial repercussions. the prevailing frames of time and scale thus seem to exacerbate the interferences between competing water uses. 5. conclusion and discussion of findings as introduced, the difficulty of synchronizing diverse programs of action and underlying system understandings creates a need for integrative spatial concepts. cas-based thinking seems helpful for such integration – primarily when it is deployed to articulate systemic diversity. intersections analysis approaches subsystem interactions in fine-grained fashion, appreciating that the co-evolution between subsystems relies on the intertwinement of actor networks and ultimately on the system understandings of actors. yet it remains a methodological challenge to make the complexity-based understanding operative and productive. the merits of intersections analysis will therefore be evaluated through its results, i.e. its intended inputs for the development of complexitysensitive spatial concepts. comparing the three configurations, the following conclusions can be drawn: 1. all of the configurations can be observed to display mixtures of interaction modes. notwithstanding the circumstance that some configurations can be typified as mainly symbiotic or interferential, none of them exhibits these co-evolution types exclusively. interaction modes are likely to coincide or blend with others – and not seldom can they be turned into their flipside: interferences can be overcome through creative solutions, symbioses may collapse under increasing scarcity, or under involved actors’ zero-sum framings of subsystem interactions. 2. second, it is striking how symbioses and interferences tend to be momentary and location-specific. the delta’s various water-based activities only interfere at particular places, and the symbioses of the ‘living with nature’ configuration can equally be article_14-18.indd 92 03/03/15 3:06 pm b. pel et al. / intersections in delta development 93 considered as only temporary ‘windows of opportunity’ within an otherwise rather interferential actor constellation. for the development of spatial concepts this holds the reminder that tailored designs are called for – also involving geographical specification of what has been inferred along systems-theoretical lines. 3. third, all three configurations display issues of distributive justice. this is most apparent in the ‘settlement & employment’ configuration, in which the emerged juxtaposition of ‘red circle’ vs. ‘blue-green heart’ structures creates cleavages between local development ambitions and supra-local visions that seek to curb those in the service of balanced development at a higher systems level (compare van assche duineveld, beunen, & teampau, 2011). the other two configurations display similarly difficult issues of distributive justice, manifesting especially in the tensions between local economic development, and overall protection of ecological values. these issues pose a caveat to spatial design: spatial concepts better be developed with due consideration of the ways to negotiate the attendant financial aspects. 4. fourth and finally, the historical dimension proves most relevant to the configurations developed. many interferences are rooted in actors’ diverging interpretations of past developments: these give rise to expectations and established rights, but are also invoked for the preservation of ecological values. furthermore, some actors seem to take the emerged spatial-economic distribution as a guideline for further developments, whereas others rather seek to mitigate interregional unevenness. finally, the nationally organized flood protection can be considered to result directly from the 1953 flooding catastrophe, which remains present in the minds of delta inhabitants. these diverging readings of history do not warrant resignation into path dependencies, however. rather, they suggest that spatial concepts should somehow address historically evolved tensions, and express what systemic trends they seek to reinforce or mitigate. the above observations show more concretely how cas-based thinking, combined with techniques from stakeholder analysis, can inform the development of complexitysensitive spatial concepts. systemic configurations, and the intersections they consist of, are shown to be dynamic and contested, and each of the four observations provides guidance for spatial design. intersections analysis can thus be said to put transformative diversity and geographical concreteness (stirling, 2011; coenen & truffer, 2012) into practice: the three configurations elude further condensation into singular transition pathways, but constitute rather irreducible subsystem clusters that account for parts of overall delta evolution. more specifically, even the brief descriptions of the developed configurations reflect attentiveness to situated agency in complex systems (teisman et al., 2009; jørgensen, 2012). arguably, intersections analysis is well-equipped to address the synchronization challenges (edelenbos et al., 2013) through spatial concepts: intersections analysis articulates actors’ different outlooks in terms of time, sector and scale, as foundational inputs to spatial concepts that ultimately are to be carried by such diverse actor constellations. highlighting actors’ either developmental or conservational attitudes, for example, it also brings out how some seek to foster desirable system transitions, whilst others rather anticipate undesirable ones. the attention to this basic distinction meets smith and stirling (2010), who consider it a neglected issue in transitions research. article_14-18.indd 93 03/03/15 3:06 pm 94 b. pel et al. / intersections in delta development notwithstanding these merits, there remains considerable scope for methodological improvement in intersections analysis. first of all, the aim for rich and dynamic representation of subsystem interactions is compromised in the process of converging onto configurations. one way to secure sufficient detail could be to apply a more rigorous differentiation within the interference/symbiosis distinction, such as the 6-fold categorization of sandén and hillman (2011). second, the observed momentary and place-specific character of interferences and symbioses can be taken to warrant greater geographical concreteness. of course, such pursuit of detail may be difficult to combine with analysis on the level of the delta as a whole. still, this issue of comprehensiveness versus detail could be accommodated by inserting targeted analyses of particular systemic ‘hotspots’, as methodological complements (cf. broesi et al., 2013). third, the identification of systemic configurations needs to be distinguished from the actual development of spatial concepts. intersections analysis, however dynamic in outlook, theorizes mainly what under current actor constellations is possible, whilst paying less attention to what may be desirable under future circumstances. this reflects its grounding in present-oriented stakeholder methodologies. also in this respect it deserves to be explored how spatial design, with its typically creative and counterfactual thinking, could function as a complement. more generally, this point indicates the importance of inserting other perspectives into intersections analysis. we have used elements of the method in a participative design process, for example, engaging stakeholders in our cas-based approach to spatial design (cf. pel et al., 2013). apart from its benefits in terms of societal relevance, this process also allowed for practical testing and fine-tuning of the method. to conclude, intersections analysis has definite merits as a complexity-based methodology for the development of integrative spatial concepts, yet it leaves ample scope for further refinements. the shortcomings discussed, such as the tension between comprehensiveness and requisite detail, need not be taken to disqualify the approach altogether: the decomposition of complex systems is unavoidably accompanied with such methodological trade-offs (luhmann, 1995; cilliers, 2005; carpenter, folke, scheffer, & westley, 2009). the signaled merits and shortcomings are therefore of greatest methodological value when balanced in the light of specific aims, and when used to arrive at tailored solutions. references ackoff, r. l. (1974). creating the corporate future. new york, ny: john wiley & sons. allen, p. m. (2011). cities: the visible expression of co-evolving complexity. in j. portugali, v. m. meyer, & e. stolk (eds.), complexity theories of cities have come of age: an overview with implications to urban planning and design (pp. 67–89). berlin, germany: springer. axelrod, r. (2006). the evolution of cooperation. new york, ny: basic books. bertolini, l., frank, a. i., grin, j., bell, s., scholl, b., mattila, h., . . . bertolini, l. (2012). introduction: time to think planning (education)—from marginal interface to central opportunity space? science for practice? educating professionals for practice in a complex world–a challenge for engineering and planning schools project-based learning–core university education in spatial planning and development managing planning pathologies: an educational challenge of the new apprenticeship programme in finland conclusion: time to act. planning theory & practice, 13(3), 465–490. doi: 10.1080/14649357.2012.704712 article_14-18.indd 94 03/03/15 3:06 pm b. pel et al. / intersections in delta development 95 broesi, r., neumann, d., van berkel, s., meyer, h., warmerdam, m., versluijs, h., & van der mijden, j. (2013). ruimtelijke uitwerkingen; samenhangende sets van ruimtelijke ingrepen [spatial effects: coherent sets of spatial interventions] (working document ipdd wp5). publisher: must amsterdam (in dutch) byrne, d. (2005). complexity, configurations and cases. theory, culture & society, 22(5), 95–111 carpenter, s. r., folke, c., scheffer, m., & westley, f. r. (2009). resilience: accounting for the noncomputable. ecology and society, 14(1), 13. cilliers, p. (2005). complexity, deconstruction and relativism. theory, culture & society, 22(5), 255–267. coenen, l., & truffer, b. (2012). places and spaces of sustainability transitions: geographical contributions to an emerging research and policy field. european planning studies, 20(3), 367–374. dammers, e., bregt, a. k., edelenbos, j., meyer, h., & pel, b. (2014). urbanized deltas as complex adaptive systems: implications for planning and design. built environment, 40(2), 156–168. de bruin, j. a., ten heuvelhof, e. f., & in ’t veld, r. j. (1998). process management: on process design and decision-making. schoonhoven, netherlands: academic services. deltaprogram. (2011). deltaprogramma 2012 werk aan de delta: maatregelen van nu, voorbereidingen voor morgen [delta programme 2012, working on the delta: action for today, preparations for tomorrow]. den haag, netherlands: author. (in dutch) dienst landelijk gebied. (2011). goeree-overflakkee in de delta: verkenning van kansen voor water en ruimte [goeree-overflakkee in the delta: exploration of opportunities for water and space]. den haag: dlg (in dutch) disco, c., & van der vleuten, e. (2002). the politics of wet system building: balancing interests in dutch water management from the middle ages to the present. knowledge, technology & policy. 14(4), 21–40. döpp, s., hooimeijer, f., & maas, n. (2011). urban climate framework: a system approach regarding climate proof cities. in l. gerrits & p. marks (eds.), compact i: public administration in complexity (pp. 244–260). litchfield park, az: emergent publications. edelenbos, j. (2000). shaping processes in multifunctional land use projects (doctoral dissertation). lemma: utrecht, netherlands. (in dutch) edelenbos, j., & klijn, e. h. (2009). project versus process management in public-private partnership: relation between management style and outcomes. international public management journal, 12(3), 310–331. edelenbos, j., & teisman, g. r. (2010). symposium on water governance. prologue: water government as a government’s actions between the reality of fragmentation and the need for integration. international review of administrative sciences. 77(1), 5–30. edelenbos, j., bressers, n., & scholten, p. (eds.). (2013). water governance as connective capacity. farnham, england: ashgate. folke, c., pritchard, l., jr., berkes, f., & colding, j. (2007). the problem of fit between ecosystems and institutions: ten years later. ecology and society, 12(1), 30. grin, j., rotmans, j., & schot, j. (2010). transitions to sustainable development: new directions in the study of long term transformative change. new york, ny: routledge. havenbedrijf rotterdam. (2011). havenvisie 2030: port compass. rotterdam, netherlands: author. (in dutch) hodson, m., & marvin, s. (2010). can cities shape socio-technical transitions and would we know if they were? research policy, 39(4), 477–485. holland, j. h. (1995). hidden order: how adaptation builds complexity. new york, ny: basic books. hughes, t. p. (1994). technological momentum. in m. r. smith & l. marx (eds.). does technology drive history? the dilemma of technological determinism (pp. 101–113). cambridge, ma: mit press. integral planning and design in the southwest delta. (2013). retrieved from http://ipdd.verdus.nl/pagina. asp?id=61 intermunicipal collaboration voorne putten. (2011). voorne putten (her)ontdekt! gebiedsvisie voorne putten. publisher: isvp hellevoetsluis (in dutch). jørgensen, u. (2012). mapping and navigating transitions—the multi-level perspective compared with arenas of development. research policy, 41(6), 996–1010. kickert, w. j. m., klijn, e. h., & koppenjan, j. f. m. (eds.). (1997). managing complex networks: strategies for the public sector. london, england: sage. kingdon, j. w. (1995). agendas, alternatives and public policies. new york, ny: addison-wesley. article_14-18.indd 95 03/03/15 3:06 pm 96 b. pel et al. / intersections in delta development koppenjan, j., & klijn, e. h. (2004). managing uncertainties in networks. london, england: routledge. law, j. (1992). notes on the theory of the actor network: ordering, strategy and heterogeneity. lancaster, england: centre for science studies, lancaster university. retrieved from http://www.comp.lancs.ac.uk/ sociology/papers/law-notes-on-ant.pdf luhmann, n. (1995). social systems. stanford, ca: stanford university press. meyer, v. m. (2005). reinventing the dutch delta: complexity and conflicts. built environment, 35(4), 432–451. mitchell, r. k., agle, b. r., & wood, d. j. (1997). toward a theory of stakeholder identification and salience: defining the principle of who and what really counts. academy of management review, 22(4), 853–888. mitroff, i. i. (1983). stakeholders of the organizational mind: towards a new view of organizational policy making. london, england: jossy-bass publishers. odum, e., & barrett, g. w. (1985). fundamentals of ecology. belmont, ca: thomson brooks/cole. pel, b. (2012). system innovation as synchronization: innovation attempts in the dutch traffic management field (doctoral dissertation). erasmus university rotterdam: rotterdam, netherlands. retrieved from http://hdl.handle.net/1765/37286 pel, b. (2014). intersections in system innovation: a nested-case methodology to study co-evolving innovation journeys. technology analysis & strategic management, 26(3), 307–320. pel, b., duijn, m., janssen, m., & edelenbos, j. (2013, july 3–5). constructing delta realities: joint fact finding challenges in serious game design. paper presented at the eighth international conference on interpretive policy analysis, vienna, austria. priemus, h. (2004). from a layers approach towards a network approach: a dutch contribution to spatial planning methodology. planning, practice and research, 19(3), 267–283. rammert, w. (2000). ritardando and accelerando in reflexive innovation, or: how networks synchronize the tempi of technological innovation (tuts-working papers, berlin wp-7-2000). publisher: technical university berlin. raven, r., schot, j., & berkhout, f. (2012). space and scale in socio-technical transitions. environmental innovation and societal transitions, 4, 63–78. reed, m., graves, a., dandy, n., posthumus, h., hubacek, k., morris, j., . . . stringer, l. (2009). who’s in and why? a typology of stakeholder analysis methods for natural resource management. journal of environmental management, 90, 1933–1949. samenwerkingsorgaan hoeksche waard. (2009). structuurvisie hoeksche waard: ruimtelijk plan. publisher: shw hoeksche waard (in dutch) sandén, b., & hillman, k. (2011). a framework for analysis of multi-mode interaction among technologies with examples from the history of alternative transport fuels in sweden. research policy, 40(3), 403–414. savage, g. t., nix, t. w., whitehead, c. j., & blair, j. d. (1991). strategies for assessing and managing organizational stakeholders. academy of management executive, 5(2), 61–75. scharpf, f. (1997). games real actors play: actor-centered institutionalism in policy research. boulder: westview. scheffer, m. (2009). critical transitions in nature and society. princeton university press. smith, a., & stirling, a. (2010). the politics of social-ecological resilience and sustainable socio-technical transitions. ecology and society, 15(1), 11. späth, p., & rohracher, h. (2010). ‘energy regions’: the transformative power of regional discourses on socio-technical futures. research policy, 39(4), 449–458. stirling, a. (2011). pluralising progress: from integrative transitions to transformative diversity. environmental innovation and societal transitions, 1(1), 82–88. susskind, l., & cruikshank, j. (1987). breaking the impasse: consensual approaches to resolving public disputes. new york, ny: basic books. teisman, g. r., & edelenbos, j. (2011). towards a perspective of system synchronization in water governance. international review of administrative sciences, 1, 101–118. teisman, g., van buuren, m. w., & gerrits, l. (eds.). (2009). managing complex governance systems: dynamics, self-organization and coevolution in public investments. new york, ny: routledge. article_14-18.indd 96 03/03/15 3:06 pm b. pel et al. / intersections in delta development 97 termeer, c., dewulf, a., & van lieshout, m. (2010). disentangling scale approaches in governance research: comparing monocentric, multilevel and adaptive governance. ecology and society, 15(4), 29. uhl-bien, m., marion, r., & mckelvey, b. (2007). complexity leadership theory: shifting leadership from the industrial age to the knowledge age. the leadership quarterly, 18(3), 298–318. ulrich, w. (1983). critical heuristics of social planning: a new approach to practical philosophy. bern, switzerland: haupt. van assche, k., duineveld, m., beunen, r. & teampau, p. (2011). delineating locals: transformations of knowledge/power and the governance of the danube delta. journal of environmental policy & planning, 13(1), 1–21. van buuren, m. w., pel, b., verkerk, j., & edelenbos, j. (2014). werkdocument ipdd wp4 deel 1: governance systeemanalyse ipdd [working document wp4 ipdd part 1: governance system analysis ipdd]. (in dutch) van de ven, g. p. (ed.). (2004). man-made lowlands: history of water management and land reclamation in the netherlands. utrecht, netherlands: matrijs. van leeuwen, c., & van buuren, m. w. (2013). connecting time spans in regional water governance: managing projects as stepping-stones to a climate proof delta region. in j. edelenbos, n. bressers, & p. scholten (eds.), water governance as connective capacity (pp. 191–210). farnham, england: ashgate. van meerkerk, i., van buuren, a., & edelenbos, j. (2013). the influence of water managers’ boundary judgments on the (non)realization of adaptive water governance: an analysis of the dutch case haringvliet sluices. water resource management, 27, 2179–2194. van schaick, j., & klaasen, i. (2011). the dutch layers approach to spatial planning and design: a fruitful planning tool or a temporary phenomenon? european planning studies, 19(10), 1775–1796. vanelslander, t., kuipers, b., hintjens, j., & van der horst, m. (2011). ruimtelijk-economische en logistieke analyse: de vlaams-nederlandse delta in 2040. antwerpen, belgium: university of antwerp. walby, s. (2007). complexity, systems theory, and multiple intersecting social inequalities. philosophy of the social sciences, 37(4), 449–470. wereld natuurfonds. (2010). met open armen: voor het belang van veiligheid, natuur en economie [with open arms: the importance of safety, nature and economy]. zeist, netherlands: author. (in dutch) zonneveld, w., & verwest, f. (2005). tussen droom en retoriek [between dream and rhetoric]. den haag, netherlands/rotterdam, netherlands: ruimtelijk planbureau/nai uitgevers. (in dutch) article_14-18.indd 97 03/03/15 3:06 pm article_14-18.indd 98 03/03/15 3:06 pm legitimacy and legitimation practices: an analysis of tsmo networks 1 complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 university of bamberg press legitimacy and legitimation practices: an analysis of tsmo networks author: takuma shibaike university of calgary email: takumi.s.com@gmail.com private transnational organizations have grown in number and in influence. however, sociologists and political scientists often study them separately, either as transnational social movement organizations (tsmos) or the larger category of international non-governmental organizations (ingos). in this paper, i examine the determinants of tsmo legitimacy by drawing on the literature on ingos. in so doing, i call for bridging the disciplinary gap between sociology and political science. empirically, i find that legitimation benefits already prominent organizations more than those that are not. networking thus helps reproduce the hierarchy among the tsmos, challenging the earlier notion that tsmos are horizontally networked. however, i also find that southern tsmos are more likely to gain legitimacy than northern tsmos once they are visible to their peers. the analysis of tsmos thus cautions our bias to study northern ingos and generalize the findings to ingo population. overall, my findings reveal that the incentives and strategies that ingo research has documented exist among tsmos despite their counter-hegemonic ambitions. keywords: international nongovernmental organizations; organizational legitimacy; organizational networks introduction private, non-profit transnational organizations are regularly engaged in the pursuit of social change, working on human rights, eco-systems, gender equity, and more. the work of such transnational organizations has been noted in seminal studies of transnational social movements (boli & thomas, 1999; keck & sikkink, 1998; tarrow, 2005). the typical picture of these groups is of principled actors seeking to hold the powerful to account (smith & wiest, 2012). existing research documents that some private organizations consistently wield vast authority, commanding the attention of global media outlets, policymakers, corporations, and their peers (bob, 2005; carpenter, 2014; dauvergne & lebaron, 2014; stroup & wong, 2017). both sociologists and political scientists have extensively analyzed these organizations but used different terms to describe them, with the former preferring the term “transnational social movement organization” (tsmo) and the latter employing “international nongovernmental organization” (ingo). conceptually speaking, tsmos are a special subset of ingos, defined as ingos “explicitly involved in work to change the dominant political and social order” (smith & wiest, 2012: 46). as ingos include prosystem organizations, tsmos and ingos may seem to fall into different conceptual categories. however, little empirical investigation has been done to demonstrate the differences. the question is not just about the terminology of transnational private organizations. the difference in the terminology also signifies different ways in which research is conducted among sociologists and political scientists. while sociologists have studied the processes in which tsmos organize progressive movements (caniglia, 2001; smith, 2001, 2002; smith, chatfield, & pagnucco, 1997; smith, gemici, plummer, & hughes, 2018; smith, plummer, & hughes, 2017), research on the differences in organizational legitimacy has been relatively scarce. by contrast, political scientists have focused on the differences in organizational legitimacy, demonstrating the effects of ingo legitimacy on international norms and regulations (bob, 2005; carpenter, 2014; stroup & wong, 2017). the analysis of organizational legitimacy is important because neither tsmos nor ingos have coercive power to singlehandedly affect social changes. instead, they must rely on legitimacy—a right to act—acknowledged by their audiences. although the analysis of social movement processes offers a useful mailto:takumi.s.com@gmail.com http://dx.doi.org/10.20377/cgn-114 2 university of bamberg press insight into tsmo behaviors, the analysis of organizational legitimacy helps us understand the effect of tsmos on the broader political and social phenomena. taken together, i ask why some tsmos gain legitimacy while others do not. in so doing, i draw on the literature on ingo research. this paper has two main contributions. first, it reveals the extent to which the choices of tsmos influence their own legitimacy. in recent years, the legitimacy of civil society as a whole has been challenged (deloffre & schmitz, 2019), but we know little about how individual organizations build or defend their legitimacy. this paper evaluates whether tsmos are capable of shaping their own legitimacy through strategic choices, or their legitimacy is shaped by structural pressure exerted by shared expectations about organizations (dimaggio & powell, 1983). second, this paper emphasizes the need to bring the literatures on ingos and tsmos closer together. to do this, i leverage a newly updated tsmo dataset1 and evaluate whether explanations for ingo legitimacy are able to account for the variation in tsmo legitimacy. in particular, i examine whether the anti-systemic goals of tsmos challenge the current understandings of ingos. the results of my analysis below show that legitimation through networking (i.e. claiming a connection to another organization) is relatively rare but effective among tsmos. my empirical evidence suggests that networking is a deliberate attempt to leverage connectivity for organizational legitimacy rather than driven by functional needs. however, i also find that the effect of networking on organizational legitimacy is larger for prominent tsmos than for low-profile ones, suggesting that the existing hierarchy among tsmos is persistent. while some scholars characterize tsmo networks as horizontal (smith, 2002; smith et al., 2018), my findings support the view that the division and inequality are salient characteristics of tsmo networks (hughes, paxton, quinsaat, & reith, 2018). these findings are consistent with the established canon explaining ingo behaviors (lake & wong, 2009; murdie, 2013), underscoring the benefit of studying ingos and tsmos together the following sections are organized as follows. first, i define tsmos and discuss how i conceptualize and measure tsmo legitimacy. second, i briefly review the network-based studies of ingos and tsmos to motivate my analysis. i then empirically evaluate arguments about organizational legitimacy through quantitative analysis. i demonstrate that legitimacy is primarily associated with legitimation attempts. however, legitimation advantages already prominent tsmos rather than those that are not, helping the reproduction of hierarchy among the tsmo community. the final section discusses the implications of my findings. understanding the legitimacy of transnational social movement organizations the concept of tsmos was introduced more than two decades ago by jackie smith and her colleagues (smith et al., 1997). smith and wiest (2012: 46) define tsmos in terms of their goals and objectives. a transnational private group is a tsmo if it is “explicitly involved in work to change the dominant political and social order.” the definition of tsmos by itself does not imply ideological orientations, but smith and wiest (2012) recognize that they are mostly progressive and even antisystemic, since right-wing, nationalist groups tend to avoid publicizing their work (exceptionally, see bob (2012) for global right-wing mobilizations). organizations whose primary focus is research, religion, service provision, or education are also excluded (smith & wiest, 2012: 70). tsmos are thus a subset of the national and ingo population with an anti-systemic focus. currently, there is no database that catalogs the entire population of ingos. scholars often use the yearbook of international organizations as a relatively comprehensive dataset on ingos (for its 1 smith, jackie, dawn wiest, melanie m. hughes, samantha plummer, brittany duncan. 2017. transnational social movement organizations dataset (tsmod), 1953-2013. [computer file]. pittsburgh, pa: world historical dataverse [distributor]. and hughes, melanie m., samantha plummer, and jackie smith. 2017. transnational social movement organization networks (tsmonet), version i, 1993, 2003, & 2013. [computer file]. pittsburgh, pa: world historical dataverse [distributor]. complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 3 university of bamberg press sampling bias, see bush and hadden’s (2019) comparison with us charity database). the yearbook updates every year and covers both intergovernmental organizations (igos) and private organizations. for private organizations to be included, they must hold three or more offices internationally or conduct internationally-oriented activities. however, the yearbook does not publish its data longitudinally or in a manner that is amenable to cross-sectional studies. in building the initial tsmo dataset, smith and her colleagues reviewed the historical records of the yearbook and sampled the organizations that fit the above definition. in the context of transnational organizations, legitimacy is defined as the belief that an actor is right to act in the eyes of audiences. it does not mean that legitimate actors always induce behavioral compliance by their audiences (lake, 2009), but their claims will be heard and taken seriously. if the brookings institution offers a view on american politics, for example, other think tanks may contrast their positions with brookings’ to legitimate them as challengers. as hudson (2001: 348) points out, legitimacy is “a socially-constructed quality that can be ascribed to an ngo by stakeholders coming from different perspectives.” as such, legitimacy is inherently relational, must be conferred by others, and cannot be claimed singlehandedly. although the study of tsmos tends to assume that any tsmo is legitimate, substantial differences exist between individual tsmos. to measure the legitimacy of tsmos, i use a dataset on networks of tsmos (smith et al., 2018). this network data also comes from the yearbook, but importantly, network ties are self-reported collaboration between tsmos. the nature of self-reporting could harm measurement validity if we analyze actual cooperation and collaboration among tsmos. however, selfreporting works well for the purpose of measuring organizational legitimacy. for example, if a reports a tie to b, b should also have collaboration with a, but this does not necessarily mean that b reports a tie with a. tsmos do not have to validate their ties when reporting to the yearbook, so they may aspirationally name other organizations as a way to legitimate themselves before their peers. conversely, even if a tsmo works with other organizations, it may not mention some unimportant ones. from this vantage point, the number of incoming ties can be interpreted as an indicator of organizational legitimacy. the more legitimate a tsmo, the more likely it is to have incoming ties from other organizations seeking to benefit from association with a legitimate actor. this legitimation practice is widely adopted by private organizations, including for-profit firms, which often use association with high-status firms as a quality signaling in market settings (kilduff & krackhardt, 1994; podolny, 1993). to use a more concrete example of amnesty international, tsmos may be interested in amnesty international’s global reach and stated commitment to representation of diverse views on human rights. other might appreciate amnesty’s effectiveness in securing attention from global media outlets (powers, 2018). whatever the reason may be, tsmos like amnesty are likely to end up with having more incoming ties than outgoing ties because peers want to leverage connections with legitimate organizations. conversely, i use outgoing ties to approximate the intensity of legitimation efforts. my use of peer networks builds upon existing efforts to use incoming ties as an indicator of organizational legitimacy. mitchell and stroup (2017) measure organizational reputation based on the self-reporting of ingos during interviews with ingo staff. similarly, murdie and davis (2012) create a matrix of ingo network ties based on the yearbook data, and stroup and wong (2017) provide a ranking of ingos in terms of authority based on the number of incoming ties. to be sure, legitimacy may be observable in other places. for example, differences in ingo legitimacy can be observed from the frequency by which media outlets quote ingos: legitimate ingos are more often cited to boost the credibility of news stories (thrall, stecula, & sweet, 2014). however, a focus on peer networks has two major advantages. first, data on nonpeer organizations would introduce selection bias. we cannot observe tsmos if they do not receive any attention, even if they are engaged in legitimation practices. the tsmo dataset allows us to observe negative cases since all tsmos are observable regardless of the attention they receive. second, because of counter-hegemonic ambitions, complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 4 university of bamberg press tsmos may be less interested in the response of pro-system actors. pro-system actors, such as states, corporations, and ingos that frequently partner with states and corporations, typically undermine rather than help transnational social movements (smith et al., 2017). this context in effect creates a hard test for my hypotheses based on ingo research. because ingos include pro-system actors, if the insights of ingo research are able to account for tsmo legitimacy among peers, they should also work when we analyze tsmo legitimacy among other groups. understanding legitimacy in transnational private organizations existing research has adopted network perspectives to understand the legitimacy of ingos. i highlight four findings from network theories to motivate my analysis of tsmo networks. first, the choice of strategies within networks can shape structural properties, such as international norms, which in turn affect the legitimacy of agents. for example, goddard (2006) argues that entrepreneurial agents in international networks are able to effect changes in normative structures. others have explored how organizational strategies, in the context of networks, affect advocacy output (luxon & wong, 2017) or their ability to affect international norms (wong, 2012). although ingos and tsmos are constrained by international structures, network theory allows us to examine if and how organizations can effect changes in broader social structures. second, networks between ingos and other types of actors can be seen as co-constitutive of legitimacy. for instance, given substantial attention to the “democratic deficit” in global governance, igos are increasingly “opening up” to ingo participation (bernauer, böhmelt, & koubi, 2013; tallberg, sommerer, & squatrito, 2013). ingos generate accountability for igos, and in turn, participating ingos increase their voice in global governance. research has documented that some tsmos also leverage their networks to increase their legitimacy. smith (2001) argues that appeals to international law provided legitimacy to the tsmos protesting the wto in seattle. similarly, the work of environmental tsmos was legitimated by national elites and marginalized states (caniglia, 2001; smith, 2014). third, networks among ingos are frequently analyzed as attempts to build legitimacy. in the analysis of what she terms free-riding behavior, murdie (2014b) argues that outgoing ties by human rights ingos to other ingos are attempts to increase their legitimacy. carpenter (2011) also argues that agenda setting by ingos at the center of advocacy networks create “bandwagoning effect” where smaller ingos join the coalition and bolster their legitimacy by joining the advocacy winners. finally, the effect of networks on legitimacy is well-established in sociology, although the focus tended to be domestic organizations. kamens (1977) shows how universities and colleges create certain discourse about education that legitimates particular organizational structures. organizations also leverage their connections to influential companies and local communities (galaskiewicz, 1979). similar to igo-ingo relations, corporations and charitable organizations have mutually co-constitutive legitimacy (galaskiewicz, wasserman, rauschenbach, bielefeld, & mullaney, 1985; miles & cameron, 1982). others show how the structure of networks affected the legitimacy of smallto medium-sized corporations in the united states (human & provan, 2000). in short, the existing network literature on ingos and organizational sociology has demonstrated how ties among actors, while also useful for other purposes, are mechanisms for building and defending legitimacy. complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 5 university of bamberg press who is legitimate? in this section, i describe different levels of legitimacy among tsmos, using the latest tsmo dataset.2 the dataset by smith et al. (2018) has two files. one file reports a variety of organizational attributes, including the location of headquarters, the issue areas of operation, and the years of founding, every odd number year from 1981 to 2013. the other file reports tsmo networks in 1993, 2003, and 2013.3 these networks capture all sorts of “horizontal,” collaborative self-reported connections between tsmos. in other words, they do not indicate formal hierarchies, such as founder-founded relationship or official membership to an umbrella organization.4 as i discussed above, however, the nature of selfreported ties creates a de facto hierarchy among tsmos in terms of organizational legitimacy. both files are based on the entries of the yearbook. i merged these two data files as cross-sectional data to overcome observation bias. because data on tsmo networks only include organizations that have ties, they miss a large number of organizations that are not involved in any networking efforts. to avoid the selection bias of observing only positive cases, i needed to compare tsmos that engaged in networking and ones that did not. thus, my analysis treats incoming and outgoing ties as organizational attributes. since network data are available only in 1993, 2003, and 2013, i also took the subset of organizational data in these years. the number of organizations in my dataset is 4,946 (1,045, 1,879, and 2,022 in 1993, 2003, and 2013, respectively). figure 1: histogram for legitimacy of tsmos descriptively, the tsmos that are legitimate or involved in legitimation are relatively uncommon. we can see this from both incoming ties and outgoing ties (see table a1 for summary statistics). as figure 1 shows, the majority of the tsmos receive no tie (73.1%). this pattern by itself challenges the widespread portrayal of tsmos as a highly networked community (carroll, 2007; juris, 2008; wolfson & funke, 2017). more importantly, it suggests that organizational legitimacy is rare, or very few tsmos are legitimate enough to induce others to claiming association. yet, the finding is consistent with existing 2 the same as above. 3 these years were chosen by the coders, smith et al. (2018). i focus on the ties that are described as any collaborative relationships, coded as “1” in the tsmonet dataset. 4 for more, see the codebook of the tsmonet. hughes, melanie m., samantha plummer, and jackie smith. 2017. transnational social movement organization networks (tsmonet), version i, 1993, 2003, & 2013. [computer file]. pittsburgh, pa: world-historical dataverse [distributor] complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 6 university of bamberg press ingo research. for example, stroup and wong (2017) find that only a handful of ingos have authority before multiple audiences and the vast majority of them have none. in terms of legitimation, the majority of the tsmos also do not send ties (67.1%), as shown in figure 2. some tsmos have extreme values due to the self-reporting nature of the yearbook, suggesting that the strategy of networking is not uniformly distributed among tsmos. the low number of outgoing ties suggests that tsmos do not just “name drop.” it represents costly networking, in which tsmos strategically send ties to other ingos rather than sending them automatically by working in the same issue area. figure 2: histogram for networking effort of tsmos the scarcity of observed networking behavior among tsmos does not necessarily mean that they lack an interest in developing and defending their legitimacy. as a point of comparison to ingos, i use stroup and wong’s (2017) list of top fourteen ingos in terms of the amount of attention paid by a variety of audiences, such as states, corporations, and peer ingos. among the fourteen ingos, eleven are identified as tsmos.5 as table 1 reports, not all leading ingos are highly legitimate in the tsmo community, which suggests that tsmos might be unique in terms of their ideas about what a legitimate organization should look like (see also table a2 for 1993 and 2003 data and figure a1 for network diagram). as discussed above, smith and wiest (2012) define tsmos in terms of their counter-hegemonic goals and objectives, a position that is not always welcomed by other actors, especially states and corporations. below, i take a closer look to investigate which factors explain the legitimacy of tsmos. 5 ten tmos were leading ingos in 1993, and they were 11 in 2003 and 2013. international campaign to ban land mines were not included in1993, presumably because it was too young (established in 1992). complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 7 university of bamberg press table 1: legitimacy ranking of tsmos from 2013 data what explains legitimacy? if some tsmos stake their legitimacy on being counter-hegemonic actors, what are the organizational attributes that satisfy such expectations? here, i derive several hypotheses from the literature on ingos. first, networking behavior, such as making a claim to work with other ingos, can be used as an attempt to increase legitimacy. for example, bob (2005) argues that insurgent groups seek to obtain the endorsement of highly legitimate ingos. i expect that the more they send ties, the more others will respond in kind, thereby creating legitimacy (through incoming ties) for themselves. h1: the more ties a tsmo will send, the more legitimacy it will gain among its peers. another plausible explanation is that affiliation with igos increases tsmo legitimacy by sending credible signals. murdie (2014a) finds that ingos with igo consultative status are more likely to be effective in international development, as consultative status works as a costly signal to local and international donors. it is costly because the requirements for such consultative status are heavy, and an ingo must have the capacity to go through the long process of obtaining consultative status. moreover, consultative status gives ingos standing before states in igos and helps them link to other ingos. murdie (2013) finds that human rights ngos with igo consultative status are more likely to be connected with one another. however, as social movements frequently target igos as neoliberal institutions (ayres, 2004), consultative status may tarnish tsmo legitimacy by signaling pro-system positions. ingo signaling behavior is measured by the number of igo affiliations. complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 8 university of bamberg press h2: a tsmo will gain more legitimacy as the number of igo affiliations increase. the divide between the global north and south is a critical element that counterhegemonic movements seek to address (evans, 2012; smith et al., 2018; smith & wiest, 2012). and yet, the existing literature suggests that the connection with northern ingos is essential for southern ingos to achieve their goals, mainly because northern ingos enjoy larger social and economic resources necessary to influence states and igos (bob, 2005; carpenter, 2007; keck & sikkink, 1998). northern tsmos may therefore induce networking behavior of others. here, tsmos are treated as northern ingos if they have primary or secondary headquarters in the global north. h3: tsmos located in the global north are more legitimate than those located in the global south. finally, organizational age may be associated with tsmo legitimacy, as tsmos may be able to establish their brand associated with their issue areas of operation over time. household names, such as amnesty international and oxfam, were indeed established more than half a century ago. operating for a long time also signals an ability to survive in a changing environment. organizational age is computed as the data year (1993, 2003, or 2013) minus the year of tsmo founding.6 h4: on average, older tsmos are more legitimate than younger tsmos. estimation methods to test the hypothesized relationships, i control for other factors that might confound tsmo legitimacy. first, the size of a tsmo can shape prominence and thus organizational legitimacy. typically, size is seen as budget, but the dataset does not report the budget of tsmos. instead, i use the number of countries where a tsmo operates as the measure for organizational scale, with the assumption that operating in more countries indicates a greater fiscal capacity. second, tsmos working in multiple issue areas may have greater legitimacy because they can bridge different clusters of organizations (granovetter, 1977; murdie & davis, 2012). i therefore control for whether or not a tsmo operates in multiple issue areas. even though the dataset is panel-structured, the variation among the indicators within each tsmo is over time quite low. for example, once a tsmo is established in the global south, it is uncommon to move its headquarters to the global north. therefore, rather than including time fixed effects or lagged dependent variables, i resorted to a pooled regression with organizational fixed effects. organizational fixed effects account for the unobservable, idiosyncratic effects on organizational legitimacy that are specific to each tsmo (vaisey & miles, 2017). it is important to note that the dependent variable is not tie activation itself, but the number of incoming ties. in network terms, it is called indegree centrality. while tie activation is estimated with exponential random graph models (ergms) (cranmer & desmarais, 2011), indegree centrality is often estimated by regression analysis (faris & felmlee, 2011; lyle & smith, 2014). in my case, the dependent variable is a count variable with many zero observations (73.1%). to account for overdispersion and zeroinflation, i adopted a zero-inflated negative binomial (zinb) model. zinb regression assumes that observed zeros are generated by two different, unobservable processes: the zero-inflation stage estimates the latent group membership of the sample that could have never received a tie (structural zeros), while the conditional stage estimates the number of ties for the rest of the sample, some of which happened to receive no tie (incidental zeros) (hendrix & wong, 2013; long, 1997). in my interpretation, structural zeros 6 a few tsmos indicated negative ages, which were dropped from the analysis. complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 9 university of bamberg press may be generated by visibility. in other words, if a tsmo is not visible to its peers, it could never be mentioned. by contrast, incidental zeros may result from the lack of organizational attributes necessary to tsmo legitimacy, even if the organization is known by others. as a robustness check, i also conducted ergms to explicitly account for network dependency (see tables a3 and a4), but the results are consistent with my findings below. determinants of legitimacy table 2 reports the results of regression analysis. observations with missing data were removed, which gives us a total of 3,447 observations. for robustness checks, model 2 (zinb without random intercepts), model 3 (negative binomial model: nb), and model 4 (zero-inflated poisson model: zi poisson) are also reported. note that the zero-inflation stage estimates the membership for the structuralzero group, so negative coefficients mean that such variables are, in short, positively associated with visibility of a tsmo among peers. complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 10 10 university of bamberg press table 2: results of regression analyses on the legitimacy of tsmo complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 11 university of bamberg press across all models, i find robust support for h1 that networking behavior is positively associated with tsmo legitimacy. in the zero-inflation stage, outgoing ties are negatively associated with the structural-zero membership. this means that sending a tie to other tsmos can increase organizational legitimacy among peers and it also increases their visibility among the broader tsmo community. in short, networking as a legitimation practice works to gain both visibility and legitimacy. however, the marginal effect of networking might be extremely limited for smaller tsmos. figure 3 illustrates the marginal effect of networking behavior for two tsmos— amnesty international and international union of catholic esperantists (iuce)—while everything else is held at the mean for each tsmo. the marginal effects differ between these organizations because they are dependent on the values of other independent variables, but they are illustrative of the fact that already legitimate tsmos can easily legitimate themselves with networking, while low-profile ones like iuce have a much more difficult time. figure 3: predicted legitimacy of amnesty international and catholic esperantists collectively, this effect would help reproduce the hierarchy of legitimacy among tsmos despite the repeated claims that tsmos are often horizontally organized (smith, 2002; smith et al., 2018). given this finding, it also makes sense that many tsmos do not engage with networking (see figure 2) because low-profile tsmos, which are the vast majority of tsmos in the dataset, do not gain much legitimacy from it. low-profile tsmos may be better off focusing on local constituencies rather than seeking collaboration with other tsmos, which incurs transaction and administrative costs. i also find support for h2 that igo consultative status increases tsmo legitimacy. the effect of each additional affiliation with an igo is substantially larger than that of networking in the conditional stage. in the zero-inflation stage, the effect is much smaller than networking and not statistically significant; that is, no evidence supports that affiliation with igos by itself can increase the visibility of a tsmo. it is only when the tsmo is known among other tsmos that igo affiliation can increase organizational legitimacy. this may be because, while igos serve as a venue in which tsmos can claim credit over outcome (betsill & corell, 2008), civil society participation in igo meetings may not be accessible to outsider tsmos and thus does not increase organizational visibility. the results also suggest that, despite the critiques that igos are coopted by neoliberalism (bond, 2012; charnovitz, 1996; mooney, 2012; willetts, 2006), affiliation with igos does not damage tsmo legitimacy. perhaps, pragmatism is complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 12 university of bamberg press important for the legitimacy of counter-hegemonic movements, and anti-systemic positions can be more about market positioning. for example, sea shepherd conservation society, which rejects pro-system advocacy methods (eilstrup-sangiovanni & bondaroff, 2014), accepted funds from the dutch national postcode lottery, which is run by a marketing agency.7 contrary to my expectation in h3, having headquarters in the global north is negatively associated with tsmo legitimacy. in the zero-inflation stage, the substantive effect is positive but statistically insignificant. that is, i do not find that northern tsmos are more visible in the tsmo community, perhaps because the majority of tsmos are headquartered in the global north anyway (74.8%). it is still puzzling, however, that they receive fewer ties, ceteris paribus, than southern tsmos once they are known. one could argue, given that the current structure of global governance favors northern ingos (barnett & walker, 2015), southern tsmos that brought themselves to visibility might exhibit exceptional qualities. alternatively, organizations with counter-hegemonic ambitions, such as tsmos, might favor actors from the global south. the result also points out the problem of observation bias. we know that the density of ties among tsmos is higher among northern countries (hughes, peterson, harrison, & paxton, 2009), but it does not follow that northern tsmos are on average more legitimate than southern tsmos. i find limited support for h4, and organizational age has two different effects in the zero-inflation and conditional stages. in the zero-inflation stage, organizational age is negatively associated with the structural zero group, meaning that tsmos are more likely to gain visibility as long as they stay active. surprisingly, however, in the conditional stage, organizational age is negatively associated with legitimacy. there are two possible reasons for this. first, older tsmos have likely amassed long track records that other tsmos can exploit to challenge their legitimacy. second, organizations that have survived for a long time may have developed a narrow, path-dependent vision. if tsmos have been successful in effecting change in the past, they could be trapped in the past success model (yanacopulos, 2015). more broadly, this finding cautions us against extrapolating our knowledge from leading ingos. for example, amnesty international and the friends of the earth were established more than 50 years ago, but these tsmos with high levels of legitimacy should be seen as anomalies rather than typical longstanding organizations. organizational size, measured by the number of countries in which tsmos operate, had a statistically significant effect on organizational legitimacy. this is unsurprising in the sense that large organizations should have relatively large operational budgets that allows them to collaborate with many other tsmos. however, the effect of organizational size on visibility was statistically inconsistent. overall, the marginal effect of organizational size was smaller than networking or igo affiliation. the result thus strongly suggests that being a large organization is not the only determinant of organizational legitimacy. it does seem to help, but strategic actions, such as networking and igo affiliation, have greater impacts on the legitimacy of tsmos. finally, i did not find the evidence of brokerage by tsmos working in multiple issue areas, such as human rights and development. while murdie and davis (2012) find that such "hybrid" organizations play an important role in bridging different network clusters, multi-issue tsmos are not necessarily more legitimate than the rest of tsmos. perhaps, brokerage is not a widely adopted tactic among tsmos, as hadden (2015) finds in the networks of climate change organizations. in appendix 3, i conducted an analysis based on ergms to estimate the chance of legitimation between two tsmos, which generally supports the findings presented here. 7 http://www.theideatree.ca/sea-shepherds-8-3-million-euro-award-and-the-idea-tree/ (accessed june 11, 2019) complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 http://www.theideatree.ca/sea-shepherds-8-3-million-euro-award-and-the-idea-tree/ 13 university of bamberg press conclusion in this paper, i evaluated self-reported networks as proxies for the legitimacy of tsmos. the selfreported nature of tsmo data allowed me to exploit the incentives and disincentives of associating with other organizations in a publicly accessible forum. a tsmo that reports a network tie is, whether genuine or not, engaging in a strategy of legitimation. i demonstrated the insights of ingo research can also explain how different strategies and organizational attributes affect tsmo legitimacy. the main empirical finding is that, while tsmos can improve their legitimacy by leveraging connectivity to other tsmos, such networking helps those that are already prominent rather than lesser known tsmos. there are also a few counter-intuitive findings. among tsmos, southern organizations are more likely to gain legitimacy than northern tsmos and organizational age is negatively associated with legitimacy (although it does help organizational visibility). the analysis begs further research as to why the hierarchy of tsmos is so persistent despite their counter-hegemonic ambitions. i suspect it has to do with the credibility of the claims (gourevitch, lake, & stein, 2012). perhaps, there is a bias in how tsmos interpret claims to have network ties. a legitimate tsmo’s claims to work with other organizations should be true, 1) as their reputations negate their need to make spurious claims about partners; and 2) their prominence makes it more likely that misrepresentations will be reported. such a display of collaborative behavior might further increase the legitimacy of the tsmo, potentially creating “free-riders” (murdie, 2014b). other tsmos, on the other hand, may aspire towards having connections with better-known peers, but paradoxically, the lack of erstwhile interpretation makes it hard for others to interpret whether a reported tie is genuine or not. moreover, even if such tsmos genuinely reported ties, the networked tsmo may not see them as important and legitimate enough to report on a public forum. in short, peer organizations may not want to react to some tsmos based on their claims, but they trust other tsmos because of different levels of legitimacy. this insight of the winners keep on winning seems to be a logic that is true in other social relations, such as corporations and individuals (sauder, lynn, & podolny, 2012). more broadly, structural pressure to maintain the hierarchy among tsmos is persistent, and strategic choices have positive but limited effects on most tsmos. the results also suggest that explanations for ingo behaviors generally work for tsmos despite the latter’s inclination against the status quo, with the possible exception that northern tsmos are less likely to be legitimated than southern tsmos. this finding illuminates the challenges of being an antihegemonic organization. in other words, it is no surprise that tsmos prefer to work with southern organizations in rejection of the current world order. it is more surprising that other determinants of ingo legitimacy, such as igo affiliations, translate to positive effects on tsmo legitimacy. perhaps, tsmos are strategic about how they can achieve counter-hegemonic goals, even if doing so requires some collaboration with pro-system actors. my analysis also contributes to efforts to understand the networking behavior of transnational private organizations. i have shown that sending an outgoing tie is an attempt at legitimation that actually works despite its limited effect. this is a different conclusion from a prominent interpretation in the ingo literature. murdie (2014b) argues that ingos with more outgoing ties than incoming ties are “free-riders.” she finds that ingos that have been less active in advocacy events in the past are more likely to report outgoing ties, which she argues as evidence of these ingos not actively contributing to the network. while murdie may be partially correct that free-riding could happen around extremely prominent tsmos, my finding that outgoing ties increase the expected number of incoming ties offers a different insight. thus, this paper contributes to the call for more research on the strategic concerns of tsmos and ingos. while the structural pressure exerted by shared expectations about tsmos is strong, it does not complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 14 university of bamberg press mean that tsmos are impossible to construct their own legitimacy. recent research on ingos has emphasized the structural properties of issue areas, such as the density of organizational populations, in legitimating certain forms of organizations (abbott, green, & keohane, 2016; bush & hadden, 2019), but the agency of ingos should not be left aside. most importantly, if tsmos collectively act, as often stated in their goals and objectives, their definition of legitimacy may shift in the way that would be more conducive to the emergence of entrepreneurial transnational private organizations. acknowledgement i thank jackie smith, samantha plummer, and melanie hughes for making the latest tsmo dataset available. i also thank wendy wong, sarah stroup, and the participants of conferences at the university of pittsburgh, isa, and arnova for constructive comments. references abbott, kenneth w., green, jessica f., & keohane, robert o. (2016). organizational ecology and institutional change in global governance. international organization, 70(2), 247277. ayres, jeffrey m. (2004). framing collective action against neoliberalism: the case of the antiglobalization movement. journal of world-systems research, 10(1), 11-34. barnett, michael, & walker, peter. (2015). regime change for humanitarian aid. foreign affairs, 94, 130-141. bernauer, thomas, böhmelt, tobias, & koubi, vally. (2013). is there a democracy–civil society paradox in global environmental governance? global environmental politics, 13(1), 88107. betsill, michele merrill, & corell, elisabeth. (2008). ngo diplomacy: the influence of nongovernmental organizations in international environmental negotiations. cambridge, ma: mit press. bob, clifford. (2005). the marketing of rebellion: insurgents, media, and international activism. new york, ny: cambridge university press. ———. (2012). the global right wing and the clash of world politics. new york, ny: cambridge university press. boli, john, & thomas, george m. (1999). constructing world culture: international nongovernmental organizations since 1875: stanford university press. bond, patrick. (2012). politics of climate justice: paralysis above, movement below. scottsville, south africa: university of kwazulunatal press. bush, sarah sunn, & hadden, jennifer. (2019). density and decline in the founding of international ngos in the united states. international studies quarterly, 63(4), 11331146. caniglia, beth. (2001). informal alliances vs. institutional ties: the effects of elite alliances on environmental tsmo networks. mobilization: an international quarterly, 6(1), 37-54. carpenter, r. charli. (2007). setting the advocacy agenda: theorizing issue emergence and nonemergence in transnational advocacy networks. international studies quarterly, 51(1), 99-120. ———. (2011). vetting the advocacy agenda: network centrality and the paradox of weapons norms. international organization, 65(1), 69-102. ———. (2014). "lost" causes: agenda setting and agenda-vetting in the global issues networks. ithaca, ny: cornell university press. carroll, william. (2007). hegemony and counter-hegemony in a global field. studies in social justice, 1(1), 36. charnovitz, steve. (1996). two centuries of participation: ngos and international governance. michigan journal of international law, 18, 183-286. cranmer, skyler j., & desmarais, bruce a. (2011). inferential network analysis with exponential random graph models. political analysis, 19(1), 66-86. dauvergne, peter, & lebaron, genevieve. (2014). protest inc.: the corporatization of activism. cambridge, uk: polity press. deloffre, maryam zarnegar, & schmitz, hans peter. (2019). ingo legitimacy: challenges and responses. in t. davies (ed.), routledge handbook of ngos and international relations: routledge. dimaggio, paul j., & powell, walter w. (1983). the iron cage revisited: institutional isomorphism and collective rationality in organization fields. american sociological review, 48(2), 147-160. evans, peter. (2012). counter-hegemonic globalization. in g. ritzer (ed.), the wiley-blackwell encyclopedia of globalization. hoboken, nj: wiley-blackwell. complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 15 university of bamberg press faris, robert, & felmlee, diane. (2011). status struggles: network centrality and gender segregation in same-and cross-gender aggression. american sociological review, 76(1), 48-73. galaskiewicz, joseph. (1979). the structure of community organizational networks. social forces, 57(4), 1346-1364. galaskiewicz, joseph, wasserman, stanley, rauschenbach, barbara, bielefeld, wolfgang, & mullaney, patti. (1985). the influence of corporate power, social status, and market position on corporate interlocks in a regional network. social forces, 64(2), 403-431. goddard, stacie e. (2006). uncommon ground: indivisible territory and the politics of legitimacy. international organization, 60(1), 35-68. gourevitch, peter a., lake, david a., & stein, janice gross (eds.). (2012). the credibility of transnational ngos: when virtue is not enough. new york, ny: cambridge university press. granovetter, mark s. (1977). the strength of weak ties american journal of sociology (vol. 78, pp. 1360-1380): elsevier. hadden, jennifer. (2015). networks in contention: the divisive politics of global climate change. new york, ny: cambridge university press. hendrix, cullen s., & wong, wendy h. (2013). when is the pen truly mighty? regime type and the efficacy of naming and shaming in curbing human rights abuses. british journal of political science, 43(3), 651-672. hudson, alan. (2001). ngos’ transnational advocacy networks: from ‘legitimacy’to ‘political responsibility’? global networks, 1(4), 331-352. hughes, melanie m., paxton, pamela, quinsaat, sharon, & reith, nicholas. (2018). does the global north still dominate the international women’s movement? a network analysis of women’s international nongovernmental organizations, 1978–2008. mobilization: an international quarterly, 23(1), 1-21. hughes, melanie m., peterson, lindsey, harrison, jill ann, & paxton, pamela. (2009). power and relation in the world polity: the ingo network country score, 1978–1998. social forces, 87(4), 1711-1742. human, sherrie e., & provan, keith g. (2000). legitimacy building in the evolution of smallfirm multilateral networks: a comparative study of success and demise. administrative science quarterly, 45(2), 327-365. juris, jeffrey s. (2008). networking futures: the movements against corporate globalization: duke university press. kamens, david h. (1977). legitimating myths and educational organization: the relationship between organizational ideology and formal structure. american sociological review, 208-219. keck, margaret, & sikkink, kathryn. (1998). activists beyond borders: advocacy networks in international politics. ithaca, ny: cornell university press. kilduff, martin, & krackhardt, david. (1994). bringing the individual back in: a structural analysis of the internal market for reputation in organizations. academy of management journal, 37(1), 87-108. lake, david a. (2009). hierarchy in international relations. ithaca, ny: cornell university press. lake, david a., & wong, wendy h. (2009). the politics of networks: interests, power, and human rights norms. in m. kahler (ed.), networked politics: agency, power, and governance. ithaca, ny: cornell university press. long, j. scott. (1997). regression models for categorical and limited dependent variables. thousand oaks, ca: sage publications. luxon, emily m., & wong, wendy h. (2017). agenda-setting in greenpeace and amnesty: the limits of centralisation in international ngos. global society, 1-31. lyle, henry f., & smith, eric a. (2014). the reputational and social network benefits of prosociality in an andean community. proceedings of the national academy of sciences, 111(13), 4820-4825. miles, robert h, & cameron, kim s. (1982). coffin nails and corporate strategies: prentice hall. mitchell, george e., & stroup, sarah s. (2017). the reputations of ngos: peer evaluations of effectiveness. the review of international organizations, 12(3), 397-419. mooney, pat. (2012). civil society strategies and the stockholm syndrome. development dialogue, 61(september), 327-331. murdie, amanda. (2013). the ties that bind: a network analysis of human rights international nongovernmental organizations. british journal of political science, 44(1), 1-27. ———. (2014a). help or harm: the human security effects of international ngos. stanford, ca: stanford university press. ———. (2014b). the ties that bind: a network analysis of human rights international nongovernmental organizations. british journal of political science, 44(01), 1-27. murdie, amanda, & davis, david. (2012). shaming and blaming: using events data to assess the impact of human rights ingos. international studies quarterly, 56(1), 1-16. podolny, joel m. (1993). a status-based model of market competition. american journal of sociology, 98(4), 829-872. powers, matthew. (2018). ngos as newsmakers: the changing landscape of international news. new york, ny: columbia university press. sauder, michael, lynn, freda, & podolny, joel m. (2012). status: insights from organizational sociology. annual review of sociology, 38, 267-283. smith, jackie. (2001). globalizing resistance: the battle of seattle and the future of social movements. mobilization: an international quarterly, 6(1), 1-19. complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 16 university of bamberg press ———. (2002). bridging global divides? strategic framing and solidarity in transnational social movement organizations. international sociology, 17(4), 505-528. ———. (2014). counter-hegemonic networks and the transformation of global climate politics: rethinking movement-state relations. global discourse, 4(2-3), 120-138. smith, jackie, chatfield, charles, & pagnucco, ron. (1997). transnational social movements and global politics: solidarity beyond the state. syracuse, ny: syracuse university press. smith, jackie, gemici, basak, plummer, samantha, & hughes, melanie m. (2018). transnational social movement organizations and counter-hegemonic struggles today. journal of world-systems research, 24(2), 372-403. smith, jackie, plummer, samantha, & hughes, melanie m. (2017). transnational social movements and changing organizational fields in the late twentieth and early twentyfirst centuries. global networks, 17(1), 3-22. smith, jackie, & wiest, dawn. (2012). social movements in the world-system: the politics of crisis and transformation. new york, ny: russell sage foundation. stroup, sarah s., & wong, wendy h. (2017). the authority trap: strategic choices of international ngos. ithaca, ny: cornell university press. tallberg, jonas, sommerer, thomas, & squatrito, theresa. (2013). the opening up of international organizations. cambridge, uk: cambridge university press. tarrow, sidney. (2005). the new transnational activism. new york, ny: cambridge university press. thrall, a. trevor, stecula, dominik, & sweet, diana. (2014). may we have your attention please? human-rights ngos and the problem of global communication. the international journal of press/politics, 19(2), 135-159. willetts, peter. (2006). the cardoso report on the un and civil society: functionalism, global corporatism, or global democracy? global governance: a review of multilateralism and international organizations, 12(3), 305-324. wolfson, todd, & funke, peter n. (2017). contemporary social movements and media: the emergent nomadic political logic and its nervous system. wong, wendy h. (2012). internal affairs: how the structure of ngos transforms human rights. ithaca, ny: cornell university press. yanacopulos, helen. (2015). international ngo engagement, advocacy, activism: the faces and spaces of change: springer. complexity, governance & networks – vol. 8, no. 1 (2022), p. 1-24 doi: http://dx.doi.org/10.20377/cgn-114 complexity, governance & networks – vol.8, no. 1 (2022), p.1-24 17 doi: 10.20377/cgn-114 university of bamberg press appendix 1: statistic n mean sd min max # of incoming ties 4,938 0.607 1.711 0 30 # of outgoing ties 4,938 1.448 5.439 0 181 igo membership 3,651 2.053 3.159 0 44 hq north 4,756 0.744 0.436 0 1 organizational age 4,756 24.672 21.687 0 174 multiple issue areas 4,938 0.212 0.409 0 1 country count 4,919 21.837 28.605 0 188 table a1: summary statistics 1993 table a2: legitimacy ranking of tsmos complexity, governance & networks – vol.8, no. 1 (2022), p.1-24 18 doi: 10.20377/cgn-114 university of bamberg press 2003 table a2: legitimacy ranking of tsmos complexity, governance & networks – vol.8, no. 1 (2022), p.1-24 19 doi: 10.20377/cgn-114 university of bamberg press 1993 figure a1: network diagrams green nodes are leading ingos identified by stroup and wong (2017). node size is proportionate to legitimacy (# of incoming ties). isolates are removed. graphs were generated by the force atlas algorithm in gephi. complexity, governance & networks – vol.8, no. 1 (2022), p.1-24 20 doi: 10.20377/cgn-114 university of bamberg press 2003 figure a1: network diagrams green nodes are leading ingos identified by stroup and wong (2017). node size is proportionate to legitimacy (# of incoming ties). isolates are removed. graphs were generated by the force atlas algorithm in gephi. complexity, governance & networks – vol.8, no. 1 (2022), p.1-24 21 doi: 10.20377/cgn-114 university of bamberg press 2013 figure a1: network diagrams green nodes are leading ingos identified by stroup and wong (2017). node size is proportionate to legitimacy (# of incoming ties). isolates are removed. graphs were generated by the force atlas algorithm in gephi. complexity, governance & networks – vol.8, no. 1 (2022), p.1-24 22 doi: 10.20377/cgn-114 university of bamberg press appendix 2: exponential random graph models i conduct the analysis of exponential random graph models (ergms) to estimate the chance of having a tie between two tsmos. although the dependent variable is different from the zinb model discussed in the main text, it measures a similar outcome: what drives tsmos to confer legitimacy on another organization. the advantage of ergms is that they can estimate the effects of different logics of tie activation, while explicitly taking account for network structures through markov chain monte carlo simulations (cranmer & desmarais, 2011). ergms assume that the observed network is just one realization of many possible networks and simulate a network many times to evaluate if a tie between two nodes occurs randomly or systematically. first of all, i consider reciprocity. this network statistic estimates the likelihood of receiving a tie back when a tsmo sends a tie to another tsmo: when j sends a tie to i, what is the chance of i sending a tie back to j? in my interpretation, this represents the mechanism of direct legitimation. i also consider three types of network clustering. the first clustering is a mechanism by which tmsos sends a tie because the tie recipient successfully signals its legitimacy through its association with a legitimate tsmo. in network terms, this relationship is called transitive: i sends a tie to j when both i and j send a tie to k. in my interpretation, this represents the case where k is locally seen as a legitimate tsmo, and thus being associated with k increases the chance of legitimation by peers. the second clustering is called cyclical: i sends a tie to j when j sends a tie to k and k sends a tie to i. i doubt this is a common phenomenon because deferential relationship should not be circular. i included this network statistic to evaluate how much more likely transitive relations are to occur relative to other kinds of triadic relations. finally, i also consider network centralization because the network data, as suggested in figures 1 and 2, have highly uneven distribution of ties among tsmos.8 the primary purpose of this network statistic to help markov chain monte carlo estimation converge by constraining the degree distribution of a network. to explore homophily effects, i included headquarter locations (global north) and united nations economic and social council (ecosoc) affiliation, given that these are salient explanations in the regression analysis above. i chose ecosoc since it is an igo that is wellknown and used widely in the study of ingos (murdie, 2014a). i also account for organizational age, which estimates the chance of a tie when age difference between tsmos increase by one year.9 finally, i consider the issue areas of operation to evaluate functionalism. this is statistically equivalent to the inclusion of other organizational attributes, just like the case of homophily. i sperate this mechanism from others because in real-world practices the functional logic that motivates a tie among tsmos should not be organizational similarities per se, but the functional needs to achieve something in the issue area in question. table a3 reports the results of estimations. because the size of a network varies by year, the results are also reported by year. coefficients are changes in the log-odds of a tie, so they can be interpreted in the same way logistic regression coefficients are interpreted. the number of edges serve as a baseline likelihood; when simulation includes this network statistic alone, the likelihood of a tie equals the density of a network, which can be understood as a random assignment of realized ties in the network.10 8 i use indegree distribution since the network is a directed network. 9 more accurately, it estimates a change in log-odds of a tie. 10 network density = # of the observed ties / # of all possible ties. in a way, this is like the “intercept” of regression analysis. complexity, governance & networks – vol.8, no. 1 (2022), p.1-24 23 doi: 10.20377/cgn-114 university of bamberg press table a3: results of ergm for 1993, 2003, and 2013 across all years, i find that a tie is much more likely to occur as a result of reciprocity than the other mechanisms i examined. cyclical relations are statistically significant, but the coefficient is negative. this means that a tie is unlikely in a cyclical relationship, which is in line with my argument about how legitimation works. in fact, transitive relations have the second largest effect among independent variables. that is, association with a legitimate organization does help increase peer legitimacy. to look at the effects more carefully, i derive the chance of tie activation from the cumulative probability distribution of the latent variables of interest. in 2013, the chance of a tie in any given dyad (i.e., baseline likelihood) is 0.1%. the very low baseline is expected given that the network is both large and directed. the chance of a tie, however, increases to 16.7% when i consider reciprocity. that is, when a sends a tie to b, b sends a tie back to a with a 16.7% chance.11 this is significantly higher than that of transitivity (0.7%), in which association with a legitimate tsmo increases the chance of tie activation. thus, networking increases the legitimacy of a tsmo directly rather than indirectly. signaling through a third party seems a reasonable strategy at first glance, but perhaps because peers are more knowledgeable about other organizations around them, it does not have a substantively important effect. the chance of a tie via homophily is also quite low. the ecosoc affiliation is negatively associated with tie activation, and the statistical significance is inconsistent (not significant in 2013). a tie is negligibly more likely among northern tsmos, and so is the effect of organizational age. however, the 11 the chance of reciprocity is very high in 1993. i suspect this is due to a smaller size of the network (441 vs. 509, 727). note that network matrices expand quickly with an additional node. complexity, governance & networks – vol.8, no. 1 (2022), p.1-24 24 doi: 10.20377/cgn-114 university of bamberg press chance of a tie via functionalism is slightly higher, 0.4%. overall, while i confirm that different mechanisms are at work in tie activation, direct networking is by far the most effect way of increasing legitimacy. table a3 summarizes the estimated chance of tie activation. year baseline reciprocity northern tsmo issue area 1993 0.1% 67.4% 0.1% 0.3% 2003 0.1% 54.2% 0.2% 0.7% 2013 0.1% 15.1% 0.1% 0.4% table a4: probability of a tie based on reciprocity and shared attributes isomorphic properties of network governance: comparing two watershed governance initiatives in the lake champlain basin using institutional network analysis complexity, governance & networks (2014) 99–118 99 doi: 10.7564/14-cgn12 isomorphic properties of network governance: comparing two watershed governance initiatives in the lake champlain basin using institutional network analysis christopher kolibaa, *, adam reynoldsb, asim ziac, steven scheinertd university of vermont e-mail a: ckoliba@uvm.edu; email b: azreynol@uvm.edu; email c: azia@uvm.edu; email d : sscheine@uvm.edu in this paper a comparison of the two planned networks that appear in watershed planning documents for the lake champlain basin in 2010 one plan (2010 tmdl) was developed by a regulatory network initiated by the united states environmental protection agency (usepa) and state legislature. the second plan (2010 ofa) was developed by a watershed partnership network spanning the governmental, nonprofit, and business sectors. this paper asks if these two planning networks reify themselves in the plans they create? the extent to which the structural and functional properties of the networks in this study are mirrored in the plans that they produce is measured. using textual data mining techniques and institutional network analysis the authors examine measures of network centrality, develop a visual analysis of network structures and clusters, and examine statistical comparisons of the task structures found across the two planned networks. institutional isomorphism theory is used to anticipate and explain any mirroring effects observed in the data. a comparison of policy tool identification, actor characteristics, and task structures for each plan is rendered. findings suggest evidence of structural isomorphism, but not policy tool isomorphism occurring between the two planning regimes and possible explanations for these findings are given. key words: network governance, policy tools, partnership networks, regulatory networks, institutional isomorphism those who have charted the trajectory of environmental policy and governance in democratic societies have noted the growing role that multi-institutional networks have played in carrying out a wide range of policy functions (durant, fiorino, & o’leary, 2004; gerlak, 2006). the environmental network literature has included studies of networks * corresponding author. article_14-12.indd 99 03/03/15 3:20 pm 100 c. koliba et al. / isomorphic properties of network governance that address certain collective action problems (ostrom, 1990, 2007; lubell, 2004), networks operating as advocacy coalitions (sabatier, leach, lubell, & pelkey, 2005; sabatier, weible, & ficker, 2005), networks designed to regulate behavior through the execution of hierarchical, command and control ties (fiorino, 2006; coglianese, 2001), watershed governance networks (imperial, 2005) and thick descriptions of environmental partnerships (koontz et al., 2004; imperial, 1999, 2005; leach & sabatier, 2005). across this literature, at least two kinds of environmental networks have been said to persist: those kinds of environmental networks that carry out regulatory functions through command and control, hierarchically aligned ties that are used to direct collective action, and those environmental networks that tend to rely on partnership, collaborative and horizontal ties to collaborate for collective action (fiorino, 2006). to date, little has been done to advance the field’s capacity to compare the properties of a regulatory network to a partnership network that are both charged with addressing the same environmental problem, covering the same geographic landscape, and encompassing similar jurisdictions during the same timeframe. in this study, the regulatory and partnership networks to be compared share a common policy function: the development of comprehensive plans that are designed to reduce the levels of phosphorus loading into lake champlain. furthermore, understanding how and why the structural and functional prosperities of different types of governance networks enable or temper experimentation and innovation has been noted as being of critical importance for practicing public administrators, researchers and theorists (rhodes, 2007). in this study, the capacity of regulatory and partnership networks to experiment and innovate is observed through the comparison of policy plans generated by each network. specifically, this study examines whether the regulatory and partnership networks responsible for the 2010 total maximum daily loading (tmdl) plan and 2010 opportunities for action (ofa) plan respectively replicate their internal governance structures into the planned networks that they create. this process of translation of process-based elements of organizational dynamics into static representations of organizational form and function is described here as “reification” (wenger,  1998). the theory of institutional isomorphism (dimaggio & powell, 1983) is drawn on to anticipate and explain the transmission of the governance structures of planning networks into the plans they reproduce. two hypotheses are proposed that examine the isomorphic properties of the planned network structures (h1) and the designation of policy tools in these two plans (h2). to study the properties of the networks produced in the plans textual data mining techniques (merkl, 2000) are used to render a comparative institutional network analysis of the two sets of networks. methods for identifying policy tools, the actors designated for initiating and carrying out specific policy tasks, and the task structures themselves are advanced. standard measures of network density, visual analysis of network structures and clusters, and statistical comparisons of the task structures found across the two planned networks are produced and used to provide basic descriptions of plan properties and to critically examine the two hypotheses. this article begins with an articulation of the two governance network isomorphism hypothesis used in this study and introduce the conceptual framework used to study them. article_14-12.indd 100 03/03/15 3:20 pm c. koliba et al. / isomorphic properties of network governance 101 the methods employed in this study are then discussed. the regulatory and partnership networks responsible for producing the two plans are presented, followed by a detailed account of each plan’s properties. the paper concludes with an analysis of the two hypotheses and some implications for further theory development and research. 1. governance network isomorphism hypotheses testing institutional isomorphism has been defined as, “a constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions” (dimaggio & powell, 1983, p. 149). the use of institutional isomorphism theory in public administration and policy studies literature has been generally relegated to the studies of the organization serving as the unit of analysis (ashworth, boyne, & delbridge, 2009; frumkin & galaskiewicz, 2004; villadsen, 2013). these applications of isomorphism theory to public sector organizations add contextual richness, but do not explain isomorphic tendencies across institutional structures of a higher, more complex order of scale: the multi-organizational network. this present study shifts the unit of analysis in isomorphism theory from the single institution to a multi-institutional, multi-sector network context. by expanding the unit of analysis to encompass actors spanning many levels of governmental, private and nonprofit actors, an opportunity to capture the relationship between policy tool selection and governance structure is provided. in another study, villadsen (2011) suggests that public governance networks are sources of isomorphic pressure unto themselves, and that network structures of particular form can either increase or decrease isomorphic tendencies. he pays particular attention to the role of elected officials as pollinators of network form and function. this study builds on this line of research. the central question to be asked in this study is: do environmental planning networks tend to reify themselves in the plans they create? drawing on distinctions found within the environmental governance and environmental partnership literature, this study distinguishes between regulatory and partnership networks and treats these distinctions as dependent variables. regulatory networks are predominantly shaped by hierarchical relationships between regulators and those that are regulated, be they industries, organizations and institutions, or individuals. fiorino describes how these hierarchical ties function: “regulatory agencies issue rules that are binding on defined classes of entities. agencies create a system of inspections and reporting to monitor compliance with their rules. entities that fail to comply are subject to penalties” (1999, p. 7). translating hierarchical ties into networks structures, we find regulatory networks dominated by hierarchical ties, appearing as mono-centric network structures, characterized by a singular network hub with the “regulator” at the center serving as the principal over its regulated agents. within the context of environmental regulation of water quality, a regulatory network is established on principal-agent ties through which federal agencies set benchmarks and call on state agencies to adopt suitable management plans to address environmental problems. in the case of study undertaken here, regulation is carried out article_14-12.indd 101 03/03/15 3:20 pm 102 c. koliba et al. / isomorphic properties of network governance by the usepa and the vermont agency for natural resources (vt anr) under policy objectives and directives stated within the nonpoint source pollution amendments to the clean water act in 1987. the usepa is then obligated to require the state to plan and implement actions (usually through the form of a total maximum daily load [tmdl]) to achieve the state standard for that water body.  the regulatory network in this study is responsible for developing, and ultimately implementing, federal level usepa directed tmdl standards for the vermont portion of the lake champlain basin. the regulatory plan to be studied here is the 2010 vermont lake champlain basin tmdl plan. in contrast to the more top down regulatory framework, environmental partnership networks are collaboratively governed (ansell & gash, 2007) and structured through mostly voluntary ties built on reputation, trust, and norms of reciprocity. in this study, we refer to these as partnership networks that often appear as polycentric, and often times, less dense, network structures (ostrom, 2010). in recent decades, watershed partnership approaches toward designing and implementing policy solutions have become more common (lubell & fulton, 2007; sabatier, leach, et al., 2005; sabatier, weible, et al., 2005). empirical comparisons of watershed partnerships have tended to focus on the composition of partnership characteristics and their inferred relationship to performance goals (koontz et al., 2004; imperial, 2005). partnership networks exist to carry out a broader range of policy functions: pooling resources, disseminating information, and otherwise aligning interests with public policy needs. in the case of the study undertaken here, these functions are carried out through the operations of the lake champlain basin program (lcbp, 2014) and its network of stakeholders. the lcbp serves as the network administrative organization (nao) (provan & kenis, 2008) or backbone organization of a multi jurisdictional watershed partnership that issues comprehensive watershed management plans for the entire basin every five years. the partnership plan to be studied here is the 2010 opportunity for action (ofa). it is hypothesized that the structural characteristics of the two different planning networks in this study can serve as a predictor of their planned network characteristics. more specifically, this paper asks to what extent the more centralized structure of the regulatory network replicates this structure in the plan that it creates? the “network structure isomorphism” hypothesis for this study is stated below as: h1: the planned network designed by the tmdl regulatory network will exhibit more centralized and denser network properties than those designed by the lcbp partnership network. it has become widely accepted that the selection and use of a policy tool can shape the structure of a governance network by giving form to certain types of relational ties, and by determining the types of tasks to be undertaken up by specific network actors (salamon, 2002; bressers & o’toole, 2005; howlett, 2005; koliba, meek, & zia, 2010). regulatory tools give structure to principal-agent ties where a regulatory authority has some measure of power over others. the regulatory policy tools identified in this study include: environmental and economic regulation, and permits. more collaboratively-oriented policy tools also persist. article_14-12.indd 102 03/03/15 3:20 pm c. koliba et al. / isomorphic properties of network governance 103 public information as a policy strategy ultimately relies on the diffusion of information through social networks (weiss, 2002) that rely on a willingness to consume and distribute information. a range of other policy tools has been identified across the literature (salamon, 2002; howlett, 2005; birkland, 2001) that tend to perpetuate some combination of principal-agent and collaborative ties. these “mixed tie” policy tools include: contracts, grants, loans and guarantees and tax incentives, all of which are coded for in this study. building on the relationship between policy tool structure and governance arrangements implied in this literature, a second hypothesis is posed that examines how regulatory authority is reified and passed on in the form of regulatory policy tools. it is expected that the regulatory network will recommend more regulatory policies (such as environmental and economic regulation, and permits) than the partnership network, which will tend to recommend policy tools that perpetuate the use of collaborative ties, (such as public information, as well as other “mixed ties” tools such as grants, contracts, and loans and guarantees). this “policy tool isomorphism” hypothesis is stated below as: h2: planned networks designed by the tmdl network will recommend a higher percentage of regulatory tools than the lcbp network. to carry out this comparative analysis, a governance network analysis framework (koliba et al., 2011) is used and is summarized in the table below (see table 1). a “governance network” is defined as a relatively stable pattern of coordinated action and resource exchanges involving policy actors crossing different social scales, which are drawn from the public, private, or non-profit sectors and across geographic levels. these actors interact through a variety of competitive, command and control, cooperative, and negotiated arrangements for purposes anchored in one or more facets of the policy stream (p. 53). key elements of governance network analysis framework to be explained here include: 1) the policy functions that the network takes (problem formation, policy design, policy implementation, policy evaluation) (bovaird, 2005); 2) the policy tools that shape the design process and appear in policy plans (salamon, 2002); 3) the sectoral and jurisdictional composition of the actors in the network; and 4) the task structures that are assigned to actors in these plans. table 1 (below) lays out these parameters along with highlighted notes. these governance network variables and parameters are used to describe the patterns of interaction of both plans, and examine the two isomorphism hypotheses introduced at the beginning of the article. the next section describes the methods used to undertake this analysis. 2. methods this study integrates institutional network analysis, text data mining, and qualitative case study research, along with elements of the governance network framework described above. comparative case study analysis methods were used to compare the two planning networks. a combination of stakeholder interviews, source document analysis, and article_14-12.indd 103 03/03/15 3:20 pm t ab le 1 g ov er na nc e n et w or k v ar ia bl es a nd p ar am et er s e m pl oy ed v ar ia bl e p ar am et er s ap pl ie d in t hi s st ud y n ot es p ol ic y f un ct io ns p ol ic y de si gn d ra w s on t he m aj or a ss um pt io n m ad e in t he n et w or k go ve rn an ce l it er at ur e th at m ul ti in st it ut io na l ne tw or ks e xi st a cr os s al l po li cy f un ct io ns : pr ob le m d efi ni ti on , p ol ic y de si gn , po li cy i m pl em en ta ti on a nd p ol ic y ev al ua ti on s ta ge s (b ov ai rd , 2 00 5; k o l ib a e t a l ., 20 10 ). d om in an t p at te rn of n et w or k in te ra ct io n r eg ul at or y p ar tn er sh ip m on oc en tr ic n et w or k p ol yc en tr ic n et w or k d ra w s on t he a rt ic ul at io n of n et w or ks a s a se ri es o f no de s an d ti es . r eg ul at or y ne tw or ks a re m os t of te n ch ar ac te ri ze d by t he ir m on oce nt ri c, m or e de ns el y cl us te re d st ru ct ur es . p ar tn er sh ip n et w or ks ar e m or e of te n ch ar ac te ri ze d by t he ir p ol yc en tr ic , l es s de ns el y cl us te re d st ru ct ur es . p ol ic y t oo ls e co no m ic r eg ul at io n e nv ir on m en ta l r eg ul at io n p er m it s p ub li c in fo rm at io n g ra nt s c on tr ac ts l oa n gu ar an te e t ax i nc en ti ve s p ri nc ip al -a ge nt t ie s c ol la bo ra ti ve t ie s m ix ed t ie s d ra w s on a c om m on ly e xc ep te d fr am ew or k fo r di ff er en ti at in g po lic y to ol s (s al am on , 2 00 2) a nd a rt ic ul at es t he r el at io ns hi p be tw ee n po li cy t oo l an d ne tw or k ti e. t yp e of n et w or k a ct or s in te rn at io na l g ov er ni ng b od y f ed er al g ov er nm en t s ta te /p ro vi nc e g ov er nm en t r eg io na l g ov er nm en t l oc al g ov er nm en t p ri va te e nt er pr is e n g o /n on -p ro fi t c it iz en r es ea rc he r d ra w s on c ur re nt a pp re ci at io n th at n et w or k go ve rn an ce u nf ol ds w it hi n ne tw or ks o f in st itu ti on al a ct or s th at s pa n th e pu bl ic , p ri va te a nd n on pr ofi t se ct or s, a s w el l as d if fe re nt s oc ia l sc al es ( in di vi du al t o or ga ni za ti on al ). t as k st ru ct ur e in it ia ti ng im pl em en ti ng a pp li es p ri nc ip al -a ge nt t he or y to t he l ev el o f ta sk s tr uc tu re . i ni ti at in g ac to rs a re p ri nc ip al s as cr ib ed r es po ns ib il it y fo r ca rr yi ng o ut a t as k to b e un de rt ak en b y an a ge nt r es po nd in g ac to r. 104 article_14-12.indd 104 03/03/15 3:20 pm c. koliba et al. / isomorphic properties of network governance 105 participant observation were triangulated to discern the network governance structures of the two planning networks. these qualitative data were coded for critical events that informed the design of each of the two plans under study. visualizations of these policy design networks were rendered from these qualitative assessments (see figures 1 and 2). figure 1. tmdl policy design network governance structure for the 2010 vermont tmdl figure 2. lake champlain basin program governance structure article_14-12.indd 105 03/03/15 3:20 pm 106 c. koliba et al. / isomorphic properties of network governance instititional network analysis techniques are used to describe and evaluate the networks of institutions found in the planning documents created by the tmdl and ofa planning networks. the application of instititional network analysis techniques to the study of governance networks has become widely adopted following the pioneering application of social network analysis to emergency management by comfort (2002, 2007) and kapucu (2006a, b). these applications have been used to study the jurisdictional and sectoral composition of institutional actors. in at least one case, institutional network analysis has been used to study the fidelity of implemented emergency management responses to those found in national emergency management plans (kapucu & demiroz, 2011). this study builds on comfort and kapucu’s approaches to textual data mining to code for actors, but we deepen this methodology by adding a network layer of “task structures” that appear in planning documents and contain a significant level of detail in them. a task structure is defined here as those bimodal relationships between two or more actors united around a common task. the combination of these bimodal relationships form patterns of network structure that may be described, compared, and contrasted. in highly structured planning documents like the ones studied here, policy recommendations may be accompanied by more detailed implementation strategies that identify who is to intiate a given task and who is to respond to the intiation to carry out the task. in strategies where actors are explicitly identified, a network structure is inferred to represent the planned relationships among these actors. *ora software (carley, 2001–2011) was used for matrix algebra and metric calculation. the network visuals were generated in ucinet (borgatti, everett, & freeman, 2002). measures of network density, nodal centrality, betweenness and eigenvector values are generated. 3. the governance of phosphorus loading in lake champlain the networks studied here share a common policy aim and geographic context: to protect and manage water quality in the lake champlain basin, located on the boarder of vermont, new york and quebec. both sets of networks rely on the same shared data regarding the scope, scale and causes of nutrient loading into lake champlain. both sets of networks were born from a shared concern for the impacts of this nutrient loading on water quality in the lake, particularly the significant algae blooms that comprise water quality for drinking, swimming and other forms of recreation. although land use contributors to phosphorus loading in the lake champlain basin vary across the region, vermont contributes the majority of phosphorus loads to lake champlain (lcbp, 2008). this variation in loading responsibility between land use types, as well as the cross-jurisdictional complexity of two state (vt and ny) regions and one province (quebec) region, prevents the use of blanket regulatory approaches, making the situation ripe for collaborative watershed management. in the sections to follow, a description of the regulatory and partnership networks that have been established to mitigate the flow of phosphorus pollution into the lake is provided. article_14-12.indd 106 03/03/15 3:20 pm c. koliba et al. / isomorphic properties of network governance 107 3.1. 2010 tmdl regulatory planning network the regulatory authority of the usepa dominates the regulatory network (see figure  1). in section 303(d) of the clean water act, the usepa requires all states to identify waters that are “impaired,” that is, which do not meet the state water quality standards. once identified, states must analyze and set total maximum daily load (tmdl) targets for each pollutant to the water body. tmdls must include: (1) an estimate of pollutant loads from all significant point and non-point sources; (2) a link between pollutant sources and their impacts on water quality; (3) allocation of allowable pollutant loads among sources; (4) actions for achieving water quality standards; and (5) opportunities for public participation (usepa, 1991). in this way, the usepa establishes limits for pollutant levels and places a burden on the state to reduce those levels. vermont was issued a tmdl directed at nonpoint source pollution into lake champlain in 2002. following a lawsuit filed by the conservation law foundation in 2008, the 2002 tmdl was pressured to be withdrawn and replaced with a more robust plan. the 2002 tmdl was officially rejected by the usepa in 2011. the 2010 tmdl plan was developed under the assumption that the 2002 tmdl may be rejected by the usepa. catalyzed into action as a result of the lawsuit, the vermont state legislature directed the vt anr to work with the vermont agency of agriculture, food, and markets (aafm) and other state agencies to design this implementation plan as a way to stave off a rejection of the 2002 tmdl. figure 1 provides a visualization for the multi-institutional context for the creation of the 2010 tmdl plan, with the anr serving as a lead organization in a collaborative planning process that included extensive consultation with the public at large, special interest groups and other state agencies. this network catalyzed as a result of mandate derived directly through the vt state legislature and indirectly through the regulatory authority of the usepa. these regulatory pressures held the lead organization, the anr, as chiefly responsible for devising a suitable plan. the extent to which these coercive pressures were translated in the plan chiefly devised by the anr is the subject of this particular study. 3.2. 2010 ofa partnership planning network the lcbp is designed along the “management conference” model of watershed management (koontz et al., 2004), in which a network administrative organization is created to generate and manage knowledge about watershed health for adaptive management purposes. an act of congress led to the passage of the lake champlain special designation act of 1990 (public law 101–596, 1990), sponsored by us senators from both new york and vermont, created the lcbp.1 the governments of vermont, new york and quebec entered into a partnership facilitated through the lcbp steering committee (lcsc) and citizens’ advisory committees to craft a science-based action plan entitled “opportunities for action” (ofa) (lcbp, 2010). the lcbp does not have regulatory 1 the lcbp is also known as the lake champlain management conference (lcmc). article_14-12.indd 107 03/03/15 3:20 pm 108 c. koliba et al. / isomorphic properties of network governance authority, but rather sponsors research, manages grant programs, provides educational outreach, convenes public meetings, and compiles watershed data and maps into a comprehensive atlas (lcbp, 2014). by design, the lcbp is an organization deeply involved in the coordination of a partnership network and serves as a classic network administrative organization (provan & kenis, 2008). the lcbp is administered jointly by several agencies: usepa regions i and ii; new york state department of environmental conservation (nys dec); vt anr; quebec ministry of environment; and new england interstate water pollution control commission (neiwpcc). administration is carried out by a professional staff, some of whom have shared reporting roles with state, provincial, and federal agencies. the lcsc guides the program’s policy and planning activities and serves an executive role in a network of committees that includes a cac from each jurisdiction (vermont, new york and quebec), as well as a technical advisory committee (tac), an education and outreach committee, and a cultural heritage and recreation advisory committee. this network of committees, shown in figure 2 below, creates a management structure that is not centrally administered, but rather built on collaboration between many actors that participate in both decision-making and implementation. every seven years, a new ofa report is issued by the lcbp, which offers policy recommendations designed to improve water quality in the basin. topical foci in the ofa extend beyond nutrient loading to include control of invasive species, management of fisheries, and so on. for the purposes of this analysis, only the portions of the ofa addressing nutrient management in the lake are included in the analysis. drafts of plans are vetted in open committee meetings and public hearings along with other citizen engagement activities. the ofa plan is informed by a broad range of actors who have repeated opportunities to inform its final version. it is important to note that the usepa mandates the institution of approved tmdl plans for all impaired watersheds. usepa is responsible for assuring some form of regulatory compliance over state and local jurisdictions. while the lcbp watershed partnership ofa plans have no legal authority per se, and carry the weight of recommendations and research funding authority. these are two distinctions that we are to return to later. 4. comparing the 2010 tmdl plan and the 2010 ofa plan the comparative analysis of the two planned networks begins with an analysis of the suite of policy tools assigned to mitigate different land use behaviors contributing to phosphorus runoff. a description of the network actors by sector and jurisdiction found in each plan is then rendered. task structures in both plans are then described as patterns of initiating and implementing actor assignments. the section concludes with a comparative analysis of both planned networks, comparing both node-level and whole network properties. article_14-12.indd 108 03/03/15 3:20 pm c. koliba et al. / isomorphic properties of network governance 109 4.1. policy tool ascription by land use phosphorus loading levels into lake champlain are strongly tied to the composition of agricultural, development (residential and commercial), and forestry land uses in the region (troy et al., 2007). table 2 (below) demonstrates how the tmdl plan identifies 249 tasks, roughly 87% of which indicate a policy tool to be used. the ofa plan identifies 192 tasks, 99% of which indicate a policy tool. while this represents a greater detail in planning in the ofa than the tmdl plan, the ofa applies a smaller range of tools. the ofa identifies five different policy tools, while the tmdl plan identifies eight. public information is the most dominant tool in both plans, accounting for more than half of the identified tools in each plan, with environmental regulation and grants recommendations the next most prevalent policy tools in both plans. both plans apply permits and contracts, but only to a lesser extent. three tools appear only in the tmdl plan: economic regulation, loans and guarantees, and tax incentives. comparison of these results suggests that the mix of policy tools is similar in both plans. in reviewing the dominant policy tool selection, public information is the most used tool in all the land-use domains. the second most identified policy tool is environmental regulation, mostly used to address agriculture and development land use issues. 4.2. actor characteristics the composition of actors within both plans is shown in table 3 and 4, differentiated by actor jurisdiction and sector. the patterns found in table 3 (below) should come as no surprise, as they reflect the differences in geographic scope of the regulatory (vermont) and partnership (vermont, new york and quebec) networks. table 2 strategic profile by use of policy tools ofa (192 tasks) tmdl (249 tasks) count percent count percent economic regulation 0 0.0 1 0.4 environmental regulation 42 21.9 48 19.3 permits 9 4.7 14 5.6 public information 100 52.1 135 54.2 contracts 6 3.1 6 2.4 grants 35 18.2 36 14.5 loan guarantees 0 0.0 7 2.8 tax incentives 0 0.0 2 0.8 policy tools utilized 192 100.0 249 100.0 article_14-12.indd 109 03/03/15 3:20 pm 110 c. koliba et al. / isomorphic properties of network governance table 3 frequency of jurisdiction attribute values category tmdl ofa count percentage count percentage vermont 12 25.53 15 21.13% new york 0 0.00% 9 12.68% quebec 0 0.00% 4 5.63% u.s. 8 17.02% 12 16.90% canada 0 0.00% 0 0.00% international 1 2.13% 3 4.23% multiple or indeterminate 26 55.32% 28 39.44% total 47 100% 71 100% table 4 frequency of sector attribute values category tmdl ofa count percentage count percentage federal government 8 17.02% 9 12.68% state/province government 9 19.15% 10 14.08% regional government 4 8.51% 1 1.41% local government 6 12.77% 21 29.58% private enterprise 7 14.89% 11 15.49% ngo/non-profit 5 10.64% 10 14.08% citizen 6 12.77% 5 7.04% researcher 2 4.26% 3 4.23% international governing body 0 0.00% 1 1.41% total 47 100% 71 100% a review of the distribution of actors by sector (table 4 below) reveals some differences in plan composition. overall, government actors dominate both plans, representing 56.25% of actors in the tmdl and 55.55% of actors in the ofa. however, while the tmdl plan demonstrates a preponderance of state-level agencies, the ofa shows a stronger emphasis on local government. reliance on the remaining sectors is remarkably similar in both plans. 4.3. network properties and structural patterns the networks extracted from these two reified plans demonstrate structures that emerge from formal planning processes. measures of node relationships and network connectivity help to illuminate patterns seen in figures 3 and 4. a visual review of these article_14-12.indd 110 03/03/15 3:20 pm c. koliba et al. / isomorphic properties of network governance 111 figure 3. designed network for a phosphorus reduction network found in the tmdl plan figure 4. designed network for a phosphorus reduction network found in the ofa plan two figures shows the tmdl planned network is distributed as a star network organized around the vt anr and municipalities (a class node), along with several isolate actors that have no ties. the ofa planned network exhibits a more distributed structure with a series of bridging organizations that link jurisdictional clusters of actors, notably the ridge of federal agencies connecting the states of new york and vermont. article_14-12.indd 111 03/03/15 3:20 pm 112 c. koliba et al. / isomorphic properties of network governance centralization measures the extent to which a network approaches a perfect star structure. table 5 shows greater degree of centralization in the tmdl network than in the ofa network. this is shown visually in figures 3 and 4 in which there are several hub nodes in the ofa, while in the tmdl network there is but a single central node. greater fragmentation in the tmdl is witnessed as a subset of highly connected actors that are centralized into one component while several others remain isolated. this is quite different from the ofa, which incorporates all of the actors into the connected component through a set of regional hubs. these characteristics are quantified in the comparison of isolate count, fragmentation scores, and levels of degree, betweenness, and eigenvector centralization (see table 5). these patterns all point to a more centralized, but also more fragmented, tmdl network. a review of the top nodal actors by centrality measures (table 6) of the 2010 tmdl plan shows that the vt anr possesses the greatest number of ties (degree centrality) and has the most relationships to other highly-connected nodes (eigenvector centrality). according to this design, anr will broker with many other actors directly to carry out plan functions. in contrast, the ofa relies on several nodes for brokering (vermont, new york, quebec, lcsg, lcbp, usepa) (see table 7). table 5 network characteristics tmdl ofa node count 47 71 component count 22 1 isolate count 20 0 fragmentation 0.722 0.00 degree centralization 0.084 0.244 eigenvector centralization 0.548 0.593 table 6 ten most central organizations in the tmdl network degree centrality eigenvector centrality vt anr 0.097 vtanr 0.646 municipalities 0.097 municipalities 0.610 vlct 0.043 homeowners 0.448 homeowners 0.043 rpcs 0.408 developers 0.043 vlct 0.408 rpcs 0.043 developers 0.386 swcds 0.043 swcds 0.346 town road crews 0.043 landowners 0.246 federal agencies 0.032 federal agencies 0.213 vtrans 0.022 state agencies 0.213 article_14-12.indd 112 03/03/15 3:20 pm c. koliba et al. / isomorphic properties of network governance 113 several actors are present in both tables, indicating their centrality to both plans. an example is the “municipalities” class node that is primarily identified as an implementing actor. its centrality indicates that the node is targeted by a range of interventions in both plans. a similar node, “farmers,” is highly central in the ofa, but does not appear as a top node in the tmdl plan. this reflects the tmdl plan’s reliance on aafm as a bureaucratic channel to address agriculture, while the ofa utilized a more varied approach. 5. network and policy tool isomorphism this analysis suggests some mixed results regarding the two network isomorphism hypotheses. in the 2010 tmdl planned network (figure 5), the largest assemblage of the network is a highly dense cluster of actors with relatively high number of connections, while the remainder of the graph is highly fragmented, including many isolates. as noted, the design of the 2010 tmdl plan was led primarily by the vt anr. the evidence presented here suggests that the same monocentric structure that shaped the planning network appears in the planned network, with anr as the central locus of initiating activity. while a visual inspection of the 2010 ofa (figure 6) reveals a spine of federal agencies linked to two main hubs—vermont and new york—with four smaller hubs: quebec, the lcbp, the usepa, and lake champlain sea grant (lcsg). given the visual evidence and the comparative centrality scores found in table 6, it appears that there is sufficient reason to affirm the h1 presented earlier: the regulatory network embodied in the regulatory network produced a more centralized planned network structure. interestingly, in follow up conversations with the lead author of the 2010 tmdl plan, there was a concerted effort to decenter anr as the center piece of the plan. although anr serves as the most central hub in the planned network, its centrality is tempered somewhat by the intentional efforts of the lead author (the then head of the clean and clear program) to collectivize responsibility for phosphorus reduction. table 7 ten most central organizations in the ofa network degree centrality eigenvector centrality vermont 0.257 vermont 0.694 new york 0.243 new york 0.683 quebec 0.064 farmers 0.348 lcbp 0.064 usda-nrcs 0.289 lcsg 0.057 usfws 0.264 usepa 0.050 lcbp 0.260 farmers 0.050 usepa 0.238 usda-nrcs 0.043 municipalities 0.227 usfws 0.043 quebec 0.225 municipalities 0.029 wrps 0.218 article_14-12.indd 113 03/03/15 3:20 pm 114 c. koliba et al. / isomorphic properties of network governance looking at the distribution of policy tools found in each plan (see table 2), a more complex picture of policy tool distribution is painted. the more coercive regulatory policy tools that were coded for (economic regulation, environmental regulation, and permitting) accounted for 21.91% of the policy tools identified in the 2010 ofa and just 19.3% of the policy tools identified in the 2010 tmdl plan. this distribution suggests that h2 is not proven and that, in fact, the overall distribution of policy tools within each of the two plans is remarkably similar, suggesting here that a predisposition toward some policy tools over others prevails across both networks. we provide some possible explanations for this below. dimaggio and powell (1983) distinguish between three types of institutional isomorphism. coercive isomorphism is associated with pressures bought on by regulation, legal sanction, or mandate, whereby organizations adopt certain structures and functions to accommodate regulatory, legal, or bureaucratic compliance. mimetic isomorphism is said to arise under uncertainty when risk may be reduced by copying pre-existing, presumably successful, organizational forms. normative isomorphism is driven by professional standards and expectations regarding best practices and high performance. all three drivers of isomorphic behavior may be in play here to explain both observed phenomena: evidence of network structure isomorphism at the level of perpetuation of regulatory and partnership network structure; and, the lack of evidence to assert that policy tool isomorphism persists across partnership or regulatory network forms. the origins of the 2010 tmdl planning networks originated through legislative mandate, a source of coercive isomorphic pressure. existing federal level tmdl legislation (appearing in the 1987 amendment to the clean water act) outlines a centralized design model in which states are required to draft management plans to be approved by the usepa. designees of each state are then presumed to assume a central role. greater coercive pressure is at play in the tmdl network. the clean and clear program of the vt anr at the time was mandated by the vermont state legislature to develop an implementation plan for the original 2002 tmdl. usepa pressures to achieve benchmarks for phosphorous loading may have encouraged the use of a lead organization network structure, where a clear line of accountably can be drawn to anr. coercive isomorphism appears in planned networks when and where formal or informal expectations are laid out by principals and provided to agents to carry out plans. however, the expectations of the principals overseeing the planning networks’ activities (usepa, vermont state legislature, and the principal parties to the governance of the lcbp) over the selection of policy tools used to address the problem of phosphorus loading into lake champlain appear to be very similar. it may suggest that other network ties are at work across the region that have been not accounted for in the current study. it may also arise out of a common set of interest groups and network actors appearing in both the regulatory and partnership networks. for example, municipalities, which are the main locus of land use zoning in vermont and new york, play a prominent feature in both plans. local control over land use decision making may be a tacit assumption driving the homogeneity of policy tool choice. deeper examination of these factors is called for. article_14-12.indd 114 03/03/15 3:20 pm c. koliba et al. / isomorphic properties of network governance 115 it is likely that some form of mimicry is at work to explain the isomorphism present. mimicry occurs when best practices are proffered by planners, policy makers and thought leaders within a given field. these best practices strategies may be drawn from other watershed regions, practices thought to be most effective, and/or practices that have been undertaken in the past. it should also be noted that nonpoint source pollution, whether it is focused on mitigating agricultural runoff or storm water runoff from development, is said to be best addressed through a professionally determined set of best management practices (bmps). these bmps are promulgated through professional networks that have evolved among environmental engineers, agronomists, and watershed management professionals. these same professionals staff the committees and programs charged with drafting these plans. the dissemination of bmps may be described as a form of mimicry. thus, mimicry appears to be in place in the standard accepted practices established for watershed management and governance, and may be a major factor in explaining the homogeneity in the selection of policy tools in both plans. in the context of plan development phases in both networks, best management practices for stormwater and agriculture run-off were widely shared and examined among water quality experts drawn from the public, private, and nonprofit sectors. normative isomorphism may be found in the translation of common norms, values, and beliefs between key stakeholders. the implementation plans that were created and vetted through both networks were subject to substantial dialogue and debate. both planning processes preferred to rely on a consensus of a core group of stakeholders whose perspectives were informed through lengthy public input, peer review, and negotiation. it has been noted how recent environmental governance and management approaches have tended to avoid the utilization of command and control regulatory approaches (gerlak, 2006) and instead rely on more reflexive strategies (fiorino, 1999). these reflexive strategies surface in the stronger deference to the use of incentives and market-based mechanisms in both plans. several key actors in the lake champlain basin were participants in both planning networks. in follow up discussions with the lead author of the 2010 tmdl plan sheds particular light on this subject. when asked about the apparent similarities between the 2010 tmdl plan and the 2010 ofa she remarked that the overlaps in the plans were likely the result of having some of the same lead actors in designing the plans participating in the same planning meetings. in other words, there was an intentional effort to align the plans together. this observation demonstrates the role of that individual network managers may take, when positioned right, in steering the overall composition of an implementation plan. 6. conclusions this study presents the first time that a regulatory network and a partnership planning network focusing on the same environmental problem—operating within similar jurisdictional regions, and functioning at the same point in time—have been studied. institutional article_14-12.indd 115 03/03/15 3:20 pm 116 c. koliba et al. / isomorphic properties of network governance network analysis was used to compare the two plans resulting from these two planning networks. the results of this study suggest that institutional isomorphism theory can be applied to the study of multi-institutional networks and that it may be useful to explain how networks evolve from one stage of development to another. in order to deepen our understanding of network isomorphism it will be essential to render a more robust comparative analysis of a network’s evolution over time using consistent institutional network analysis methods for comparative purposes. in the present study, the informational asymmetry between the qualitative description of the planning networks and a statistical description of the planned networks is acknowledged. the methods of institutional network analysis advanced here allow for the mapping of networks reified in formal plans in a manner that accounts for initiating and implementing actors. our methods advance the field’s capacity to measure plan characteristics, including ways that document sectors and jurisdictions, policy tools for achieving outcome goals, and task structures. the task structure framework found in figures 1 can be applied to other types of plans, reports, and detailed descriptions of implementation tasks that offer significant details about actors and policy tools. when used as a decision support, these measures can clarify the structural needs of proposed actions and indicate plan weaknesses prior to implementation. in this way it can help align policy structures with contextual constraints and provide insight into the role that plans play in policy implementation. 7. acknowledgements this work was supported by the national science foundation, epscor program, grant eps 1101317, and the james m. jeffords policy research center at the university of vermont. the co-authors of this paper wish to thank julie moore, eric howe and bill howland, and to paper reviewers for their thoughtful comments on earlier iterations of this paper. references ansell, c., & gash, a. (2007). collaborative governance in theory and practice. journal of public administration and theory, 18(4), 543–571. ashworth, r., boyne, g., & delbridge, r. (2009). escape from the iron cage? organizational change and isomorphic pressures in the public sector. journal of public administration research and theory, 19(1), 165–187. birkland, t. a. (2001). an introduction to the policy process: theories, concepts, and models of public policy making. new york, ny: m. e. sharpe. borgatti, s. p., everett, m. g., & freeman, l. c. (2002). ucinet for windows: software for social network analysis. harvard, ma: analytic technologies. bovaird, t. (2005). public governance: balancing stakeholder power in a network society. international review of administrative sciences, 71(2), 217–228. bressers, h. t. a., & o’toole, l. j. j. (2005). instrument selection and implementation in a networked context. in p. eliadis, m. m. hill, & m. howlett (eds.), designing governance: from instruments to governance. montreal, quebec, canada: mcgill-queen’s university press. article_14-12.indd 116 03/03/15 3:20 pm c. koliba et al. / isomorphic properties of network governance 117 carley, k. m. (2011). *ora [computer software]. pittsburgh, pa: center for computational analysis of social and organizational systems, carnegie mellon university. coglianese, c. (2001). is consensus an appropriate basis for regulatory policy? in e. orts & k. deketalaere (eds.), environmental contracts: comparative approaches to regulatory innovation in the united states and europe (pp. 93–113). the hague, the netherlands: kluwer law international. comfort, l. k. (2002). managing intergovernmental responses to terrorism and other events. publius, 32(4), 29–51. comfort, l. k. (2007). crisis management in hindsight: cognition, communication, coordination, and control [special issue]. public administration review, 67, 189–197. dimaggio, p. j., & powell, w. w. (1983). the iron cage revisited: institutional isomorphism and collective rationality in organizational fields. american sociological review, 48(2), 147–160. durant, r. f., fiorino, d. j., & o’leary, r. (2004). environmental governance reconsidered: challenges, choices, and opportunities. cambridge, ma: mit press. fiorino, d. j. (1999). rethinking environmental regulation: perspectives on law and governance. harvard environmental law review, 23(2), 441–469. fiorino, d. j. (2006). the new environmental regulation. cambridge, ma: mit press. frumkin, p., & galaskiewicz, j. (2004). institutional isomorphism and public sector organizations. journal of public administration research and theory, 14(3), 283–307. gerlak, a. k. (2006). federalism and u. s. water policy: lessons for the twenty-first century. publius, 36(2), 231–257. great lakes critical programs act of 1990, public law 101–596. (1990). retrieved from http://thomas.loc. gov/cgi-bin/bdquery/l?d101:./list/bd/d101pl.lst:596%28public_laws%29 howlett, m. (2005). what is a policy instrument? tools, mixes, and implementation-styles. in p. eliadis, m. m. hill, & m. howlett (eds.), designing government: from instruments to governance (pp. 31–50). montreal, quebec, canada: mcgill-queen’s university press. imperial, m. t. (1999). institutional analysis and ecosystem-based management: the institutional analysis and development framework. environmental management, 24(2), 449–465. imperial, m. t. (2005). using collaboration as a governance strategy: lessons from six watershed management programs. administration & society, 37(3), 281–320. kapucu, n. (2006a). interagency communication networks during emergencies: boundary spanners in multiagency coordination. american review of public administration, 36(2), 207–225. kapucu, n. (2006b). public-nonprofit partnerships for collective action in dynamic contexts of emergencies. public management, 84(1), 205–220. kapucu, n., & demiroz, f. (2011). measuring performance for collaborative public management using network analysis methods and tools. public performance & management review, 34(4), 549–579. koliba, c., meek, j. w., & zia, a. (2010). governance networks in public administration and public policy. boca raton, fl: crc press. koontz, t. m., steelman, t. a., carmin, j., korfmarcher, k. s., moseley, c., & thomas, c. w. (2004). collaborative environmental management: what roles for government? washington, dc: resources for the future. lake champlain basin program. (1996). opportunities for action: an evolving plan for the future of the lake champlain basin: pollution prevention, control & restoration plan–october. retrieved from http://plan .lcbp.org/ofa-database/ lake champlain steering committee. (2010). opportunities for action: an evolving plan for the future of the lake champlain basin. retrieved from http://www.lcbp.org/pdfs/opportunitiesforaction2010.pdf lake champlain basin program. (2014). the lake champlain basin atlas. retrieved from http://www.lcbp.org leach, w. d., & sabatier, p. a. (2005). are trust and social capital the keys to success? watershed partnerships in california and washington. in p. a. sabatier, w. focht, m. lubell, z. trachtenberg, a. vedlitz, & m. matlock (eds.), swimming upstream: collaborative approaches to watershed management (pp. 233–258). cambridge, ma: mit press. linder, s. h., & rosenau, p. v. (2000). mapping the terrain of the public-private policy partnership. in p. v. rosenau (ed.), public-private policy partnerships (pp. 1–18). cambridge, ma: mit press. article_14-12.indd 117 03/03/15 3:20 pm 118 c. koliba et al. / isomorphic properties of network governance lubell, m. (2004). collaborative environmental institutions: all talk and no action? journal of policy analysis and management, 23(3), 549–573. lubell, m., & fulton, a. (2007). local policy networks and agricultural watershed management. journal of public administration research and theory, 18(4), 673–696. merkl, d. (2000). text data mining. in r. dale, h. moisl, & h. somers (eds.), handbook of natural language processing (pp. 923–938). new york, ny: mercel dekker. ostrom, e. (1990). governing the commons: the evolution of institutions for collective action. cambridge, ma: cambridge university press. ostrom, e. (2007). institutional rational choice: an assessment of the institutional analysis and development framework. in p.a. sabatier (ed.), theories of the policy process (2nd ed., pp. 21–64). boulder, co: westview press. ostrom, e. (2010). beyond markets and states: polycentric governance of complex economic systems. the american economic review, 100(3), 641–672. provan, k. g., & kenis, p. (2008). modes of network governance: structure, management and effectiveness. journal of public administration research and theory, 18(2), 229–252. rhodes, r. (2007). understanding governance: ten years on. organization studies, 28(8), 1243–1264. sabatier, p. a., leach, w. d., lubell, m., & pelkey, n. w. (2005). theoretical frameworks for explaining partnership success. in p. a. sabatier, w. focht, m. lubell, z. trachtenberg, a. vedlitz, & m. matlock (eds.), swimming upstream: collaborative approaches to watershed management (pp. 173–200). cambridge, ma: mit press. sabatier, p. a., weible, c., & ficker, j. (2005). eras of water management in the united states: implications for collaborative watershed approaches. in p. a. sabatier, w. focht, m. lubell, z. trachtenberg, a. vedlitz, & m. matlock (eds.), swimming upstream: collaborative approaches to watershed management (pp. 23–52). cambridge, ma: mit press. salamon, l. m. (2002). the new governance and the tools of public action: an introduction. in l. m. salamon (ed.), the tools of government: a guide to new governance (pp. 1–47). oxford, england: oxford university press. troy, a., wang, d., capen, d., o’neil-dunne, j., & macfaden, s. (2007). updating the lake champlain basin land use data to improve prediction of phosphorus loading (report to the lake champlain basin program). grand isle, vt: lake champlain basin program. retrieved from http://www.uvm.edu/giee/ pubpdfs/troy_2007_lake_champlain_basin_program.pdf united states environmental protection agency. (1991). guidance for water quality-based decisions: the tmdl process (epa 440/4-91-001). washington, dc: assessment and watershed protection agency. vermont agency for natural resources. (2010). revised implementation plan: lake champlain phosphorus tmdl. retrieved from http://www.watershedmanagement.vt.gov/erp/docs/erp_revisedtmdl.pdf villadsen, a. r. (2011). structural embeddedness of political top executives as explanation of policy isomorphism. journal of public administration research and theory, 21(4), 573–599. villadsen, a. r. (2013). similarity or difference? the relation between structure and strategy isomorphism in public organizations [special issue]. british journal of management, 24, s62–s75. weiss, j. a. (2002). public information. in l. a. salamon (ed.), the tools of government: a guide to the new governance (pp. 217–254). oxford, england: oxford university press. wenger, e. (1998). communities of practice: learning, meaning, and identity. cambridge: cambridge university press. article_14-12.indd 118 03/03/15 3:20 pm strategic policymaking within complex timescapes 15 university of bamberg press strategic policymaking within complex timescapes author: graham room department of social and policy sciences, university of bath, united kingdom email: hssgjr@bath.ac.uk the complex interactions of social life serve to structure in time the opportunities, threats, and constraints that confront different social actors. this article examines policymaking within such complex “timescapes.” it does so by reference to the policy sciences literature and the complexity literature. the former is rich in its qualitative treatment of policymaking and time but has been slow to exploit the analytical tools of the complexity literature. the latter has been slow to appreciate the complexity of timescapes and to recognise the importance of power and the struggle for positional advantage. this article develops a synthesis of the two, combining their analytical power. it shows how this can illuminate the policy world, both conceptually and practically. it draws on hirschman’s treatment of the “interlocking vicious circles” that hold development back and the “upward spirals” that can—when the time is right—be mobilised. it sets this within an appreciation of political economy and institutionalism. it concludes by considering the practical tools available for policy makers to navigate complex timescapes—and how the social scientist can subject those policy decisions to stringent scrutiny. keywords: complexity; landscapes; timescapes; policymaking introduction it is common to assume, when speaking of time, that it is enough to distinguish the past, present and future, the short-term and the long-term. we also speak as though the rhythms of time— and the calendar of social and political life—are the same for all actors. in general, they are not. it is not just that the nomenclatures of time—the day of the week, the hour of the day—are socially constructed, as with the many other classificatory activities of human societies (adam, 1990). it is more that the complex interactions of social life serve to structure in time the opportunities and constraints that confront social actors and the threats to which they are exposed1. exogenous shocks also intrude, exposing the resilience and vulnerability of different actors. spatial and temporal patterns are thus closely interrelated what prigogine describes as the “timing of space” (prigogine and stengers, 1984: 17). it follows that the language of fitness landscapes, employed by writers on complex systems, could be complemented by the language of complex “timescapes” (adam, 1998)2. this article examines policymaking within such complex “timescapes.” it does so by reference to the policy sciences literature and the complexity literature. the former is rich in its qualitative treatment of policymaking and time but has been slow to exploit the analytical tools of the complexity literature. the latter has been slow to appreciate the complexity of timescapes and to recognise the importance of power and the struggle for positional advantage—central tenets of the policy sciences. the first half of this article develops a synthesis of the two, combining their analytical power. the second half shows how this can illuminate the policy world, both conceptually and practically3. 1 this re-works marx’s affirmation (inspired by hegel) that men make their own history, not under circumstances of their own choice, but under circumstances directly encountered, given and transmitted from the past. 2 there is an argument for using the uglier but otherwise preferable neologism of complex “time-landscapes.” my argument is after all that landscapes and timescapes are socially co-produced. given however that the focus here is on making the case for the temporal dimension of such co-production, it seemed appropriate to use the simpler language of “complex timescapes.” 3 this argument was first developed in the author’s 2011 book, complexity, institutions and public policy (starting with section 1.4). chapter 1 concluded: “the first concern of the book is conceptual and ontological, integrating complexity science and institutionalism. i argue that each is in need of the other, as far as social science is concerned. complexity science is bereft of an adequate treatment of institutions; institutionalism needs the formal dynamic modelling of complexity science …. i argue that there are remarkable—but so far i think largely unremarked—convergences between the two, which make complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 mailto:hssgjr@bath.ac.uk 16 university of bamberg press before proceeding to that scholarly exercise however, it may be helpful to consider in more everyday terms in what sense “timescapes” are indeed woven into the interactions of social life. notice first that all moments are not the same. some are critical, in giving access to opportunities or avoiding danger, if only we can act. “there is a tide in the affairs of men, that taken at the flood leads on to fortune … and we must take the current when it serves, or lose our ventures”4. but most of the time we just get on with the demands of our everyday lives, while watching and waiting for those brief moments, when decisive action is called for5. it may also be that, in lenin’s words, “there are decades where nothing happens; and there are weeks where decades happen.” but these decades and weeks are not the same for everyone; they depend upon the specific arena in which a person finds themself. even within a particular arena, you may be dealing with no more than the routines of everyday life, while i face a critical moment. nevertheless, how i then decide to act may, in due course, impinge in major ways on your quiet and untroubled existence, confronting you, in turn, with decisive choices. but “in due course” is the crucial qualification—it takes time for such effects to work their way through the system and become manifest. that timescale may be very long-term, as for example with our actions on climate change, and the responsibility we exercise towards those yet unborn. we play out our lives within a variety of local arenas—family, workplace, local community— each with its own strategic dilemmas and calendar. we all face moments when it is our turn to choose which card to play, but some have more turns than others and hold more trump cards. an incumbent prime minister may have some freedom of manoeuvre over when to call an election; a committee chair some freedom in deciding what items to include on the agenda and which to postpone6. the rest of us have to adjust to what these better-placed actors do. nevertheless, even these adjustments – when played out across a myriad of minor actors – can in turn change the landscape that major actors confront. we try to multi-task, hoping that these various rhythms do not produce a tangled log-jam, all requiring our focused attention and strategic imagination at the same time. double-booking in our diary may cause social embarrassment (the very rationale for having a diary is to decompose our busy lives into manageable, bite-sized, and non-overlapping chunks); much worse is to find that the rhythms of one arena and the attention these demand crowd out all other concerns, however pressing. one obvious example has been the calendar of brexit negotiations in the uk, sucking the air out of the political system and incapacitating it from addressing other pressing needs; likewise with the coronavirus pandemic of 2020. i have described such complex timescapes by reference to the dilemmas and dynamics of social action for those involved. history is here not just a calendar of what happened when and in what order (in toynbee’s words, just “one damn thing after another”); it is the sequence of acts and scenes, within the interlocking dramas that social actors co-produce. one person’s critical moment can coincide with another person’s routine existence. nevertheless, moments and spaces of rapid reconfiguration tend to bring many out of their everyday placidity, whether driven by hope or by fear of change. here the struggle for positional advantage intensifies; new alliances emerge. within the larger drama, a myriad of sub-plots unfold, which could then become the seed of altogether different dynamics. in short: it may sometimes be sufficient—practically and analytically—to treat timescapes as though they were simple, and do no more than distinguish past, present, and future, the short-term and the long-term, with an appropriate time discount rate. but often not. plausible the project of integrating them.” my 2016 book, agile actors on complex terrains, took this integration project further, with particular reference to power and the struggle for positional advantage. the present article builds on those antecedents, to address the question of policy timescapes. 4 shakespeare, julius caesar, act 4 5 watching and waiting as a dynamic system develops, and intervening at key moments, is something all of us have watched in the children’s playground—as a parent pushes a child on a swing, timing each push to ensure that it reinforces the swing, or dampening it down, when it is time for the ride to end. 6 recall lukes’ second dimension of power (lukes, 2005). complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 17 university of bamberg press policy timescapes: theoretical scope and focus the foregoing applies to all of us, to a greater or lesser degree. it applies with particular force to actors who shape and implement public policies. “policy” is a somewhat fuzzy notion—and one used in a diversity of ways. it is in part about the choice of goals, the means for achieving them, the obstacles to implementation. it is also about organisational contexts, capabilities, and constraints, and about the power and interests of different stakeholders. much conventional policy analysis relies on the assessment of individual policies, viewed by reference to their “impact” on a target population. these assessments assume that the rest of the landscape remains more or less fixed. policy forecasts often extrapolate from such assessments, projecting them into the future. these simple models and forecasts may offer some limited insights for the placid backwaters of our societies; but not where we need them most, in the arenas of dramatic and convulsive change, where the boundaries of policies are actively contested. the language that policy makers themselves use betrays their awareness of the complex timescapes they face. during the covid pandemic, national leaders and their public health advisers have had to make sense of the rise and fall in the rates of infection, the critical junctures when lockdowns should be imposed or relaxed, and the consequences of these restrictions for the functioning of the health service, the economy and public morale. what indicators to use of such levels of infection; what evidence that different sorts of lock-down will sufficiently reduce social mingling; what signs that morale is breaking down or that public indiscipline is spreading? will the pandemic be over soon, or are we in for a long haul? the language of climate change likewise betrays anguished acknowledgement of a no less tangled mix7. this article takes stock of the treatment of timescapes within the policy science literature. it goes on to consider timescapes within the complexity literature—and the points of convergence between the two. on these conceptual foundations, it examines policymaking within complex timescapes and the tools and capabilities on which policy practitioners can draw. the article shows how this can illuminate the policy world, both conceptually and practically. it draws on hirschman’s treatment of the “interlocking vicious circles” that hold development back and the “upward spirals” that can—when the time is right—be mobilised. it sets this within an appreciation of political economy and institutionalism. the article concludes by considering the analytical and practical tools that policy makers might use, to navigate complex land-timescapes. how are they to recognise critical junctures and tipping points, wicked problems and smart solutions? how should they decide when to wait, when, and where to act? and not least, how is the social scientist to subject those decisions to stringent scrutiny? conceptual frameworks: the policy sciences the policy literature offers several idealised accounts of policymaking. each sheds some light on the policy world—policymaking as a process of evidence assessment and rational analysis (bardach, 2005), as a process of experimentation and “muddling through” (lindblom, 1959), as the manipulation of symbols (edelman, 1964), as the exercise of brute power (klein, 2007). none of these however brings centre-stage the complexity of policy timescapes. nevertheless, within the policy science literature there are some critics of that mainstream who show how this can be done. i take three classic treatments—and one more recent contribution. in his study politics in time, pierson (2004) argues for sensitivity to a variety of time-scales in the explanation of social phenomena. he questions the general focus in social science upon the immediate, where both causes and effects are manifest in the short-term. he challenges the tendency to attempt explanation by reference to “variables” that “are ripped from their temporal context” (p 1); and he questions whether “the social significance of historical processes can be easily incorporated 7 https://www.pnas.org/content/115/33/8252?mod=article_inline complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 https://www.pnas.org/content/115/33/8252?mod=article_inline 18 university of bamberg press into the ‘values’ of particular ‘variables’ at a moment in time” (p 78). he notes that the current focus on short-term and immediate causes and effects was not always so; he cites the modernisation literature (p 98) and its collapse in face of the critique it faced, in terms of functionalism and teleology. pierson underlines the importance of sequencing. the timing and direction of institutional reforms will shape future administrative capabilities and constrain subsequent options for social change. it may be difficult to reverse them because of the procedural time and trouble involved and the benefits they generate for significant constituencies. many of the studies he reviews stress the fixity of institutional legacies: clay rather than putty, swords that only with difficulty can be turned into ploughshares. pierson also however deals with situations in which, notwithstanding such path dependency, there may be occasional bursts of change, set in motion—but maybe at some remove—by earlier and external critical events. slow changes in background social and economic conditions may have effects that are far from gradual; at particular thresholds or “tipping points,” there are sudden avalanches of change (mcgee and jones, 2019). at times pierson adopts the language of attractors and trajectories that writers on complex systems employ. he may not offer a toolkit of formal models. nevertheless, his study reveals the convergence between historical institutionalism and complexity science and it reinforces the case for a shared analytic approach to social dynamics. abbott’s time matters (2001) deals with trajectories and attractors, turning points and fateful choices. he considers the time scale—long-term or more immediate—within which particular processes unfold, whether as effects or as the causes of other processes. he refers to triggers and switches—events and processes which create thresholds or tipping points, critical junctures, and bifurcations. he discusses the whole sequence of developments, by virtue of which an agent redeploys—but never escapes—the endowments of the past. abbott (1977) elsewhere considers how social dynamics can be analysed by reference to both time and space. for the chicago school of sociology, some social dynamics played out within neighbourhoods whose connections with the larger socio-economic system were stable and rather limited. in consequence, it was possible to study the sequence of stages through which each such neighbourhood typically developed, with only limited reference to these larger contexts. other neighbourhoods—and other social spaces such as organisations and professions for example—involve regular abrasion against environing factors; they variously pursue their “careers” in relation to that larger context, or have careers thrust upon them. finally, such abrasion may be so great that the research focus is necessarily on the interactional processes themselves—the “macro” co-evolution of neighbourhoods—and the cascades of change that such interactions—not always predictably— unleash across the city. kingdon (1984) (agendas, alternatives and public policies) has also become an established classic of policy timescapes. social problems, policy ideas, and political agendas each develop in their own way, but when they happen to align, the proponents of a particular policy idea may find that a window of opportunity opens, as political actors embrace it, for addressing a particular social problem. kingdon uses a series of case studies of major national policy changes in the us, to explore how these alignments develop—and the combination of patience and agility displayed by the proponents of particular policy ideas. nevertheless, there are limitations to kingdon’s treatment. power and political economy affect the politics of policy change, but he leaves them rather in the background. the focus moreover is on the cockpit of political power, washington dc—that is where his empirical case studies were conducted, but somewhat to the neglect of the dynamics of social change in the wider society. kingdon is right to study the social construction of political reality—but that reality is not just the construct of politicians. kingdon offers qualitative lessons drawn from his case studies about the development of conjunctures and the timescapes of the policy world. especially in the second edition, he also acknowledges the new literature on complex systems as a source of analytical insights. to that extent, he foreshadows the argument for complexity approaches to be integrated with institutional analysis and political economy (2011, 2016). complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 19 university of bamberg press finally, among more recent writers, jacobs (2011) (governing for the long term) examines policies which have their costs and benefits spread over the short and long run, the distributional consequences for different social groups, and the political consequences for politicians concerned with their own electoral survival. he assumes rational citizens, who will heavily discount the promise of long-run future benefits, and rational political leaders, who will avoid imposing unpopular short-run costs. he asks whether those leaders will ever “govern for the long-term.” in grappling with the question of how and whether it is ever rational for political leaders and citizens to embrace policies for the long-term, jacobs confronts some aspects of what i am calling the complexity of policy landscapes and timescapes. first, any policy decision to invest in new production possibilities will shift the society to a new set of capabilities. this will in turn change the consumption possibilities available in future periods. jacobs points out that with pensions policies for example, many scholars have studied the distributional question in terms of transfers from today’s workers to today’s pensioners, rather than as an inter-temporal re-allocation of consumption between today and tomorrow, by reference to the new capabilities landscape that has by then emerged. it is he argues misleading to assume that the long-term is connected to the short term only by the discount rate that is applied while starting from an assumption of rational actors, much of jacobs’ discussion reveals how remote from the real world that assumption is. rationality is heavily bounded, surrounded by much uncertainty. policy choices made now will set in train a myriad responses among social and economic actors on the ground, whose interactive consequences will become evident only in the course of time. there will be challenges and resistance from organized interests, watching for opportunities to act at critical moments, to sabotage the policy or to ensure that its costs are diverted onto other groups. jacobs acknowledges (pages 53-57) the “complex causal chains” involved in these dynamics. like kingdon, he does not offer a clear analytic for handling these; nevertheless, he makes reference to discussions of “punctuated equilibria”—a key feature of the literature on complex systems—and to the importance, again, of integrating this with institutional analysis and political economy. jacobs concludes with the need to study political ideas, the mental maps that policy makers use and the patterns of information processing that characterize human cognition. no less important is the quality of political leadership, appealing to moral commitments and practical interests and telling a persuasive story of a future to be achieved and futures to be avoided—futures that, as jacobs argues, are no mere extrapolations from the present, but involve shifts to a different landscape. to this extent, political leaders may seek to lay claim to—or even we might say to occupy—one of those futures, laying out its contours and binding the community to its realisation. these four policy scientists all recognise the bounded rationality of the policy actors they study. they focus primarily on the bounds that arise from the limited information available and the limits of human cognitive powers. the main limitations however to rational decision-making and foresight are those that arise from the complex social dynamics that these actors confront, the counter-intuitive processes of emergence they entail and the complex timescapes within which they present moments for decisive action. all four writers are innovators in handling these fundamental ontological and epistemological challenges. they acknowledge the complex dynamics exhibited by the social and policy systems in which these actors find themselves. the models that they use of such complex dynamics are however rather general and they make little explicit use of the rich stream of models issuing from complexity science. it is therefore to those models that we now turn, in an effort to forge a closer synthesis of the two literatures (room, 2011: ch 1). conceptual frameworks: fitness landscapes critical junctures are moments when the landscape shifts, to reveal new opportunities or dangers, where decisive action is both possible and necessary. icebergs come together to create a temporary bridge, over which we can move if we hurry; or an ice floe may break up and oblige us to complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 20 university of bamberg press scramble immediately to safety. in contrast, if everything is connected in a continuous ice sheet, we can venture forth whenever we want—time is much simpler. critical moments are thus intimately bound up with the particular landscapes within which actors find themselves: social time and social space are co-produced. to discuss complex timescapes it will therefore be convenient first to consider complex landscapes, where there is a more wellestablished literature (for a critical overview, see gerrits and marks, 2015). the biological: kauffman kauffman (1993) applies the literature on complex systems to biological systems and evolutionary dynamics. his first concern is with the evolutionary fitness of the populations of different species. a fitness landscape displays how the genetic variations underlying population characteristics (the horizontal plane in figures 1 and 2 below) affect their fitness (the vertical dimension) in the daily struggle for existence. the horizontal plane of our diagram is limited to just two dimensions—whereas of course in the real world the genotype of even the simplest organisms will vary across many dimensions. in darwinian models of evolutionary development, populations will (over many generations) throw up variations in each of these dimensions. where greater overall fitness results from one of these variations, the population of the species in question can advance to a higher point on the fitness landscape (figure 1). if this is generally the case, the fitness landscape will display a single peak, towards which all journeys lead. if however a fitness improvement in relation to one endowment means a decrease in relation to another – if in other words there are trade-offs between these different genetic investments – the fitness landscape, instead of offering a smooth ascent, will instead be rugged. any evolutionary journey across such a landscape is replete with points where the ‘choice’ of one line of ascent may ultimately lead to a low fitness peak, from which no subsequent improvements of position are readily available (figure 2). there is strong path dependency in these transitions; and the sequencing of moves matters. figure 1 figure 2 the contrast between the single peak and the rugged fitness landscape is what kauffman refers to as his nk model. this is however not all. kauffman goes on to consider how the fitness landscapes that different species face may be linked through the food webs in which they are involved. this is his nk(c) model. one species may, for example, thrive the more that another thrives, as with flowering plants and insects; their co-evolution will typically involve increasing specialization around this particular interdependence. in the case of predators and prey however, the increasing predatory capability of the one species means that the prey survives only insofar as its population develops greater complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 21 university of bamberg press capabilities of escape or disguise strategies which then in turn incite further shifts in the capabilities of the predator population. the fitness landscape of each species can thus be re-shaped by the shifts that other species make across their own fitness landscapes. what previously appeared to be a single peak fitness landscape may become increasingly rugged, so that a species gets stuck on a low peak – or indeed, it may find the peak breaking up and collapsing beneath its feet, so to speak. alternatively, it may discover particular areas of its fitness landscape are being raised to new heights, a narrow and brief window of opportunity for advance. the result of these co-evolutions may be a stable state, where the participating species discover a shared ecosystem that allows both of them to thrive, or a “red queen” dynamic, an arms race where each must keep moving fast, if it is to stay in the same place8. between these two extremes, there lie many different forms of emergent structure. periods of equilibrium are punctuated by moments of more or less sudden change; shifts by one species may open up or close off opportunities for fitness improvement by the other, even if it takes time for these effects to work their way through and become manifest. how shall we understand the temporal dimension of such fitness landscapes? within the nk model, the single peak “mount fuji” in figure 1 involves no trade-offs between different genetic endowments—hence the smoothness of the ascent, from any direction, at any time, towards the peak. in contrast, the rugged landscape of figure 2 involves strong interactions and trade-offs between those endowments—and the danger of ending up on a low fitness peak. past investments both enable and limit what future ascents are possible. nevertheless, these constraints of the past are a given—no brief moments when they lift and allow a new ascent. contrast that with the temporal dimension of the nk(c) model. here the movement of one species across its fitness landscape produces movements in the fitness landscape of another— changing the patterns of ruggedness that now confront that second species. this moreover is a dynamic and temporal process, producing periods of stasis and blockage but then maybe new opportunities for ascent, as the landscape shifts beneath the feet, so to speak, of the species in question, if only for a moment. it is here that kauffman’s fitness landscape becomes a complex timescape also. the social: potts and crouch these fitness landscapes can provide a powerful toolkit for thinking about complex landscapes within not only biological but also social systems. but how are we to move from the language of genetic endowments, fitness, and food webs to the social and policy world? the answer is in terms of capabilities, positional advantage and institutional connections (room, 2016: ch 5). these can be illustrated through the work of potts (2000). he offers, from the standpoint of complex systems, a critique of orthodox economics and walrasian equilibrium. the latter assumes an economic space where all buyers and sellers have free and ready access to each other. this is a fully connected landscape, a “well-mixed pot.” this in turn allows all markets to clear. futures markets with appropriate discount rates extend this beyond the present. neither space nor time has any structure; they are smooth and featureless. against this, potts argues that in the real world, economic space is a “non-integral” network— not everything is connected to everything else. while some nodes have multiple connections (the hubs or seaways of the world), others are accessed only via one or two connections (the remote and wellinsulated hinterlands). potts’ non-integral space is kauffman’s rugged landscape, where easy transitions are often barred by deep valleys. it is from this standpoint, and inspired by the literature on complex systems, that potts brings together diverse critiques of walrasian orthodoxy (including for example simon on decomposable systems (earl and potts, 2004).) 8 kauffman reserves the designation of red queen to extreme versions of such arms races, where those involved are “forever doomed by their own best efforts to… deform ... their own landscapes” (kauffman, 1995: 223). however, some evolutionary biologists would use the term somewhat more generally, to mean evolutionary change that is continuous due to antagonistic feedbacks between two co-evolving partners. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 22 university of bamberg press potts proceeds to shift the ontology of economic life away from walrasian exchange, back to production and capital accumulation, as in the classical tradition of economics. instead of buying and selling in markets, the primary focus is now on economic actors and firms endowed with capabilities: there are resonances here with penrose (1959). these they acquire, develop, and recombine, so as to build positional advantage in terms of technological and market dominance. in the fitness landscapes in the figures above, the horizontal plane now captures these diverse bundles of capabilities, the vertical (fitness) dimension displays the positional advantage the economic actor enjoys, in relation to rivals9. for potts, those capabilities are nested and activated within systems of micro-, meso-, and macro-rules (dopfer and potts, 2008: ch 3), including property relations. social and policy scientists will more commonly speak of institutional rules. it is these that decompose economic and social space, but also then mediate the dynamic interdependencies of different spaces. they thereby constrain and channel the exercise of capabilities. moreover, by tethering social actors within rule-based environments, they enable others to exercise imperative coordination and to dominate a wider social canvas. the stronger these tethers, the less scope there is for the social actors caught within them to interact outside those rules and the easier for the powerful to shape and predict the future. crouch (2005) offers an account of agile institutional entrepreneurs that closely parallels potts and his technology entrepreneurs. like potts, he places centre-stage creative actors who, armed with mental models of how the world works, scan the array of available materials with a view to combining them in new ways. in the case of crouch, the available materials are institutions; they are combined using systems of governance, selected by reference to the composite actors they bring together and the capacities and knowledge assets they enable them to deploy. both potts and crouch insist that actors thus find themselves on terrains that are “complex.” modelling the environment in which actors finds themselves must therefore be in sufficiently “fine detail” to respect this complexity and must not treat actors as though they were located in a topographically uniform “abstract space” (crouch, 2005: 101). both think of the entrepreneur as searching for new institutional or technological combinations using simplified templates. there are differences between potts and crouch, but the parallels are more important. purpose, power and positional advantage kauffman uses fitness landscapes to explore the evolution of different species and populations. potts and crouch adapt this approach to the world of technological and institutional transformation and struggles for positional advantage. micro-actors experiment with a myriad innovations, but the macro-world evolves through emergent processes, hardly less blind than the evolution of darwinian biology. history is made largely behind their backs; only to a limited extent can they glimpse and act upon its emerging contours. larger actors—corporate and governmental—can in greater degree shape the world to their own ends. to apply the notion of fitness landscapes to the social and political world requires that we recognise not only of the blind dynamics of the micro-interactions between small actors, but also the purposeful interventions by big actors, as they build their positional advantage. it is in terms of such positional advantage that room (2016: ch 5) reviews the conceptual value of fitness landscapes. he starts with the “one-dimensional view” inspired by hirsch (1977): positional advantage as a monotonic pecking order. on a fitness landscape, this is positional advantage as the vertical height of the landscape at any point. there is also however a “two-dimensional view” concerned with access and exclusion, connections across boundaries. on a fitness landscape, this means easy movement to new positions, rather than being isolated on a low fitness peak of kauffman’s nk rugged landscape (see also burt, 2004 on "structural holes"). 9 there is a large literature in complexity economics that explores the implications of this: see for example hidalgo (2007), with reference to national technological capabilities and global positional competition. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 23 university of bamberg press finally however, there is a “three-dimensional view” of positional advantage, as the scope to shape the co-evolutionary dynamics of linked fitness landscapes. here we are back with kauffman’s nk(c) model – but with social actors seeking to re-work the system parameters, so as to steer those dynamics to their own advantage. this is a process with periods of stasis and blockage, but then maybe opportunities for ascent: in short, a complex timescape. these purposeful struggles for positional dominance are the stuff of political economy and policy analysis. even as they watch and wait, these strategic actors look to enforce their occupancy of the institutional landscape, and thereby to “occupy the future.” fitness landand timescapes, when thus extended to the social and policy world, are drenched in inequalities of power. path dependency chains of the past or capabilities for the future? path dependency is a central preoccupation of many policy scientists, but also of writers on complexity. it will pay us to examine this is greater depth, as we attempt to forge a mutually enriching synthesis of these two intellectual traditions. for historical institutionalists in particular, such path dependency operates primarily via the institutions within which policy is developed. this sort of lock-in is well-illustrated by pierson’s writings on the us congress (hacker and pierson, 2010). path dependency is no less a preoccupation of complexity writers, examining the extent to which a complex system carries within it a “memory” of its origins. this includes the rugged landscape of kauffman, where depending on the evolutionary route followed, a species may find itself caught on a low fitness peak, with no easy escape. for both literatures, path dependency involves the constraining chains of the past, hemming us in and constraining options for the present and the future. jacobs however notices that any policy decision to invest in new production possibilities will shift the society towards a new future capabilities landscape. reforms to pension policies should therefore be viewed as a long-term re-allocation of consumption between today and tomorrow, within whatever landscape of new capabilities has by then emerged. (this is very much within the economics tradition of alfred marshall and keynes, where the “short-term” is defined in relation to fixed capabilities, while the long-term allows for changes in capital and capabilities.) also within the institutionalist literature, a similar perspective is offered by crouch (2005), who as we have seen recognises institutions as a source of change in public policy, in the hands of agile institutional entrepreneurs. the same is true of agile actors trapped on an nk rugged fitness landscape. for simplicity, we commonly visualize such a landscape as having just two dimensions of capabilities, set in the horizontal plane. new forms of capability will bring additional dimensions, however; a low peak can then become a saddle point, through which the system can pivot, to take advantage of this new range of capabilities and thus improve its fitness10. this might arise in the biological world from sexual recombination. in the social and technological world, such additional dimensions are limited only by human inventiveness and investment in new capabilities. the existing situation is therefore not so much a prison as a launch-pad, a pivot to new ascents. embedded within it, we enjoy the security and familiarity from which we can make sense of the world; but we can also identify the ways in which the present order is constraining the potentialities of the present and the new capabilities that will be needed, in order to escape these constraints. typically, such a break-out will build on existing assets but combine them with new or external assets in new ways. by these novel combinations of the existing and the new, we transform our capabilities and organise them into new configurations. path dependency is here not a lock-in, an 10 this conversion of peaks into saddles was recognised by fisher, whom gavrilets (2004: 36-7) however criticises for suggesting that the new ascents thus made available would enable a population to move towards a single global fitness peak. whether such a global peak exists is of course a quite separate matter. nevertheless, having criticised fisher on the question of a global peak, gavrilets does not himself pursue the implications of successive monotonic ascents as new dimensions of genotypic space open up. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 24 university of bamberg press impasse, constraining us to more of the same; instead, it provides a vantage point from which we discover pathways to innovation, and we escape from the chains of the past. here therefore the rugged nk fitness landscape offers a new dawn, with the possibility of new ascents. indeed, under these new conditions, new co-evolutionary interactions may also develop with other landscapes, of the sort that are modelled by kauffman’s nk(c) model and its complex timescapes. policymaking on complex timescapes consider now the practical implications for policymaking, viewed as a strategic walk. this recognises choice and agency in relation to a range of possible futures. it recognises multiple actors pursuing interdependent but potentially conflicting aims, each of whose plans of action must be taken into account. it acknowledges uncertainty in relation to the larger social, economic, and physical environment, within which the strategy will unfold. it recognises fateful choices—the paths not taken, the windows of opportunity ignored, the relationships of mutual benefit never explored. to speak thus of strategy involves recognising the complexity of timescapes. this is quite different from a timeless world, where decision-making can be boiled down to the rational calculation and comparison of alternative inputs and outcomes, and the associated costs and benefits. instead, it acknowledges that the interactions among multiple actors unfold over time, and they cannot be entirely predicted. the strategic policymaker must therefore wait and watch—and intervene only when the moment is right. the leader who has a new strategic vision each day will soon exhaust the emotional energy of their followers, unless the environment they face is so redolent with danger or with promise that frequent adjustments to strategy are self-evidently necessary. the leader, who waits too long, risks being brushed aside—"too little too late.” what then is a strategic policymaker? what is it to engage with policy on a complex timescape? hirschman’s heterodox ideas on development policy, as set out in the strategy of economic development (1958), have in recent years attracted renewed interest, in part through the work of edelman (2013). hirschman asks: what are the preconditions for economic development? the mainstream literature answers in terms of particular resources—natural resources, capital, entrepreneurs, etc. but, hirschman points out, it proves difficult to agree empirically on which of these is key; and indeed, once development gets going, somehow they all fairly readily appear. hirschman therefore offers an alternative view. what seems to matter much more than any particular resource are the “interlocking vicious circles” that hold development back, and, in contrast, the “upward spirals” that can bring forth all the resources that are needed. the focus should therefore be not on the resources themselves but on the “essential dynamic and strategic aspects of the development process” (p 6). hirschman adds however that many of these resources may be latent rather than immediately available. development depends on mobilising and combining these purposefully, but also in a spirit of experimentation—trying out different makeshift adaptations and finding which ones will work. it is this capacity—to mobilise and combine, to adapt and redeploy—that hirschman therefore places at the centre of development strategy. it is inappropriate to embrace a strategy of “balanced growth,” hoping to make simultaneous progress across all sectors. instead, “if the economy is to be kept moving ahead, the task of development policy is to maintain tensions, disproportions and disequilibria,” mobilising in force against particular critical points (p 66). we must moreover locate this within an appreciation of the exercise of social, economic, and political power within the society concerned. this in turn requires recognition of the alternative possible futures that development strategies may offer, the political choices and trade-offs involved and the key turning points. hirschman, to repeat, refers to the “interlocking vicious circles” that hold development back. this is reminiscent of the language of “wicked problems” (rittel and webber, 1973). policy makers may want to disentangle the convoluted mess that history has produced. that is sometimes possible in a physical system—for example, a tangled skein of threads—but it is impossible to unscramble an omelette, however well one understands the processes by which it was produced. it is also rare in most social situations; in the real world no groundhog days are possible, re-setting the clock to zero. this is complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 25 university of bamberg press why parties to a historical conflict, if they are to make a new beginning, often need to reach for a shared account of the hurts entailed by their shared history—and the responsibility and wickedness it involved. here the deployment of models of complex interactive change may well have a role, especially in making evident the interaction of many factors, in producing those injustices. hirschman also however refers to the “upward spirals” that can bring forth all the resources that are needed. these he places at the centre of attention for the development policy strategist. so also, alongside the wicked problems that the policy maker faces, there may be corresponding “smart solutions,” whereby careful attention to the interrelationships of these problems, “upward spirals” of dynamic interdependence can be set in motion, through which these problems can be jointly addressed. this might in a physical system mean disentangling a skein of threads and then using them to weave a new garment. nevertheless, wholesale disentangling is not always necessary or appropriate, even where it is possible. the tangles themselves may have produced combinations that can now be re-worked, with each other and with other resources, to create something novel and useful. this is why hirschman expects development strategists to be sufficiently agile and imaginative, to break out from these tangled constraints and path dependencies, mobilising new micro-interactions among them, new disequilibria that confound pessimism and complacency. this requires recognition of the alternative possible futures that development strategies offer, the political choices and trade-offs that are involved and the key pivot points. the more we conceive of the policy maker in hirschman’s terms, the more appropriate the earlier conceptual discussion of fitness landscapes in sections 4 and 5 now appears. that discussion culminated in kauffman’s nk(c) model—applied not to co-evolving species, but to social actors seeking to re-work the system parameters within a complex timescape. the policy maker is constrained by the interlocking blockages that hinder progress but hopes from there to see what new combinations of capabilities—and what new institutional connections—will transform low fitness peaks into saddle points for fresh advance. recall also the discussion of path dependency and the chains of the past, entangling us and blocking any forward movement. no less, recall our argument that path dependency is janus-faced: so that nestling among those chains, innovators may be able to identify how—and when—a breakout into new lines of ascent may become possible. the moments—the critical junctures—at which such turns can be made may be brief. this applies to many of the major interlocking policy problems that our complex societies face, including climate change and the covid pandemic. nevertheless, wait and watch for the mists to clear—the moment for decisive action may be there for the taking. tools for policymaking faced with such challenges, how are policy makers to navigate these complex land-timescapes? how are they to recognise critical junctures and tipping points, wicked problems, and smart solutions? how should they decide when to wait, when and where to act? what analytical and practical tools can we, as social scientists, provide? how can we enhance the strategic imagination of the policy maker and the elucidation of alternative societal trajectories? at the analytical level, the scholarship that has developed over recent years offers many models of complex systems that are applicable to the policy world: not only kauffman’s fitness landscapes, but others such as agent-based modelling and cellular automata, in the tradition of schelling (1978) and holland (1995) and evolving networks (jain and krishna, 2003). policy researchers have applied these models to simulated and empirical datasets. they also engage actively with the policy world, expanding the range of “mental models” that policy actors can deploy, as they consider how their complex world may unfold11. the present article challenges all these writers on complexity—using their wide range of approaches, beyond that of fitness landscapes—to take much more seriously “the timing of space.” it 11 for example, cecan at the university of surrey: https://www.cecan.ac.uk/ complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 https://www.cecan.ac.uk/ 26 university of bamberg press also however insists on embedding this within an appreciation of political economy and power, something that they often miss (room, 2015). to exercise power is to expand one’s own freedom of manoeuvre and to limit the times when others have the opportunity to choose which card to play. power is the luxury of being free to decide when to act to maximum advantage, rather than being rushed by the throng. at the practical level, the need is for “weak signals” of impending shifts in the policy landscape—critical junctures and pivot points, where the policy maker can act decisively. this is a search for tools of practical monitoring and navigation—maybe a whole dashboard, or at least some simple rules of thumb. scheffer argues that there are generic symptoms of complex adaptive systems that are approaching a “tipping point” (scheffer, bascompte et al., 2009). these include for example a slowdown in the recovery rate after small perturbations and increased variance in the pattern of fluctuations. fisher (2011) looks for “weak signals” of impending change: surface signs of deeper dynamics that are underway. he begins with the natural world—toads migrating in response to earth tremors for example—but also climate change and biodiversity. examples from the social world might include the reputation of a school falling and parents hastily switching their allegiance elsewhere (room and britton, 2006). in some developing countries, signs of impending distress include the eating of the seed corn for next year, or the age of marriage for daughters falling (chambers, 1989, bevan and sseweya, 1995, indra and buchugnani, 1997, carney, 1998, moser, 1998). these signals are however all very context-specific. these rules of thumb often refer to thresholds (room, 2016: ch 6.4). in relation to the covid pandemic, thresholds include the r-value; in relation to climate change, rates of ocean warming12. these indicators are often embedded within a prognosis of the timescale within which action, if taken, can be expected to bring the disaster in question under control, or the sequence of consequences to be expected, if that timescale is missed (werners, pfenninger et al., 2013, steffen, rockström et al., 2018). their selection is always however contestable, predicated as they are on our incomplete understanding of the complex dynamics in question. not that this is a language that is only for the potential disasters that face our societies. the policy landscape also encompasses the technology ecosystem. here the literature in complexity economics, referred to earlier, is the basis for policy-related research into national and regional technological capabilities and global positional competition, and the indicators by reference to which policymakers can identify new and profitable niches, wherein they can re-position their regional and national economies (pugliese and tübke, 2019). conclusion the article began by asking what analytical and practical tools the policy makers might use to navigate complex timescapes. how are they to recognise critical junctures and tipping points, wicked problems and smart solutions? how should they decide when to wait, when, and where to act? and not least, how is the social scientist to subject those decisions to stringent scrutiny? these questions are not new. on the contrary, they have long been central to sociological debate—perhaps most obviously in weber’s lectures on science and politics as vocations at the end of ww1 (gerth and mills, 1948). weber addressed himself to germany’s political leaders, facing a mass of tangled and wicked problems, amidst the ashes of germany’s defeat. his task as a social scientist was not to compare alternative ways of tackling some specific policy problem or to produce an evidence-based assessment of an intervention. it was, instead, to identify the interlocking and complex problems that germany face, to illuminate the available policy choices, to notice the “weak signals” of the societal shifts that were underway, and to foresee the 12 https://www.theguardian.com/environment/2020/jan/13/ocean-temperatures-hit-record-high-as-rate-of-heatingaccelerates complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 https://www.theguardian.com/environment/2020/jan/13/ocean-temperatures-hit-record-high-as-rate-of-heating-accelerates https://www.theguardian.com/environment/2020/jan/13/ocean-temperatures-hit-record-high-as-rate-of-heating-accelerates 27 university of bamberg press shifting coalitions of interests that these would likely produce. this remains the fundamental task of the social scientist, in relation to the policy world. the task of the policy maker—at least in more or less democratic societies—is to interpret the world but also to change it and to provide a justificatory account as to the journey and the destination, a justificatory account that the social scientist will then however expose to stringent scrutiny. the conceptual framework elaborated in this paper is to be judged, not least, by the extent to which it helps to sharpen that scrutiny. references abbott, a. (1977). "of time and space: the contemporary relevance of the chicago school." social forces 75(4): 1149 1182. abbott, a. (2001). time matters: on theory and method. chicago, university of chicago press. adam, b. (1990). time and social theory. cambridge, polity press. adam, b. (1998). timescapes of modernity: the environment and invisible hazards. london, routledge. adelman, j. (2013). worldly philosopher: the odyssey of albert o hirschman. princeton, princeton university press. bardach, e. (2005). a practical guide for policy analysis (second edition). washington dc, cq press. bevan, p. and a. sseweya (1995). understanding poverty in uganda. oxford, centre for the study of african economies. burt, r. s. (2004). "structural holes and good ideas." american journal of sociology 110(2): 349-399. carney, d., ed. (1998). sustainable rural livelihoods. london, uk department for international development (dfid). chambers, r. (1989). "vulnerability: how the poor cope." ids bulletin 20(2): 1-7. crouch, c. (2005). capitalist diversity and change: recombinant governance and institutional entrepreneurs. oxford, oxford university press. dopfer, k. and j. potts (2008). the general theory of economic evolution. london, routledge. earl, p. e. and j. potts (2004). bounded rationality and decomposability: the basis for integrating cognitive and evolutionary economics. models of a man: essays in memory of herbert a simon. m. augier and j. g. march. cambridge, mass, mit press: 317-333. edelman, m. (1964). the symbolic uses of politics. urbana, university of illinois press. fisher, l. (2011). crashes, crises and calamities. new york, basic books. gavrilets, s. (2004). fitness landscapes and the origin of species. princeton, princeton university press. gerrits, l. and p. marks (2015). "the evolution of wright’s (1932) adaptive field to contemporary interpretations and uses of fitness landscapes in the social sciences." biol philos 30: 459–479 gerth, h. h. and c. w. mills (1948). from max weber: essays in sociology. london, routledge. hacker, j. s. and p. pierson (2010). winner-take-all politics. new york, simon and schuster. hidalgo, c. a., et al. (2007). "the product space conditions the development of nations." www.sciencemag.org 317. hirsch, f. (1977). social limits to growth. london, routledge and kegan paul. hirschman, a. o. (1958). the strategy of economic development. new haven, yale university press. holland, j. (1995). hidden order: how adaptation builds complexity. new york, basic books indra, d. m. and n. buchugnani (1997). "rural landlessness, extended entitlements and inter-household relations in south asia." the journal of peasant studies 24(3): 25-64. jacobs, a. (2011). governing for the long term. cambridge, cambridge university press. jain, s. and s. krishna (2003). graph theory and the evolution of autocatalytic networks. handbook of graphs and networks. s. bornholdt and h. g. schuster. weinheim, wiley-vch: 355-395. kauffman, s. a. (1993). the origins of order: self-organisation and selection in evolution. oxford, oxford university press. kauffman, s. a. (1995). at home in the universe: the search for laws of self-organisation and complexity. harmondsworth, penguin. kingdon, j. w. (1984). agendas, alternatives and public policies. boston, little, brown and company klein, n. (2007). the shock doctrine. london, penguin lindblom, c. e. (1959). "the science of "muddling through"." public administration review 19. lukes, s. m. (2005). power: a radical view (second edition). basingstoke, palgrave macmillan. mcgee, z. a. and b. d. jones (2019). "reconceptualizing the policy subsystem: integration with complexity theory and social network analysis." policy studies journal 47(s1): s139-158. moser, c. (1998). "the asset vulnerability framework: reassessing urban poverty reduction strategies." world development 26(1): 1-19. penrose, e. (1959). the theory of the growth of the firm. oxford, blackwell & mott. pierson, p. (2004). politics in time. princeton, princeton university press. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 http://www.sciencemag.org/ 28 university of bamberg press potts, j. (2000). the new evolutionary microeconomics: complexity, competence and adaptive behaviour. cheltenham, edward elgar. prigogine, i. and i. stengers (1984). order out of chaos. london, heinemann. pugliese, e. and a. tübke (2019). "economic complexity to address current challenges in innovation systems: a novel empirical strategy for regional development.” jrc policy insights: industrial r&i. rittel, h. w. j. and m. m. webber (1973). "dilemmas in a general theory of planning." policy sciences 4(2): 155-169. room, g. (2011). complexity, institutions and public policy: agile decision-making in a turbulent world. cheltenham, edward elgar. room, g. (2015). complexity, power and policy. handbook complexity and public policy. r. geyer and p. cairney. cheltenham, edward elgar. room, g. (2016). agile actors on complex terrains: transformative realism and public policy. london, routledge. room, g. and n. britton (2006). "the dynamics of social exclusion." international journal of social welfare 15(4): 280-289. scheffer, m., et al. (2009). "early warning signals for critical transitions." nature 461(3): 53-59. schelling, t. c. (1978). micromotives and macrobehaviour. london, w.w.norton. steffen, w., et al. (2018). "trajectories of the earth system in the anthropocene." proceedings of the national academy of sciences 115(33): 8252-8259. werners, s. e., et al. (2013). "thresholds, tipping and turning points for sustainability under climate change." current opinion in environmental sustainability 5(3-4): 334-340. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 15-28 doi: http://dx.doi.org/10.20377/cgn107 complexity in international society: theorising fragmentation and linkages in primary and secondary institutions university of bamberg press complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 complexity in international society: theorising fragmentation and linkages in primary and secondary institutions author: dennis r. schmidt swansea university, united kingdom email: d.r.schimdt@swansea.ac.uk this article seeks to contribute to theorising the institutional structure of international society by exploring synergies between complex systems thinking and the english school theory of international relations (ir). suggesting that the english school already embraces key conceptual insights from complexity theory, most notably relational and adaptive systems thinking, it reconfigures international society as a complex social system. to further advance the english school’s research programme on international institutions, the article introduces the notion of “law-governed emergence” and distils two effects it has on global institutional ordering practices: fragmentation and clustering. these moves help to establish complexity as a fundamental structural condition of institutional ordering at the global level, and to provide a basis for taking steps toward better understanding the nature and significance of institutional interconnections in a globalised international society. keywords: english school theory; institutional fragmentation; primary institutions, secondary institutions, complexity ir theory introduction the uptake of complexity theory in the discipline of international relations (ir) has been steady but slow. first attempts to incorporate insights from complexity science into international systems thinking were made in the 1990s but failed to generate a sustained and systematic research programme. in an effort to encourage a more systematic engagement, a number of recent contributions have sought to reconfigure debates and conceptual resources to build a more solid foundation for dialogue between complexity theory and ir theory (bousquet & curtis, 2011; kavalski, 2007). pointing to the pervasive and ever-growing randomness and uncertainty of international life, these scholars see an opportunity in complexity approaches to enhance current theoretical frameworks for the study of global politics. taking up this challenge, this article explores how complexity theory can help to advance the english school’s approach to ir theorising and, in particular, its account of international society’s institutional structure. once on the margins of the field, english school theory has seen a revival since the turn of the millennia, and it now maintains a thriving research programme centred around societal order and shared norms, rules and institutions at the global level. drawing on sociology, history, and political theory, its methodological approach to theorising international relations offers a valuable resource for studying the normative dynamics and institutional structure of world politics that is different from us-led mainstream ir theory.1 the claim of this article is two-fold. firstly, and more generally, i suggest that english school theory is inherently open to complexity thinking. this openness springs from its methodological orientation. proceeding from the basic assumption that international society is a dynamic social construct, english school scholars have privileged constitutive theorising over linear causality, historical contingency over presentism, and relational over essentialist ontology. in doing so, english school scholarship embraces and applies, though somewhat implicitly, a number of conceptual insights from complexity theory that are deemed valuable for advancing the study of international relations (kavalski 2007; bousquet and curtis 2011): open and adaptive systems thinking, historical process tracing, self-organisation, and multilevel interactions. 1 for the methodological differences between the english school and us-led ir theory, see bull's (1966) seminal essay "the case for a classical approach”. 94 mailto:d.r.schimdt@swansea.ac.uk 95 university of bamberg press secondly, and more specifically, i argue that incorporating insights from complexity theory can help to further develop the english school’s account of international institutions. while some progress has been made in understanding the reciprocal relationship between different types and levels of institutional structures (i.e. primary and secondary institutions), current english school scholarship has not fully engaged with concepts, such as fragmentation and linkages, that have been key to explaining complex governance arrangements in different issues areas. drawing more explicitly on these concepts, i suggest, can help english school scholars to improve their understanding of the dynamics and interconnections within and among international society’s different institutional ordering practices. the article is divided into three sections. the first section offers a necessarily brief overview of complexity theory in ir, introducing definitions and setting out key debates. the second section focuses on two conceptual insights from complexity theory – relational and adaptative systems thinking – and shows how they have featured into the english school’s approach to theorising the composition and evolution of the international system. here, i demonstrate how the school’s approach to the study of international relations has been built around a socio-historical understanding that emphasises changes in both the basic units of the international system and the deep-seated normative structures that drive shifts in system structures over time. recognising that current english school scholarship does not exhaust the analytical value and potential of complexity thinking, the third section considers how current debates about the institutional structure of international society can be further advanced. after setting out the english school’s conceptual distinction between primary and secondary institutions, i introduce the notion of “law-governed emergence” and distil two effects it has on global institutional ordering practices: fragmentation and clustering. the article concludes with a brief summary, pointing to some of the normative implications of complexity thinking for english school theorising. complexity theory in international relations: a brief overview complexity theory has become a broad field of study, and there exists no clear and widely accepted definition. rather than a unified theory that can be holistically applied (walby 2007, p. 35), complexity theory involves a number of concepts and approaches that have been developed in response to a particular problem: how to make sense of systems (both physical and social) that defy the logic of general laws, predictable behaviours, linear relationships, and the proportionality between cause and effect. accordingly, it has become common practice to define complexity theory in terms of the subject it seeks to study: complex systems. following a recent definition, a complex system is an “open system (…) that includes multiple elements (units) of various types intricately interconnected with one another and operating at various levels” (orsini et al 2019: 3). these systems, so the argument goes, need to be approached in distinct ways, because they display certain unique characteristics – that is, they evolve “at the edge of chaos” (lewin 1992, waldrop 1992; holland 1995), where unexpected patters, events, and outcomes are the product of self-organisation and emergent properties (orsini et al 2019: 3-4), rather than external shocks or large-scale structural transformations (kauffmann 1993). the ontological features of these systems, then, have epistemological consequences. analytically, complex systems are more than complicated, for they cannot be broken down into their individual parts (morin 1990). because the systemic whole is different than the sum of its parts (jervis 1997), the dominant scientific paradigm built around notions of causality and linear relationships is of limited value. in the social sciences, complexity theory has thus been convincingly associated with critical (or philosophical) realism and sociological naturalism (reed and harvey 1992), though different versions of complexity thinking continue to enlist a variety of epistemological approaches. building on these insights, i define complexity theory as a cognitive toolkit and associated research strategies explicitly designed to make sense of systems, both physical and social, in which structures, properties, processes, and outcomes arise from interactions among the elements of the system. this definition is consciously broad, aiming to capture the diversity of disciplinary and complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 96 university of bamberg press methodological approaches associated with complex system thinking. it also decouples complexity theory from a specific list of basic system features such as self-organisation and emergence, recognising that complexity theory can and should be applied to a broad range of systems marked by various characteristics and properties (and combinations thereof). the discipline of ir has had a patchy relationship with complexity theory, marked by continues punctual interventions rather than systematic engagement (e.g. albert, cederman, & wendt, 2010; hoffmann & riley, 2002). this is surprising, given that social, economic, and technological developments associated with globalisation have turned complexity into “a structural condition of world politics” that “provides the ontology behind challenging current research questions” (haas 2019: 5, in orsini et al). while traces of complexity thinking can be found in the early international systems theories of morton kaplan (1957) and oran young (1968), it was not until the late 1990s that ir researcher started to pay explicit attention to the phenomenon of complexity (e.g. axelrod 1997, rosenau 1990). most notably, robert jervis (1997) observed how certain system effects transformed the environment in which policy makers operate, pointing to issues and governance problems that arise from complex interconnections among social, economic, and political developments. today, much of the complexity agenda in ir is focused on exploring the dynamic relationships among international institutions (haas 2019: 7, in orsini et al). expanding webs of cooperation in diverse issue areas, ranging from refugee protection (betts, 2010; 2009), to climate change (abbott, 2012), to intellectual property rights (helfer, 2009), have produced a multitude of institutional arrangements that partially overlap and are not hierarchically ordered. so instead of focusing on discrete institutions, as early liberal regime theorists did, ir scholars have shifted the focus to the global governance architecture, understood as ‘the overarching system of public and private institutions that are valid or active in a given issue area of world politics’ (biermann et al 2009: 15). because the international systems lack centralised authority, many institutional elements of these governance systems evolved independently, covering different geographical regions and issue areas, and display different degrees of formalisation. they also frequently overlap in the sense that many issue areas and actors are within reach of more than one institutional arrangement. for example, the global institutional framework that governs international investment now involves around 3.000 bilateral treaties, lacking an overarching multilateral institutional framework that integrates these arrangements (johnston and trebilcock 2013: 622). these fragmentation dynamics have sparked rich conceptual innovations that seek to categorise and compare the nature and structure of institutional complexes (e.g. orsini et al 2013, biermann et al 2009). typically discussed under the banners of “regime complexity” and “orchestration” (haas 2019: 7, in orsini et al 2019), ir scholars have focused on the interconnections and linkages (e.g. young 1996) between individual institutions to understand the opportunities they create for governance innovations and sophisticated divisions of labour (e.g. gehring and faude 2014), as well as the negative effects of fragmentation dynamics (e.g. zürn and faude 2013). this work has also been very valuable in establishing the empirical relevance of complexity theorising: for example, by showing how specific institutions strategically respond to growing institutional proliferation by developing resources and re-defining organizational structures (e.g. betts 2009) the effects and governance of climate change have been a critical driver behind the integration of complexity theory into ir. most obviously, the governance of climate change is marked by an array of multilateral treaties, thereby displaying the classical characteristics of a fragmented global governance architecture (biermann et al 2009: 18). moreover, at the heart of the problem of climate governance lies the recognition that human and non-human (ecological) systems are deeply interwoven (kavalski 2011), producing emergent effects that cannot be traced back to one single system. this makes climate change an extremely complex phenomenon that generates unexpected non-linear societal and political events (campbell 2009) that often defy traditional modes of inquiry. while the bulk of complexity analysis in ir has focused on the meta-level of institutional complexes, aiming to make sense of the pattern and effects of functional interconnections, a number of constructivists have zoomed in on the nature of the linkages that connect institutions. peter haas’ (2013, haas and stevens 2017) work is exemplary of this approach. proceeding from the assumption complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 97 university of bamberg press that the boundaries and nature of complex issues are socially constructed, haas (2019: 7, in orsini at el) argues that the processes through which connections among institutions are formed, sustained, and modified can only be made intelligible by reference to the shared normative understandings in which they are embedded. this is an important insight, as it connects complexity theory to ir approaches that emphasise the role of shared norms, rules, and principles for the organisation of international systems. it is here where the english school approach to ir theorising enters the picture. tracing complexity in the english school’s research programme what broadly unifies english school scholarship is the core belief that international relations are not merely about power politics, fixed preferences and conflicting national interests, but also about identities and shared norms, rules, and institutions that order international life.2 as such, english school theory shares a number of theoretical assumptions and cognate research goals with ir constructivism (reus-smit, christian, 2009, p. 58): the focus on the cultural bases of state identity; the rule-governed nature of state-systems; the different forms of organising life under anarchy; and an intersubjective reading of the international world, where socially constructed institutions such as diplomacy, international law and sovereignty constitute agents’ identities and interests and ‘can exist only in virtue of certain intersubjective understandings and expectations’ (wendt, 1992, p. 412). the english school’s macro-sociological approach to theorising international affairs is perhaps best reflected in its key concept: international society. according to bull and watson’s (1984, p. 1) seminal definition, an international society exists when, “a group of states (or, more generally, a group of independent political communities) which not merely form a system, in the sense that the behaviour of each is a necessary factor in the calculation for the others, but also have established by dialogue and consent common rules and institutions for the conduct of their relations, and recognize their common interest in maintaining these arrangements.” understanding and theorising the distinct nature, emergence, and transformation of the deep-seated normative structures that constitute international society has been one of the defining features of the english school literature (clark, 2009; holsti, 2004). i will say more about those societal ordering practices, and how insights from complexity theory can help to further their study, at a later stage. for now, i will focus on showing how the english school approach to ir theorising displays quite a few of the traits that are central to complexity theory. relational and processual systems thinking systems thinking lies at the heart of complexity theory. at the simplest level, it presupposes that the elements under study cannot be considered independently from each other but are somehow connected through dynamic feedback processes and relationships, thereby forming a system. for classical sociological systems thinkers such as durkheim and parsons, those feedback processes would help to establish some kind of stability in the system. as walby (2007, p. 456) notes, “equilibrium was the norm to which a system would return if there were small a small deviation.” while complex systems theorists do not deny the existence of equilibriums and stable relationships within systems, they stress the way in which those feedback processes contribute to the evolution of a system beyond equilibrium (e.g. arthur 1994). on this view, the components of a system are entangled in complex micro-macro relationships that produce outcomes and configurations that defy causal analysis and path-dependence models (e.g. kauffmann 1993, waldrop 1992). english school theory is essentially concerned with the system level, as exemplified by the concept of international society. the explicit focus on the system level sets the english school apart from other sociological approaches to the study of ir. whereas constructivists, perhaps with the 2 for an excellent, more elaborate overview of the english school’s evolution, key concepts, modes of inquiry and eclectic methodological approach to ir theorising, see buzan (2014). complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 98 university of bamberg press notable exceptions of wendt (1999) and reus-smit (1999), have focused on the location, function and transformation of specific norms, the english school is largely concerned with “the way in which things fit together” (hurrell, 2001, p. 489). understanding the international system in societal terms comes with a relational and processual approach to systemic thinking that stresses dynamic flux in both the units and boundaries of the international system. here, the english school follows precisely the historical sociology that bousquet and curtis (2011, p. 49) identify as one of the perquisites for complexity theorising in ir – that is, a “processual ontology that privileges ‘becoming’ over ‘being’, dynamic flux over stable essence.” although the bulk of historical english school scholarship has been primarily concerned with the evolution of the society of states, its framework clearly rejects an essentialist ontology of social units. barry buzan and richard little (2000), for example, have developed a theoretical approach to the historical analysis of the international system that emphasises dynamics in both the interactions among the units of the systems (process) and the resources that spread throughout the system (interaction capacity). whereas the former relates to dynamics in what units do when they interact (e.g. changes in trade, political recognition, war), the latter is concerned with dynamics in the capacities that define what units can do (e.g. changes in transportation, communication, and weapons technologies) (buzan and little 2000, pp. 79-84). the result is a world historical perspective on international systems that introduces the idea of turning-points, such as the transformation from agrarian empires, chiefdoms, and nomadic empires to nation states, to develop a chronological account of how the international system began and evolved. this and other work (watson, 1992; wight, 1977) reinforces the claim that the dynamics of global social systems, and specific configurations of units and ordering structures within them, have always been contingent and subject to change. to get to grips with the more current systemic processes beyond the nation state associated with processes of globalisation, english school theory has begun to explore more systematically the concept of world society (buzan, 2004; weinert, 2017; williams, 2005). focused on cosmopolitan norms and institutions arising from the inter-human domain of international relations, the concept of world society has been explicitly designed to capture what bull (1977, p. 269) called the “totality of global social interaction” that includes “all parts of the human community.” bull did not flesh out the concept in detail, and there remains a good deal to be resolved in terms of rendering the concept of world society precise. yet, it is clear that world society encapsulates the values, institutions, relationships, and organizational processes that emerge from the interaction between non-state actors, including civil society, supranational organisations, and other transnational actors that operate beyond and below rigid, territorially bound hierarchical state authority structures. put simply, whereas states are the main actors in international society, world society involves institutions, processes, and normative arrangements that originate from interactions among non-state actors. here, a clear resonance can be found with ideas of decentralized networks, bottom-up processes, and forms of selforganisation frequently highlighted by complexity theory. looking more closely to complexity theory and its terminology and concepts could provide a rich resource for english school theorists to further carve out the processes of order formation and governance within the still underexploited domain of world society. adaptive systems: membership dynamics and the social construction of boundaries the processual notion of social units also carries forward into the school’s understanding of the boundaries of international society. whereas traditional systemic ir accounts (e.g. waltz 1979) have worked under the assumption that the international system is essentially closed, consisting of a few interacting parts, english school theory has always depicted international society as an open social system with boundaries that shift over time – an approach neatly reflected in bull and watson’s (1984) classical account of the expansion of international society. the ‘expansion narrative’ put forward by bull and watson is largely historical, telling the story of how colonialization and decolonization shaped the world according to european blueprints (buzan 2014, p. 45). it is also a eurocentric narrative, as complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 99 university of bamberg press bull (1984, p. 123) himself seems to concede at some point. yet, read in conjunction with more recent theoretical english school work on international social structures, it offers some important conceptual insights into how open systems evolve. according to the english school’s perspective, the dynamics that shape system boundaries in international relations are essentially normative, socially constructed through the interaction of political entities. the idea is that who is inside and who is outside international society matters, which places the focus on questions of membership in international society and the criteria for obtaining it. after all, international society is a club of clear participants (i.e. sovereign states) that have exclusive access to certain rights and institutions. the society of states that began to emerge in 15th century europe, for example, is recognised as a distinct social system with boundaries, because it involved a number of rules, norms and institutionalised ordering practices, such as the development of a common body of international law and rules around managing the balance of power, that were specific to the relations among european states (e.g. keene’s 2002). while the terminology of historical english school scholarship does not always fit easily with the more abstract conceptual language of complex systems thinking, it does cast light on the dynamic, socially constructed nature of system boundaries in international relations. barry buzan’s (2004) structural work offers further insight into the processes through which international society has expanded. building on wendt’s (1999) three cultures of anarchy, buzan (2004, pp. 139-158) identifies three mechanisms through which social structures are internalised: coercion, calculation, and belief. his model is, of course, an ideal type. in practice, system structures are internalised through a complex mixture of those mechanisms. however, by formalising social systems thinking in international relations in terms of mechanisms of internalisation, buzan’s scheme offers a concrete toolkit for investigating two key questions: how and why social systems are shared, to what degree, and with what kind of social opposition? the emergence of regional normative orders is another area in which membership dynamics shape the formation of system boundaries. in an attempt to get to grips with regional variation in international ordering practices, english school writers have recently introduced the notion of “regional international society” (schouenborg, 2012; stivachtis, 2015; stivachtis & webber, 2011). driven by the rise of regionally confined orders and governance arrangements, those accounts highlight how international norms and institutions assume different meanings and forms in different regional contexts. this is an interesting insight for ir systems thinking, because of the way in which it brings the notion of sub-systems into the picture, and the ways in which they are distinct from, but interact with, the overall international system in which they are embedded. it is important to understand here that while the boundaries of those systems are somewhat wedded to geographical context, membership in regional international societies and their distinct normative orders is ultimately defined through shared membership norms (thomas, 2017). before moving on, a few general remarks on how complex adaptive (cas) thinking can improve international society theorising seem to be in order. like complexity theory in general, cas consists of a number of approaches and concepts rather than a unified theoretical framework; figuring out exactly which of those approaches can be coherently aligned with and applied to the evolution of international society ought to be key for developing the english school as complex ir theory. the most obvious entry point here seems to be the notion of co-evolution. populated by biologist stuart kauffmann (1993), the idea of co-evolution is that as one system evolves, it necessarily changes the structure of the other system with which it interacts. as he (1993, p. 208) explains with regards to the interaction between species and environments: “fitness landscapes change because the environment changes. and the fitness landscape of one species changes because the other species that form its niche themselves adapt on their own fitness landscapes. bat and frog, predator and prey, coevolve. each adaptive move by the bats deforms the landscape of the frogs.” complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 100 university of bamberg press of course, these principles may only apply metaphorically to english school theory – after all, international society is a social construct and not a biological system. yet, as noted above, international society has evolved through processes quite familiar to co-evolutionary analysis, with normative developments within non-western regional systems changing and evolving based on the interaction with the european society of states and with the environment in which they are embedded. while the idea of co-evolution is, in principle, compatible with and holds significant analytical potential for both historical and more contemporary english school scholarship, there are some limits to the incorporation of cas into its research programme. for example, whereas some cas scholarship starts from the assumptions that social order is not always possible, nor is it necessarily desirable (reed and harvey 1994, p. 390), english school theorists’ principal concern from the outset has been to show that order under conditions of international anarchy is possible. this is not to say that the differences cannot be overcome – in fact, it would be interesting to see how the dissipative cas paradigm and its emphasis on periodic chaos and radical transformation can be reconciled with the school’s account of order – but there are methodological issues that need addressing. the above overview of systems thinking in english school theory is necessarily crude and simplified, but it hopefully shows the affinities between its approach and complexity theory. the next section seeks to harvest some of those synergies by looking at how claims about fragmentation and non-linear processes of order making can be utilised to further advance the english school’s current research programme on international institutions. advancing institutional theorising in the english school: from linear relationships to complexity and fragmentation during the last fifteen years or so there has been a conscious move within the english school towards an emphasis on institutions as offering the most fruitful conceptual lens through which the character of order in the international society can be understood. significant progress in identifying and organising the different types of institutions that structure international social life has been made, but the school is still a long way from a fully developed institutional theory of international order. looking to insights from complexity thinking could be one way of driving this project further. in what follows, i will briefly introduce the main conceptual distinctions and debates relating to institutional theorising in the english school and, subsequently, sketch out elements of a complexity inspired research agenda that focuses on patterns of institutional fragmentation and interconnections. taxonomies and debates building on hedley bull’s (1977) idea that order in international society is maintained by a number of fundamental institutions, generations of english school writers have sought to elaborate and specify the nature, identity and function of the underlying institutional elements of international social systems. though this has generated quite an extensive, and increasingly specialised, body of literature, it seems safe to say that barry buzan’s 2004 book from international to world society? has become the main touchstone for institutional debates within the english school. buzan carefully discusses the different, though frequently overlapping understandings identified by ir theorists to distil from them two different types of international institutions. the first, which he labels “primary institutions”3, are ‘relatively fundamental and durable practices, that are rather evolved more than designed (…) and are constitutive of actors and their patterns of legitimate activity in relation to each other’ (2004, p. 167). these are the institutions referred to in english school theory. as historically evolved normative ordering structures, they arise from interaction among actors and we know them ‘by their shared and principled understandings of desirable and acceptable forms of social behaviour’ (kratochwil & ruggie, 1986, p. 764). there exists no definitive list of primary institutions. bull (1977) listed five: international law, the balance of power, great power management, war, and diplomacy. other accounts have introduced sovereignty (jackson, 2003; wight, 1977), 3 the terms ‘primary institutions’ and ‘fundamental institutions’ are frequently used interchangeably in the literature. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 101 university of bamberg press international trade (holsti 2004), and more recently, environmental stewardship (falkner & buzan, 2019) as potential candidates. primary institutions, buzan goes on to argue, can be distinguished from the kind of (secondary) institutions regime theorists typically refer to – that is, “an organisation or establishment founded for a specific purpose” such as the united nations (un), the world trade organization (wto), international courts, or regional international organisations such as the european union (eu) (buzan 2004, p. 164). the distinction between the two types of institutions hinges primarily on the function they perform: the deeper layer of primary institutions he identifies is primarily associated with constitutive tasks such as establishing actors’ roles and identities, while the more immediate, secondary level institutions are sites of regulatory practices and procedure. moreover, whereas secondary institutions are associated with formalised governance arrangements, primary institutions are associated with more informal social practices. the distinction between primary and secondary institutions has generated quite an extensive, and increasingly specialised, stream of research (friedner parrat, 2017; schouenborg, 2011; spandler, 2015). discussions have primarily revolved around how primary institutions can be identified and what functions they perform. more recently, however, english school scholars have begun to question the relationship between the different types of institutions (navari & knudsen, 2018; spandler, 2015). the main issue at stake here has been change in international society, and in particular the question of the relationship between primary and secondary institutions in this process. traditionally, references to the concept of primary/secondary institutions have shared a top-down vision of change. that is, they tend to work on the assumption that primary institutions represent some more or less cohesive sets of norms, values and principles that manifest themselves in a number of corresponding formal rules and regimes. although buzan’s original account seems to suggest that there is no one-way causal relationship between primary and secondary institutions, secondary institutions are essentially presented as ‘rule-takers’ from primary institutions, with a role that is focused almost exclusively on operationalising the principles and associated normative perspective of primary institutions. this in line with earlier constructivist scholarship, which posited a hierarchical, linear relationship in which constitutional structures defining criteria for membership and legitimate state action in international society constitute primary institutions, which in turn constitute the more issue specific secondary institutions (reus-smit, 1997). the idea of embedding issue-specific regimes in overarching institutional arrangements that constitute international society’s deeper-seated structures is not new. it has been identified by liberal regime scholars as one type of mechanisms through which institutions are linked (e.g. young 1996: 23). that said, conceptual detail about these deep institutional structures and how exactly they relate to regimes is typically lacking. using the english school’s account of primary institutions can help to significantly improve the understanding of that mechanism, in particular in the context of climate governance. above all, it could help to ground the “fragmented global [climate] governance architecture” (biermann et al 2009) in an underlying account of international society’s shared norms and values. buzan and falkner (2019) have made some important progress here by conceptualising environmental stewardship as a primary institution. this move enables them to tie together functionally fragmented and geographically dispersed institutional arrangements, by relating them to a “distinctive set global values” that international society as a whole holds about global environmentalism (buzan and falkner 2019: 132). here, the english school offers real added-value to global environmental politics theory. introducing complexity: law-governed emergence coming at the problem of change in international society from the bottom-up, rather than topdown, a new wave of english school scholarship has proposed a reticular relationship between primary and secondary institutions (navari & knudsen, 2018; spandler, 2015). the idea here is that primary institutions are not necessarily analytically prior to secondary institutions, and that the policies and operations of international organisations and regimes have reciprocal implications for the evolution of complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 102 university of bamberg press international society’s deeper-seated institutional ordering practices (see figure 1). building on the theoretical claim that actors are entangled in processes of constitution and regulation that always occur in parallel, this research highlights how the politics of international organisations such as the international criminal court or the un security council frequently affect practices surrounding the balance of power and sovereignty, respectively. this argument links to complex systems thinking because it recognises cyclical paths, feedback loops and non-linearity in the relationship between different types of institutions. figure 1 the reciprocal relationship between international society’s three institutional structures drawing more explicitly on complexity thinking could push this research agenda even further. in particular, it offers a way to consider how different secondary (or primary) institutions interact with each other, and how those interactions may produce effects that cannot be attributed to any specific institution. this agenda and opportunity springs from one of the core insights of complexity research: the idea that a system is not simply the sum of its parts, but is produced, altered and transformed by the interaction of its constituent parts. the conceptual challenge for english school theorists, then, is to make sense of an institutional structure whose nature, pattern and transformation cannot be explained in terms constitutional meta principles. the place to start, i suggest, is with the notion of emergence. “emergence” sits at the heart of the complexity research programme and there is a vast body of literature dedicated to spelling out both its conceptual properties and its application in various disciplines. it is beyond the scope of this article to offer an exhaustive review of this literature here; instead, i will briefly focus on what i believe is the most useful approach for english school theorists: “law-driven emergence.” at the most basic level, the term “emergence” denotes “the arising of novel and coherent structures, patterns and properties during the process of self-organization in complex systems” (goldstein 1999, p. 49). crucially, for many complexity theorists, the process of “arising” is, at least in theory, not completely random, but unfolds according to a set of fundamental “laws” that set the boundary conditions within which new structures, patterns, and properties can arise. holland (1998, p. 2), for example, uses the analogy of a chess game to show how “a small set of rules or laws can generate surprising complexity”, suggesting that complexity in biological organisms is also governed by a small set of basic rules. in the same vein, kauffmann (2000) has identified a salient organising force in the cosmos, which he calls “the fourth law of thermodynamics.” what unifies these understandings of emergence, as corning (2002, p. 25) nicely summarises, it the belief that “some underlying inherent force, agency, tendency, or ‘law’ (…) is said to determine the course of evolutionary processes.” primary institutions e.g., international law, balance of power, diplomacy secondary institutions e.g., eu, un convention of the law of the sea constitutional structures metavalues defining legitimate statehood and rightful state action complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 103 university of bamberg press this perspective on emergence fits well with the english school’s institutional account because of the way in which it allows for conceiving of institutional structures in terms of complex social constructs that are, at least to some extent, the result of unique, self-contained interactions among their components, while preserving some of the hierarchy and order assumptions that have been so central to their cognitive concerns. in this sense, the notion of “law-drive emergency” offers a via media between constitutive hierarchy and emergent institutional complexity. on this view, and staying true to the critical realist claim of constitutive effects, the rules and principles associated with higher order institutional levels constitute the background conditions in which lower order institutional levels emerge and evolve, while recognising that the actual process of emergence and evolution is a result of complex and often unpredictable interactions among actors and institutions. some complexity theorists may rightly note that this approach runs the risk of treating emergence as an epiphenomenon. but it is, i believe, a necessary price to pay for incorporating complexity into the english school’s theoretical structure. if we follow the idea of “law-governed emergence,” then the distinct character of international society, and the institutional processes and mechanisms by which it evolves, cannot be explained exclusively by resorting to a single source, principle or, in our case, institution. instead, the dynamism of self-reproducing structures calls for putting the spotlight on the interplay between institutions and the normative consequences and social practices that follow from it. put simply, if we take the claim that social system structures reproduce themselves from within themselves (autopesis) seriously (teubner, 1992), then exploring dynamics and interconnections within and among institutions is key for theorising normative ordering dynamics within international society. in what follows, i will focus on two specific institutional consequences of emergence: fragmentation and clustering. fragmentation in primary institutions as intuited earlier, the aspect of complexity that has been most relevant to the study of international institutions is the tendency towards the fragmentation of secondary institutions. this tendency includes the segmentation of spheres of regulative activities, each with its own role structure, norms, and obligations (e.g. alter & meunier, 2009). there is no reason to assume that primary institutions are not subject to similar fragmentation dynamics. hints about how this fragmentation occurs at the level of primary institutions can be found in the regionalisation debate international society alluded to earlier. looking at the institution of sovereignty, for example, costa-buranelli (2015, pp. 499–500) shows that although it exists at the global level, “the increasing regionalization of world politics and the consequent multiplication of interpretations of global institutions” have weakened the shared normative understanding underpinning it. this means that sovereignty is not necessarily a globally integrated primary institution that mandates a universally recognised course of action. instead, sovereignty is subject to constant reframing and different interpretations, which often results in diverging sovereignty practices that are distinct to certain actors (or groups thereof). exploring fragmentation dynamics can be significant, because they are likely to produce much more fine-grained, nuanced understandings of the nature and internal structure of the many primary institutions. the easiest way to explain what i mean by this it to look briefly at the structural development of one of international society’s key institutions, international law, where efforts at normative integration via notions of jus cogens and erga omnes co-occur with an increasing degree of normative fragmentation and competition. legal commentators have understood for decades that the modern international law is marked by uneven normative development, a process that not only involves the increasing diversification of issue areas, but also the admission of new constitutive members and participants. the international law commission paid tribute to this process by initiating a study group on the theme of ‘fragmentation of international law’. marti koskenniemi (2006, p. 68), who finalised the landmark report, summarises the increasing disintegration of international law as follows: “what once appeared to be governed by “general international law” has become the field of operation for such specialist systems as “trade law”, “human rights law”, complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 104 university of bamberg press “environmental law”, “law of the sea”, “european law” and even such exotic and highly specialized knowledge as “investment law” or “international refugee law” etc. each possessing their own principles and institutions. the problem, as lawyers have seen it, is that such specialized law-making and institution-building tends to take place with relative ignorance of legislative and institutional activities in the adjoining fields and of the general principles and practices of international law.” here, we can see how “law-governed emergence” operates in practice. that is, although those free-standing regimes (secondary institutions) are in principle bound by the general rules of the primary institution in which they are embedded (international law), they have their own binding dispute settlement mechanism, essentially making and applying rules according to their own procedural norms. as koskenniemi (2006, p. 4) observed, they are “interrelated wholes of primary and secondary rules, sometimes also referred to as ‘systems’ or ‘subsystems’ of rules that cover some particular problem differently from the way it would be covered under general law.” the consequences for linear patterns of order making are exemplified by some of the inconsistent judgments and conflicting jurisprudence issued by different international courts, and the way in which this tends to undermine the value of precedent as the main “guarantor of certainty and equality of treatment” (guillaume, 2011, p. 6). for example, in judging on whether the government of serbia and montenegro was responsible for act of violence conducted by irregular militias, the international criminal tribunal for the former yugoslavia revisited the international court of justice’s (icj) ruling regarding state responsibility in the nicaragua case, triggering a dispute among the two courts about the appropriate test for determining when states are responsible for the use of irregular force. the limitations those non-linear processes of law making impose on article 59 of the icj statute for developing international public law in a coherent and systematic fashion are well known (kratochwil, 2014, pp. 93–97). considering how and why primary institutions have become fragmented is one area for future research. the first task here is to conceptualise degrees of fragmentation. the regime complexity literature offers good points of departure. for example, biermann et al (2009: 19-21) have distinguished between synergistic, cooperative, and conflictive types of fragmentation. building on this typology could also help to get to grips with fragmentation dynamics in primary institutions. the second task is empirical. english school scholars need to develop tools for comparing fragmentations in different primary institutions. this would, for example, require network analysis to map out the network structure of primary institutions, by locating nodes and identifying links among different institutional arrangements. such a method may sit uneasy with the english school’s traditional methodological focus on historical, sociological, and normative inquiry, but it offers an interesting tool for expanding the scope of the school’s research agenda. clustering primary institutions another consequence associated with emergence is institutional clustering. this has to do with the linkages primary institutions develop in a complex international society. because norms and beliefs do not always line up so as to form a consistent overarching normative structure, there is good reason to assume that the interplay between primary institutions is more often than not characterised by resistance, contestation, and alteration. williams (2015, p. 122) briefly points to the inherent tensions among primary institutions, by noting that “their constitutive roles introduce many of the tensions, dilemmas, contradictions and frictions of international relations, when competing and irreconcilable constitutive institutions pull us in different directions in the face of events and circumstances which the same institutions grant different levels of types of significance.” in contemporary international society, conflicts and tensions between different primary institutions are perhaps most evident in our debates about humanitarian interventions, in which cosmopolitan moral appeals to human rights norms frequently rub-up against formalised westphalian rules of non-intervention. nato’s intervention in kosovo in the 1990s, which was deemed “illegal but justified” by the international legal community, perhaps stands as the clearest example of the way in complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 105 university of bamberg press which a fundamental institutional ordering practice such as sovereignty can be legitimately contested under appeal to human rights (independent international commission on kosovo, 2000, p. 4). the debate about humanitarian intervention – and the normative tensions between sovereignty and human rights norms that come with it – is a good example. typically, the stalemate in those debates has been broken in favour of one institution, with some scholars privileging sovereignty and territorial integrity over the protection of individuals (jackson, 2000), while others have advocated for the occasional suspension of non-intervention claims in the face of severe humanitarian emergencies (wheeler, 2000). privileging one primary institution over another when confronted with conflict is certainly an attractive option, not at least because it has helped to formulate a more or less coherent vision of order in international society. but it comes at the expense of considering the entanglements of, and relationships between, different primary institutions. a different way to approach this is through the notion of institutional clustering. building on oran young’s (1996: 5-6) typology of institutional linkages, i define clustering as the result of combining two or more, analytically differentiable primary institutions into an institutional package. clustering of primary institutions occurs when members of international society join together analytically distinct primary institutions for the purpose of resolving conflicts among fundamental social ordering principles and practices. the emergence of the responsibility to protect (r2p) doctrine set out in 2005 world summit outcome, which ascribes a responsibility to sovereign states to protect their populations from genocide, war crimes, crimes against humanity and ethnic cleansing, is case in point. rather than having to choose between either the protection of human rights or upholding sovereignty when faced with these crimes, r2p indicates international society’s effort to combine the two institutional practices – by making sovereignty conditional upon the ability and willingness of a state to protect its populations from severe human rights abuses. we may observe similar clustering dynamics between sovereignty and environmental stewardship in the future, given that the latter calls for “a global system perspective and coordinated rules of the game that transcend borders” (buzan and falkner 2019: 149). conclusion the concepts and language of complexity theory speak directly to a number of concerns that are central to the widening and broadening of the interaction in social, economic, and political dynamics, scales and levels that characterise contemporary international relations. this article has sought to lay the groundwork for a systematic engagement between the english school of ir and complexity theory and, in particular, for an english school approach to complex systems thinking. the emphasis english school theory places on evolutionary change, the reflexivity of systems and boundaries, and relational social processes at the global level enables complex systems thinking to flourish. harvesting those synergies and using ideas and concepts from complexity could not only open up new space for debate about the current transformation processes and dynamics that shape global international society, but also push english school scholars to re-think some of their more established models and assumptions. focusing on one specific strand of current english school theorising, the article has introduced the notion of fragmentation dynamics and applied it to the study of international institutions. if primary institutions are to play a significant role in exploring the normative framework of international society, then we need to improve our understanding of their different institutional compositions and internal structures. here, the idea of fragmentation dynamics may open some critical analytical space. consideration of decentralisation and localisation offers an alternative conceptualisation of the nature and structure of international society’s fundamental ordering practices that stresses the inescapability of normative tensions and contradictions among them. once we recognise bottom-up processes of emergence, fragmentation, and decentralised interaction dynamics become part and parcel of institutional english school thinking. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 106 university of bamberg press there are some more general implications of this for english school’s intellectual structure that are potentially radical. while the conceptual developments advanced in the second half of this article have primarily been social-structural in the sense that they have focused on the formal dynamics between international society’s institutional building-blocks, they also have normative consequences. by allowing for fragmentation and competition among primary institutions, complexity presents a fundamental challenge to those who presume some sort of normative consensus about the ethical content and moral purpose of international society’s basic institutional ordering practices. if we accept the argument that even primary institutions evolve in non-linear ways and do not always line up coherently, then english school scholars need to address the way in which the values and moral principles inscribed in institutional practice emerge and transform. this connects complexity thinking to recent pluralist english school scholarship that treats ethical diversity and value conflict as both an empirical reality of international relations and a normatively desirable feature of the human condition (e.g. williams 2015). references abbott, k. w. (2012). the transnational regime complex for climate change. environment and planning c: government and policy, 30(4), 571–590. albert, m., cederman, l.-e., & wendt, a. (eds.). (2010). new systems theories of world politics. basingstoke: palgrave macmillan. alter, k. j., & meunier, s. (2009). the politics of international regime complexity. perspectives on politics, 7(1), 13–24. arthur, w. b. (1994). increasing returns and path dependence in the economy. university of michigan press. axelrod, r. m. (1997). the complexity of cooperation: agent-based models of competition and collaboration. princeton, n.j: princeton university press. betts, a. (2010). the refugee regime complex. refugee survey quarterly, 29(1), 12–37. betts, alexander. (2009). institutional proliferation and the global refugee regime. perspectives on politics, 7(1), 53–58. biermann, f., pattberg, p., asselt h., & zelli f. (2009). the fragmentation of global governance architectures: a framework for analysis. global envrionemtnal policy, 9(4), 14-39. bousquet, a., & curtis, s. (2011). beyond models and metaphors: complexity theory, systems thinking and international relations. cambridge review of international affairs, 24(1), 43–62. bull, h. (1966). international theory: the case for a classical approach. world politics, 18(3), 361–377. bull, h. (1977). the anarchical society: a study of order in world politics. london: palgrave macmillan. bull, h. & watson, a. (1984). the expansion of international society. oxford: oxford university press. buzan, b. (1993). from international system to international society: structural realism and regime theory in the english school. international organization, 47(3), 327–352. buzan, b. (2004). from international to world society?: english school theory and the social structure of globalisation. buzan, b. (2014). an introduction to the english school of international relations: the societal approach. cambridge: polity press. buzan, b., & little, r. (2000). international systems in world history: remaking the study of international relations. oxford ; new york: oxford university press. clark, i. (2009). legitimacy in international society (reprinted). oxford: oxford university press. corning, p. a. (2002). the re-emergence of emergence: a venerable concept in search of a theory. complexity, 7(6), 18–30 costa-buranelli. (2015). ‘do you know what i mean?’ ‘not exactly’: english school, global international society and the polysemy of institutions. global discourse, 5(3), 499–514. falkner, r., & buzan, b. (2019). the emergence of environmental stewardship as a primary institution of global international society. european journal of international relations, 25(1), 131-155. friedner parrat, c. (2017). on the evolution of primary institutions of international society. international studies quarterly, 61(3), 623–630. gehring, t., faude, b. (2014. a theory of emerging order within institutional complexes: how competition among regulatory international institutions leads to institutional adaptation and division of labor. review of international organisatioz,, 9, 471–498 . goldstein, j. (1999). emergence as a construct: history and issues. emergence, 1(1), 49–72. guillaume, g. (2011). the use of precedent by international judges and arbitrators. journal of international dispute settlement, 2(1), 5–23. helfer, l. r. (2009). regime shifting in the international intellectual property system. perspectives on politics, 7(1), 39–44. hoffmann, m. j., & riley, j. (2002). the science of political science: linearity or complexity in designing social inquiry. new political science, 24(2), 303–320. holland, j. (1995). hidden order: how adaptation builds complexity. addison-wesley. holland, j. (1998). emergence: from chaos to order. perseus books. holsti, k. j. (2004). taming the sovereigns: institutional change in international politics. hurrell, a. (2001). keeping history, law and political philosophy firmly within the english school. review of international studies, 27(03). complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 107 university of bamberg press independent international commission on kosovo (ed.). (2000). the kosovo report: conflict, international response, lessons learned. oxford: oxford university press. jackson, r. h. (2003). the global covenant: human conduct in a world of states. oxford: oxford university press. jervis, r. (1997). system effects: complexity in political and social life. princeton, nj: princeton univ. press. johnston, a., & trebilcock, m. (2013). fragmentation in international trade law: insights from the global investment regime. world trade review, 12(4), 621-652. doi:10.1017/s1474745613000128 kaplan, m. (1957). kaplan, m. (1957). system and process in international relations. new york: wiley. kauffman, s. a. (1993). the origins of order: self-organization and selection in evolution. oxford university press. kauffman, s. a. (2000). investigations. oxford university press. kavalski, e. (2007). the fifth debate and the emergence of complex international relations theory: notes on the application of complexity theory to the study of international life. cambridge review of international affairs, 20(3), 435–454. kavalski, e. (2011). from the cold war to global warming: observing complexity in ir. political studies review, 9(1), 1-12. keene, e. (2002). beyond the anarchical society: grotius, colonialism and order in world politics. cambridge: cambridge university press. koskenniemi, m. (2006). fragmentation of international law. difficulties arising from the diversification and expansion of international. un document a/cn.4/l.682. kratochwil, f. (2014). the status of law in world society: meditations on the role and rule of law. kratochwil, f., & ruggie, j. g. (1986). international organization: a state of the art on an art of the state. international organization, 40(4), 753–775. lewin, r. (1992). complexity: life at the edge of chaos. phoenix. little, r. (1995). neorealism and the english school:: a methodological, ontological and theoretical reassessment. european journal of international relations, 1(1), 9–34. morin, j. (1990). introduction aa la complexité. paris: le seuil. navari, c., & knudsen, t. (eds.). (2018). international organization in the anarchical society: the institutional structure of world politics. basingstoke: palgrave macmillan. orsini, a., le preste p., haas p., brosig m., pattberg p, widerberg o., gomez-mera l., morin j., harrison n, geyer r. & chandler d. (2019): forum: complex systems and international governance. international studies review. reed, m., & harvey, d. l. (1992). the new science and the old: complexity and realism in the social sciences. journal for the theory of social behaviour, 22(4), 353–380. reed, m. & harvey, d.l. (1994). the evolution of dissipative social systems. journal of social and evolutionary systems, 17(4), 371–411. reus-smit, c. (1997). the constitutional structure of international society and the nature of fundamental institutions. international organization, 51(4), 555–589. reus-smit, c. (1999). the moral purpose of the state: culture, social identity, and institutional rationality in international relations. princeton, nj: princeton univ. press. reus-smit, christian. (2009). constructivism and the english school. in c. navari (ed.), theorising international society: english school methods (pp. 58–77). basingstoke: palgrave macmillan. rosenau, j. n. (1990). turbulence in world politics: a theory of change and continuity. princeton, n.j: princeton university press. schouenborg, l. (2011). a new institutionalism? the english school as international sociological theory. international relations, 25(1), 26–44. schouenborg, l. (2012). the scandinavian international society: primary institutions and binding forces, 1815-2010. new york: routledge. spandler, k. (2015). the political international society: change in primary and secondary institutions. review of international studies, 41(3), 601–622. stivachtis. (2015). interrogating regional international societies, questioning the global international society. global discourse, 5(3), 327–340. m. (2011). regional international society in a post‐enlargement europe. journal of european integration, 33(2), 101–116. teubner, g. (1992). law as an autopoetic system. florence: the european university institute press. thomas, d. c. (2017). beyond identity: membership norms and regional organization. european journal of international relations, 23(1), 217–240. walby, s. (2007). complexity theory, systems theory, and multiple intersecting social inequalities. philosophy of the social sciences, 37(4), 449–470. waldrop, m. m. (1992). complexity: the emerging science at the edge of order and chaos. touchstone. waltz, k. (1979). theory of international politics. new york: mcgraw hill. watson, a. (1992). the evolution of international society: a comparative historical analysis. london ; new york: routledge. weinert, m. s. (2017). grounding world society: spatiality, cultural heritage, and our world as shared geographies. review of international studies, 43(3), 409–429. wendt, a. (1992). anarchy is what states make of it: the social construction of power politics. international organization, 46(2), 391–425. wendt, a. (1999). social theory of international politics. wheeler, n. j. (2000). saving strangers: humanitarian intervention in international society (reprint). oxford: oxford university press. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 108 university of bamberg press wight, m. (1977). systems of states. leicester: leicester university press. williams, j. (2005). pluralism, solidarism and the emergence of world society in english school theory. international relations, 19(1), 19–38. williams, j. (2015). ethics, diversity, and world politics: saving pluralism from itself? (first edition). oxford, united kingdom: oxford university press. young, o. (1968). a systemic approach to international politics. princeton, nj: center of international studies woodrow wilson school of public and international affairs. young, o. (1996). instituional linkages in international society: polar perspectives. global governance, 2(1), 1-23. zürn, m., & faude, b. (2013). on fragmentation, differentiation, and coordination. global environmental politics 13(3), 119130. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 94-108 doi: http://dx.doi.org/10.20377/cgn-105 the emergence of metropolitan governance: a coevolutionary analysis of the life-and-death cycles of metropolitan governance in the amsterdam metropolitan region complexity, governance & networks (2014) 57–78 57 doi: 10.7564/14-cgn13 the emergence of metropolitan governance: a coevolutionary analysis of the life-and-death cycles of metropolitan governance in the amsterdam metropolitan region. danny schippera,*, gerrits, l. m.b a erasmus university rotterdam e-mail: schipper@fsw.eur.nl b university of bamberg e-mail: lasse.gerrits@uni-bamberg.de this article develops an explanation for the emergence of metropolitan governance in the amsterdam metropolitan region. a coevolutionary framework is deployed to analyze how the governance system developed between 1982 and 2009. data was gathered from 43 policy documents, over 150 newspaper articles and 14 semi-structured interviews. the analysis focuses on the workings of selection pressures and the adaptive processes. it shows that metropolitan governance evolves in a punctuated fashion when fitness with the environment is lost and when the governance system’s ability to absorb selection pressures has eroded. keywords: coevolution, complex systems, exploitation; exploration; metropolitan governance, selection pressures 1. introduction within metropolitan regions, authority is often dispersed over a variety of institutions. this jurisdictional fragmentation is considered to be obstructing the effectiveness, efficiency, equality and democracy of metropolitan regions (e.g. brenner, 2002; savitch & vogel, 2009; tomàs, 2012). consequently, one issue resurfaces continuously in scientific literature on metropolitan governance: what is the most effective governmental structure to deal with the built and socio-economic developments of a particular region? over the past decades attempts to answer this question have led to considerable research and reporting by both academics and practitioners. this resulted in several different approaches to metropolitan governance, which differ for instance in the degree of institutionalization of cooperation (savitch & vogel, 2009; tomàs, 2012). one approach is the institutional design approach or old regionalism (buitelaar, lagendijk, & jacobs, * corresponding author. article_14-13.indd 57 03/03/15 3:04 pm 58 d. schipper et al. / the emergence of metropolitan governance 2007; kantor, 2008; savitch & vogel, 2009). this approach focuses on structural reform to create metropolitan governments that cover the functional territory of the metropolitan region in a single political unit (tomàs, 2012). the search for the perfect institutional design has also been dominant in the netherlands (faludi, 1994). but this holy grail proved elusive (tetteroo, 2003), as is evidenced by the 85 policy documents that were written on this theme since 1965 (witte, 2010), and the multiple failed attempts to reorganize governmental organization for a given tempo-spatial development. in contrast to the institutional design approach, the network approach or new regionalism has often been debated as an alternative approach to metropolitan governance (savitch & vogel, 2009; van der bol, 2010; heinelt & zimmermann, 2011). the interdependencies of actors is a central notion in the network approach (koppenjan & klijn, 2004). according to this approach cooperation between municipalities is needed to deal with the complexity of metropolitan policy problems, as these problems can’t be solved in isolation. so collective decision-making should be increasingly geared towards collaboration between governmental levels and joint action beyond the boundaries of each of the existing governmental bodies (brenner, 2002, 2003; innes & booher, 2003; salet, 2003). moreover, this approach calls for a new pro-active metropolitan planning philosophy, where both public and private actors meet in flexible arrangements to deal with the spatial-economic development of the region (tomàs, 2012). yet, the realization of joined action beyond the boundaries of the existing institutions has proven to be very hard because of differing perceptions, rules and objectives, a lack of decisiveness and the considerable competition between existing institutions and new temporal arrangements (hajer, van tatenhove, & laurent, 2004; koppenjan & klijn, 2004; van der bol, 2010). while these different theoretical approaches to metropolitan governance have their merits, they don’t provide an explanation for the highly different metropolitan governance arrangements that have emerged, even within countries (heinelt & zimmermann, 2011, p. 1175–1176). this article follows e.g. gerrits, l. m. (2011); gual and norgaard (2010); kerr (2002); lewin and volberda (1999); norgaard (1984, 1994); room (2011); volberda and lewin (2003) and argues that a coevolutionary analysis of metropolitan governance is necessary to understand the emergence of metropolitan governance systems as a complex process of reciprocal selection. while most approaches to metropolitan governance focus on structures and (institutional) rules at a given single point in time, a coevolutionary analysis takes on a longitudinal approach to explain the evolutionary process that these coordination mechanisms go through (cf. feiock, 2008; teisman et al., 2009). metropolitan governance is regarded as a dynamic process, hence it is necessary to describe the changes over time and to explain the temporal process through which change occurs (kerr, 2002). the research question of this article is: how can the emergence of metropolitan governance be explained from a coevolutionary perspective? to answer this question, the origin and development of the amsterdam metropolitan area (metropoolregio amsterdam, mra) has been studied. arguably, the mra is the most developed form of metropolitan governance in the netherlands (haran, 2010; janssen-jansen, 2011; article_14-13.indd 58 03/03/15 3:04 pm d. schipper et al. / the emergence of metropolitan governance 59 levelt & janssenjansen, 2013). its core is a voluntarily collaboration between 36 municipalities, two provinces and one city-regional authority (see figure 1). the empirical part of this research is a single in-depth case study spanning the period between 1982 and 2009. the first research step concerned an extensive longitudinal reconstruction of events in the political and administrative realm of the mra and its predecessors in the amsterdam region. this was done through an analysis of over 150 articles from national newspapers covering this topic between 1982 and 2009. the analysis focused on how events came about, which effects they generated and how they lead up to new events. this time-series provided the backbone for the in-depth analysis. the second step concerned a content analysis of the exact properties of the governance structure and processes during the time-series studied through the analysis of 43 policy papers on the mra and its predecessors. in addition, 14 semi-structured in-depth interviews were carried out with senior civil servants who were or still are involved with metropolitan governance in the amsterdam region in key positions. data from the interviews was used to reconstruct how variation and selection of policy options came about and what the motives were for making certain decisions. the interviews were recorded, transcribed and coded. a full list of documents and respondents is available from the authors. figure 1. the amsterdam metropolitan area and its location in the netherlands article_14-13.indd 59 03/03/15 3:04 pm 60 d. schipper et al. / the emergence of metropolitan governance the article is structured as follows. the following section discusses the theoretical approach of this research, where the use of coevolution as an analytical concept is explained and a coevolutionary framework developed. empirical data is presented in the form of a condensed case description in section 3. the analysis is presented in section 4. section 5 discusses the results and their implications. 2. a coevolutionary framework coevolution was originally suggested by ehrlich and raven in 1964 to describe the simultaneous evolution of species through mutual selection pressures. as a conceptual development of explanations of evolutionary change, it has found its way into other domains, ranging from organizational theories (e.g. hodgson, 2003) to urban planning (e.g. marshall, 2009). an important contribution comes from norgaard (1984, 1994, 1995, 2010), who understood that there is a reciprocal relationship between the activities of humans using physical systems for their material progress, whereupon physical systems change to the extent that humans have to respond to those changes subsequently. over time, this creates a pattern of feedback loops between two (or more) mutually dependent systems. norgaard’s work demonstrated that coevolution could have explanatory power outside biology in analyzing human decision-making processes, which has potential for understanding the coming about of governance processes and systems (e.g. sementelli, 2007). early attempts at (co-) evolutionary explanations can also be found in the realm of governance, e.g. corning (1983), daneke (1996), kaufman (1991), and wilson (1975). kaufman’s account is perhaps the most well-known. he acknowledges that actors in the public domain overestimate the influence they have on the outcomes of processes. those outcomes, kaufman says, are the result of a partly random coming together of several factors. since kaufman there have been few works within this realm were evolutionary mechanisms were applied to analyze governance processes. we believe that coevolution holds explanatory power for governance processes and we will us this as our point of departure (cf. gerrits, l. m, 2008, 2011). the concept of coevolution features a subset of mechanisms that explain how and why systems, such as metropolitan governance, change or not over longer periods of time. the central mechanisms are variation, selection and retention (cf. campbell, 1969), which form a continuous cycle. variation involves the exploration of new forms of governance, of which some forms are then selected according to their fit with the environment (gerrits, l. m, 2008). public decision-making is the chief instrument for selection, even though it is acknowledged that not every decision leads to concrete outcomes and will always be intentional (dopfer, 2005; gerrits, l. m, 2012). the coevolutionary perspective on governance systems implies that the creation of variety and selection of said variety by taking decisions, is considered to be a part of the same feedback loop. in other words, variation and selection are not separate events but intertwined in a complex way and shape the freedom to act (foster & hölz, 2004). retention serves to counteract the self-reinforcing loop between variation and selection as it involves forces that maintain certain forms and practices of governance (van de ven & poole, 1995) article_14-13.indd 60 03/03/15 3:04 pm d. schipper et al. / the emergence of metropolitan governance 61 coevolution in this research is defined as the mutual adjustment through reciprocal selection between the metropolitan governance system and the internal and external systems it interacts with. these interactions, in the shape of actions and responses between the systems, constitute feedback loops leading to changing or stable situations (gerrits, l. m, 2012, p. 78–79). there are multiple feedback loops that operate simultaneously. however coevolution thrives on positive feedback loops as these lead to changes in systems. feedback therefore becomes selection pressure as a response to an incentive leading to change (gerrits, l. m, 2008). this poses two questions: what is being selected, and how is it being selected? in other words, how should coevolution be operationalized to understand the emergence of metropolitan governance? coevolving systems are constantly subject to selection pressure. these selection pressures present a complex puzzle to the actors in terms of content, process and structure of the metropolitan governance system. as mentioned above, the selection process is steered by the actors involved through variation, selection and retention. feedback loops carry information for actors which needs to be assessed, ordered and acted upon by selecting the appropriate strategic response (kerr, 2002). to cope with the information from these selection pressures actors may deploy an existing combination of routines, goals and ideas. at the same time, it is necessary for them to evaluate those existing combinations against changing conditions in order to achieve renewed fitness with the environment. in this search for renewed fitness with the environment, a temporal situation is created where diversity in routines, ideas and goals is generated, evaluated and altered. according to van den bergh and kallis (2009), variety is in fact a combination of existing ideas from which novelty could emerge, i.e. the intended or accidental combination of ideas to create new possible pathways towards the future. as duit and galaz (2008) mention, both exploitation (the capacity to benefit from existing forms of collective action) and exploration (the capacity to nurture learning and experimentation with new forms of governance) are essential for the adaptive capacity of metropolitan governance systems. the adaptive moves will eventually lead to the selection of the immediate future state of the system involved, which is the metropolitan governance system in this case, to reach a temporal equilibrium state. at any point in time a governance system has a number of possible future states. this collection of possible future states describe a particular system state consisting of content (what should be achieved? i.e. policy content), process (how should it be achieved? i.e. degree of institutionalization) and structure (who should achieve it? i.e. organization of the arrangement; number and diversity of participants). the collection of possible system states is defined through path-dependency, chance events, and through the perception of those involved, who have certain ideals and beliefs that shape what is perceived a desirable future of the system (arthur, 1994; david, 1985; greener, 2002; pierson, 2000). as time progresses, a succession of system states allows mapping the changes (or lack thereof) in terms of content, process and structure. in table 1 shows the operationalization of the system states in terms of content, process and structure as used to map the change of the system over time. we will use these items to trace the evolution of metropolitan governance in the amsterdam region in the case description and analysis below. article_14-13.indd 61 03/03/15 3:04 pm 62 d. schipper et al. / the emergence of metropolitan governance 3. three decades of metropolitan governance what follows is a condensed presentation of the data from the case description that has been developed for this study. the case has been subdivided in six time periods. each period starts with a brief description of the selection pressures on the governance system. the attempts to adapt to these pressures by changing the configuration of the system in terms of content, process and structure are given in tables 2 to 7. the description starts in 1982 and ends in 2009, when the first mra conference was held. since 1969 the mayors and aldermen of 25 local authorities in the amsterdam region met within the informal agglomeration consultation amsterdam (iao) to discuss regional matters on a voluntary and informal basis. usually these meetings were in a response to new policies of the national government or the province of north-holland. actual collaboration between local authorities was lacking (van der veer, 1997). therefore, in 1982 the city of amsterdam presented a detailed plan to form a new regional authority, which had to reduce the growing wealth disparity within the region. the iao, although not pleased with the unilateral action of the city of amsterdam, initiated a study on the most important regional problems. following up on this research six models for collaboration were presented to tackle these regional problems. some of the models were considered quite revolutionary, but were also met with fierce resistance from most of the participating municipalities in the iao. resistance also came from the national government who rejected any governmental reform at that time. it offered an alternative in the shape of a revision of the joint provision act (wet gemeenschappelijke regeling, wgr), which was intended to stimulate and guide the coordination between the large cities and the smaller suburbs in the netherlands. consequently, there was a majority vote from the municipalities in the iao to consolidate the existing collaboration and to allow some minor adjustments in anticipation of the new joint provision act (see table 2). in december 1986 this new form of governance was founded on a policy agreement and renamed in to regional consultation amsterdam (regionaal overleg amsterdam, roa). the ongoing globalization and upcoming european integration at the end of the 1980’s, caused the main focus of the metropolitan governance to shift from national table 1 the variables used to map the configuration of the metropolitan governance systems over time in terms of content, structure and process. selection mechanisms content ● goals and ambitions ● the tasks that are (formally) assigned to the governance system. structure ● the number and diversity of participants ● organization of the arrangement process ● decision making rules ● type of arrangement (institutionalized or voluntary) article_14-13.indd 62 03/03/15 3:04 pm d. schipper et al. / the emergence of metropolitan governance 63 spatial and wealth redistribution towards the international economic position of the regions within a competitive market. the joint provision act lacked the means to act on the spatial-economic problems of the regions, because of its voluntary character. this conclusion was shared by a number of committees who had studied the socio-economic position of the dutch city regions, the most prominent one being the montijn committee. montijn concluded that a revision of the institutional structure was needed and it advised on the creation of new governmental structures at the scale of the city region. the national government responded to the advices of the committees with its own policy document (bon-1) in september 1990. in this document the national government announced that it wanted to form seven new regional governmental structures for the largest city regions in the netherlands, including the amsterdam region. with the national government now supporting governmental reform the partners in the amsterdam region couldn’t maintain their informal way of collaborating. again, a research team was asked to search for the binding elements in the region and to formulate a comprehensive plan. as a first step along this path the regional consultation amsterdam was renamed in to regional body amsterdam (regionaal orgaan amsterdam, again: roa) on the 30th of june 1992. this new form of collaboration had some major differences in comparison to the regional consultation amsterdam (see table 3). the process towards a new governmental structure gained momentum after the establishment of the regional body amsterdam. this was particularly due to the developments in the second largest city in the netherlands, rotterdam. the plans of the national government to form a new regional authority were at first met with much reluctance in the rotterdam region. it had only been five years after the abolishment of the rijnmond authority, a public authority directly elected by the people in which the region had unsuccessfully tried to coordinate specific regional matters (bordewijk, 2000). yet, informal table 2 regional consultation amsterdam. regional consultation amsterdam content the aim of the arrangement was to look after the common interest of the region in order to promote a balanced development of the region. in order to do so the arrangement had the following tasks: ● the exchange of information on shared (spatial) problems. ● collective research on regional affairs. ● central point for the announcement of regional activities. ● the initiation of (spatial) projects ● acting as a negotiating partner of the national government. structure the arrangement was formed by 23 municipalities. representatives of these municipalities joined the plenary meetings. the agenda of the meetings were drafted by a commission of up to 14 representatives. in addition there were special meetings between the aldermen of the municipalities. process the collaboration was founded on the basis of a limited governance agreement. decisions were made based on majority vote, however there were no additional rules tied to the decision making process. article_14-13.indd 63 03/03/15 3:04 pm 64 d. schipper et al. / the emergence of metropolitan governance talks about the future of the rotterdam region eventually led to a new consensus among the regional partners on the most important regional issues. however, the plans of the national government forced the region to start thinking on an institutional structure to deal with these regional matters. in an attempt to maintain the initiative the region proposed a metropolitan authority, which would perform local (especially those of rotterdam), regional and national governmental tasks. this new authority had to be able to operate independently from the province of south holland and in order to avoid a conflict of power, the city of rotterdam was to be divided into several smaller independent municipalities. these plans were highly supported by the national government and it announced, in its third policy document (bon-3) of february 1993, a special law to make this new administrative authority possible (buitelaar et al., 2007). however, the ministry of interior rejected the creation of a non-municipal or non-provincial authority, since this would mean that the constitution would have to be adjusted. for this reason, the region agreed on the establishment of a new province on the level of the metropolitan region, the so-called stadsprovincie or city province. with the ongoing developments in the rotterdam region, the national government started to raise the pressure on the amsterdam region to follow the example of rotterdam. hence, in june 1994 the ‘letter of intent city province amsterdam’ was signed by the ministry of interior, city of amsterdam, province of north holland and roa to establish a city province on january 1st, 1998 (table 4). similar to rotterdam, the partners in the amsterdam region agreed to let legislation determine the process towards the city province. this decision was made under pressure of the ministry of interior, who wanted the special law for the city province to be finished early 1995 (van der veer, 1997). soon, dozens of teams of civil servants were working on this special law. in design and pace the process was dominated by the copying of legal texts for the rotterdam region (voigt & van spijker, 2003). meanwhile, the region started to study on how the city province should be accommodated within the existing institutional fabric. it developed three policy documents which were meant to get the local authorities into line with one another. table 3 regional body amsterdam. regional institute amsterdam content the aim of the arrangement was to promote a balanced development of the region and to serve as a first step towards a more far-reaching form of collaboration. the arrangement had planning, steering and coordinating authority over: (a) spatial planning, (b) housing, (c) economic development, (d) transport and infrastructure, and (e) environmental issues. structure the arrangement was formed by 17 municipalities (this reduction was due to the merger of municipalities). it had its own council with representatives of the municipalities, a management committee, and a chairman (the mayor of amsterdam). process the collaboration was founded on the joint provision act. the regional council drafted plans that were binding on the participating municipalities. the management committee could enforce municipalities to implement these plans made by the regional council or if necessary a disputes committee could resolve the dispute. article_14-13.indd 64 03/03/15 3:04 pm d. schipper et al. / the emergence of metropolitan governance 65 hence, controversial subjects, such as authority over the national airport and the seaport of amsterdam, were postponed and consensus between the local municipalities grew (ibid.). because of the considerable time pressure behind the special law, the citizens of amsterdam weren’t consulted. as such, officials didn’t recognize the growing unrest among amsterdam’s citizens about the potential subdivision of the city. early 1995 these officials were far more concerned with the influence of the region on the special law. the production of legal texts had started to dominate the process, pushing the content to the background as well as the freedom for different interests. the amsterdam city council and its officials became worried that a powerful city council would be exchanged for a weak city province, because some of the region’s main assets, like the international airport and the port, would remain under municipal authority. therefore, they severely criticized the (secret) draft versions of the legal texts with which the negotiators of amsterdam returned (van der lans, 2006). eventually the city council even warned the junior minister that the city would reconsider its participation in the city province. they also demanded the ministry of interior to submit to seven terms that were essential to the city. not only were these terms seen as essential to gain support for the city province from the city council and its officials, but also to gain the trust from the citizens of amsterdam. these citizens had collected the 25.000 signatures needed for a citizen initiated referendum on the subdivision of amsterdam, which was to be held in may 1995. the negotiators of amsterdam believed that they could convince the citizens to vote in favor for the subdivision of amsterdam, when held to a prospect of a strong city province. but of the voters a major 92,6 percent rejected the division of the city, against 6,6 percent in favor. this result meant that the city council had to reconsider the subdivision of amsterdam, which was one of the main prerequisites in the process towards the city province. therefore, roa decided to put the formation of the city province on hold for half a year. during this recess, a roa-committee consulted politicians in the region and mayor patijn of amsterdam invited society’s representatives to his residence to voice table 4 regional body amsterdam under the framework law. regional body amsterdam under the framework law content the arrangement served as a step towards the city province. it had a wide range of tasks, of which some had been carried over by the municipalities, province and national government. overall it was a planning, steering and coordinating authority over: (a) spatial planning, (b  housing, (c) economic development, (d) transport and infrastructure, (e) environmental issues, (f) land policy, and (g) healthcare structure the administrative body was formed by 16 municipalities. it had its own council with representatives of the municipalities, a management committee, a chairman (the mayor of amsterdam), and commissions responsible for the statutory tasks. process the regional body amsterdam became an official administrative body under the framework law. municipalities were legally obliged to implement the decisions made by the regional council. article_14-13.indd 65 03/03/15 3:04 pm 66 d. schipper et al. / the emergence of metropolitan governance their opinion on the situation. these meetings and consultations resulted in a renewed consensus on the future direction of the metropolitan governance (van der lans, 2006). participants agreed that the challenges for the region in the upcoming years were of such a magnitude that metropolitan governance and guidance was urgently needed. they also agreed that the new arrangement had to be flexible and that all participants were to have the opportunity to appeal to a higher authority, which could control the process and enforce decisions (ibid.) during the recess the province of north-holland seized the opportunity to reclaim its position and to intervene in regional affairs. at first, this move surprised the roa partners, but they decided that it was better to cooperate in order to find renewed collaboration. after all, there was still a chance that the national government would enforce the city province. with a widely accepted alternative, they might convince the parliament to decide against the city province. so, the provincial governor van kemenade and the mayor of amsterdam jointly visited the municipalities in the region to talk about the future of the metropolitan governance. not only the roa-municipalities were approached by van kemenade and patijn. following the criterion of flexibility, there was no fixed boundary for the new governance system. however, the need for flexibility was hindered by roa which had statutory tasks, a bounded geographical area and which was still a formal regional administrative body. some roa officials actually regarded the new flexible governance form as a competitor and feared that they would lose their position. a two-track policy was adopted in 1998 to solve this tension. one was a formal track making it possible for roa to fulfill its legal obligations. the informal track involved a flexible network to solve the urgent (supra)-regional problems and was named regional collaboration amsterdam (regionale samenwerking amsterdam, rsa). a coordination committee (coco), chaired by patijn and van kemenade, was established to meet the need for synergy (table 5). table 5 regional collaboration amsterdam. regional collaboration amsterdam content the aim of the arrangement was to cooperate in a functional and tailor-made manner in the following areas: ● accessibility ● port and airport development ● the realization of the major housing task and the prevention of segregation. ● the international economic position of the region. ● regional labor market. ● youth support structure it was decided that no new governance structure would be created. these ad hoc arrangements should be able to adjust itself to the given problem. this new flexible approach to metropolitan governance meant that there was no fixed number or composition of participants. the coordination committee was the core administrative body. process collaboration between the municipalities was on a voluntary basis. meetings were informal and focused on cooperation, without any rules to enforce decisions. article_14-13.indd 66 03/03/15 3:04 pm d. schipper et al. / the emergence of metropolitan governance 67 during the same period the ministry of housing, spatial planning & the environment (vrom) and the ministry of transport (v&w) were preparing new national policy documents, respectively the fifth policy document on spatial planning and the national infrastructure and transport plan. this happened in a period of economic growth and increasing mobility and consequent congestion. these developments led to conceptual changes in the spatial development policy. previous policies focused on monocentric regional agglomerations (stadsgewesten). this focus was replaced by the concept of ‘network city’ (later called urban networks), which reflected the growing belief that society was developing into a network society. the cities constituting these networks were expected to collaborate and form shared spatial development policies to accommodate the development of the network society. that’s why the coordination commission initiated a platform in 1998 for politicians to share their ideas on regional spatial development. participants were the three provinces (north-holland, flevoland and utrecht), the cities of amsterdam, almere and utrecht and the two official regional authorities (roa and bru). at the first meeting of the socalled muiderslot conference participants showed each other their spatial development plans and discussed them. during the second meeting, at the end of 1998, it was decided that the different plans had to be combined into one shared regional spatial plan, which could then act as the region´s contribution to the fifth policy document. the formation of this regional plan was not without its difficulties. especially the complex mutual relations between the city, the regional authority and the province of utrecht were cause for conflicts. in preparation for the fifth policy document the ministry started to assess the qualitative and quantitative housing needs in the urban networks. for the amsterdam region it was decided, in consultation with the region, that 150.000 houses were to be built between 2010 and 2030. given this enormous task amsterdam’s alderman for spatial planning stadig was asked (during the fifth conference meeting) to lead the accommodation of all these new houses and jobs in the region. one of his first acts as chairman was the exclusion of the utrecht region, which had been playing a waiting game in the previous years. above that, the platform was renamed in to north wing consultation (noordvleugel overleg) and chaired by stadig and meijdam as representative of the province of north-holland (table 6). during the first meeting of this new platform, the aldermen for spatial planning came to an agreement on the accommodation of the houses, but when the aldermen of transport and infrastructure joined at the second conference in november 2002, it became clear that the housing and job growth would place a heavy burden on the already congested infrastructure. therefore it was decided, during the third conference in september 2003, that the majority of the houses would be built on the corridor schiphol international airport amsterdam – almere, because of the close relationship between work, living and infrastructure on this axis. investments in infrastructure were prioritized for the corridor and road pricing was investigated. moreover, a platform accessibility north wing (platform bereikbaarheid noordvleugel) was founded to lobby for these infrastructure projects. article_14-13.indd 67 03/03/15 3:04 pm 68 d. schipper et al. / the emergence of metropolitan governance in 2004, the newly elected national government presented its own national spatial development strategy (nota ruimte). in this strategy the national government recognized the region as a legitimate negotiating partner and as an area for coordinated inter-ministerial investments. the north wing scheme included eight large-scale projects and a total sum of 8 billion euros to realize them. in addition, the nota ruimte also announced a new policy document from the ministry of economic affairs, called peaks in the delta (pieken in de delta), which was the economic agenda of the national government, aimed at seizing opportunities instead of overcoming inequalities. for the north wing consultation this was an important incentive to expand the conferences with meetings on the economic opportunities of the north wing. the focus of the national government on the economy wasn’t without a reason. since 2005 alarming reports appeared on the declining competitiveness of the randstad (a polycentric urban area comprising the four largest cities in the netherlands and some smaller cities) compared to other european metropolises. these reports brought governmental restructuring back on the agenda. according to the analysis of a committee appointed by the minister of interior, this declining competitiveness of the randstad was understood to be caused by the jurisdictional fragmentation and complexity. it pleaded for the creation of a governmental authority on the scale of the randstad. but 2007 saw a shift in dutch politics. the newly elected national government considered governmental reform a waste of energy. however, it did agree on the conclusion of the committee that a short term strategy for the randstad was needed and so the inter-ministerial scheme was up-scaled table 6 north wing consultation. north wing consultation content the conferences were organized to find enough space for the development of housing and commercial areas for the period 2010-2030, while also looking at the relationship between these developments and the effects on the accessibility of the region. in order to improve the accessibility agreements had to be reached on the development of infrastructure and public transportation. structure the first muiderslot conferences are attended by the three provinces north-holland, utrecht and flevoland, the two official regional authorities bru and roa and the cities of amsterdam, almere and utrecht. the utrecht region was eventually excluded by the alderman of spatial planning of amsterdam. the north wing consultation was founded and chaired by stadig and meijdam. the administrative core group, consisting of the six largest municipalities, the provinces of north-holland and flevoland, and city-region authority, would prepare the meetings and agreements to be made. those municipalities within both provinces that could contribute to the formation of a regional spatial strategy, in response to the housing needs as defined by the national government and the region, were invited to join the conference meeting. 29 aldermen of spatial planning attended the first meeting and at the second meeting the representatives in transport and infrastructure joined. process the conference meetings are informal and participation is voluntary. agreements should be formalized in the spatial strategies of the participants. the administrative core group supervises the implementation of the agreements by the participants. article_14-13.indd 68 03/03/15 3:04 pm d. schipper et al. / the emergence of metropolitan governance 69 table 7 amsterdam metropolitan area. amsterdam metropolitan area (mra) content the amsterdam metropolitan area has to become an international competitive metropolis by 2040. five impulses were identified as being crucial for this ambition: (a) improving the metropolitan accessibility, (b) intensification and transformation of urban areas, (c) development of the metropolitan landscape, (d) a focus on sustainability and the role of climate change, and (e) economic development. the mra developed location specific investment strategies and acted as a negotiating partner of the national government. structure there are 36 municipalities, 2 provinces and 1 city-regional authority that participate in the mra. the metropolitan governance arrangement consists of three platforms: one on spatial planning, one on the accessibility of the metropolitan area and a third platform on the economic development. these three platforms each have their own core administrative group, which is made up of the two provinces, the largest cities, the city-regional authority, and in the accessibility platform the department of waterways and public works is also involved. process the conference meetings are informal and participation is voluntary. agreements should be formalized in the spatial strategies of the participants. the administrative core group supervises the implementation of the agreements by the participants. to the randstad. consequently, the north wing scheme had to be re-evaluated and was incorporated into the newly created randstad scheme, which affected the status of the north wing consultation. in an attempt to reestablish its position and to corroborate the growing metropolitan consciousness, the participants in the north wing consultation agreed (during the seventh conference in december 2007) that the amsterdam metropolitan region had to become an international competitive metropolis by 2040. to support this ambition the north wing was renamed in to the amsterdam metropolitan area (mra, metropoolregio amsterdam). for the mra, five crucial ‘impulses’ were indicated during its first conference in april 2009 (see table 7). a third platform was established to develop these impulses into policy: the platform regional economic structure (pres). together with the accessibility platform and the central administration (bkg) these three platforms are responsible for the implementation of the five impulses. the establishment of the mra and its sub-platforms, as a new but not final stage in metropolitan governance, concludes this case description. the analysis of the case is subject of the next section. 4. analysis twenty-five years of metropolitan governance in the amsterdam region has shown continuous and considerable selection pressures. it is for example important to note how the city of amsterdam has changed. early 1980s, the city was relatively poor and in a bad condition compared to its adjacent cities, and faced economic and social decline in neighborhoods as the result of the suburbanization of higher and middle-income families. consequently, the city aimed for a city province to redistribute wealth among the article_14-13.indd 69 03/03/15 3:04 pm 70 d. schipper et al. / the emergence of metropolitan governance municipalities in the region in an attempt to reverse its fortunes. but the city grew rapidly in the following decades, both economic and demographically, which strengthened the position of the city. this change caused the balance between the city and its neighboring municipalities to shift. the coordination focused increasingly on the joint accommodation of population growth for different target groups. with amsterdam becoming the proverbial engine for the region, more municipalities desired to become part of the amsterdam metropolitan area. the relationship between the region and the national government also proved quite dynamic. most notable were the ever-changing regionalization policies as developed by the national government. while the policies in the early 1980’s were aimed at the improvement of inter-municipal collaboration, for which the joint provision act was revised, this same act proved counter-productive during the 1990’s. as a result, the national policy switched to the creation of a fourth administrative layer in the form of a city province. when the implementation of the city province failed, the policies soon shifted back to inter-municipal coordination and the delegation of tasks by the national government for which the joint provision act was once again revised. another type of dynamics can be found in the planning approaches for the region as deployed by the different departments of the national government. early 1980’s, the region was supposed to be accommodating the suburbanization through concentrated urban expansion to contain urban sprawl. at the end of the 1980’s this policy shifted in order to stop the inner-city decline, and fostering the international economic position of the netherlands became a predominant goal, with the randstad and its international airport and seaport as the key features. at the end of the 1990’s, the regions were regarded as interconnected cores in a global network, the network cities, soon to be followed up with the term urban network. inter-ministerial schemes were set up to coordinate investments in these urban networks, but much debate lasted on where an urban network starts and ends. this was shown in 2008 when the national government scaled-up their national scheme. these examples show that the actors in the amsterdam metropolitan region were under continuous pressure to change the means and ends of the metropolitan governance system in terms of content, structure and process. one would expect these incentives to lead to a continuous search process in which variation is explored and selections are made in order to improve the fitness with the environment, as elaborated in section 2. however, if we look at the development of the metropolitan governance systems over time, they show a considerable capacity to absorb the pressures exerted on them by exploiting the existing governance arrangement. in the following section we will have a closer look at the trajectory of the metropolitan governance systems. the case description started with a metropolitan governance system that was muddling through. attempts to radically change the governance system all ended in failure. when the pressure on the system was severely increased by the end of the 1980s, by the city of amsterdam and the national government, the system had to adapt by letting go of the voluntary collaboration. instead a regional authority with binding terms of reference article_14-13.indd 70 03/03/15 3:04 pm d. schipper et al. / the emergence of metropolitan governance 71 was formed. although almost all members seemed to agree that this was a first step towards a more intensive form of collaboration, the region tried to keep control over their own planning, structure and goals. this attempt was reinforced by the bottom-up approach of the national government. but the system was soon pushed in a trajectory towards a city province when the rotterdam region opted for a far-reaching proposal in its attempt to stay in control over the process. the coevolutionary dynamics between the national government and the rotterdam region resulted in such a pressure on the amsterdam region that they had to let go their own open planning and goals and were forced in a more reactive role. what started out as an open and dynamic process soon turned in to a process dominated by legislation. no longer were the questions asked of what the new administrative authority should achieve and who should do that. instead, the question of how this administrative authority could be achieved became dominant. with the time pressure on the negotiators increasing, variation diminished resulting in less creativity, innovation and room for different opinions. this resulted in an ever-diminishing support for the city province, but the momentum is carried forward since very few people dare to challenge the bandwagon. the system was also unaware of the growing unrest among the citizens and their potential influence on the process. the outcome of the referendum provided a window of opportunity for most people to withdraw their support in public. in other words, the massive vote of the citizens of amsterdam against the subdivision of their city pushed the system out of its balance; since the system’s adaptive capacity was diminished it could no longer cope with the internal and external pressures. in terms of punctuated equilibrium this period is marked by an alternation between inertia and swift change during which the system can create substantially different configurations. the change to the two-track policy shows that a system that has undergone a transition carries the traces of its past state in to the new state as the regional body amsterdam was still a formal administrative body. it also highlights the self-organizing capacity of the region to create this new arrangement with a flexible structure without external control. interestingly, we can once again observe, that this capacity is practiced by the region to stay in control over its own future, e.g. to resist externally induced change towards the city province, by creating an alternative arrangement and by forging ‘new’ alliances (province of north-holland). however, the flexible governance system proved to be well developed in terms of exploration, but was lacking the capacity to transform gains from exploration into objects of exploitation (cf. duit & galaz, 2008). this made the system incapable of developing a joint spatial development strategy, thereby risking the loss of national government funding and authority over the development of the region. this pressure forced the region to hasten its decisions in terms of content, process and structure of the metropolitan governance system. in terms of content it was decided to focus on the housing task. as of structure the utrecht region was excluded from the meetings and only those municipalities that could contribute to the regional spatial strategy were invited. although the meetings stayed informal and voluntary, implementation article_14-13.indd 71 03/03/15 3:04 pm 72 d. schipper et al. / the emergence of metropolitan governance was supervised by the administrative core group. this shows that selection pressures can also have enabling principles as it had a positive effect on the north wing consultation to develop itself in to what would become the amsterdam metropolitan area or mra. so what started as an informal gathering to talk about spatial plans, eventually became the most important coordination platform in the region, making the regional collaboration amsterdam and the coordination commission redundant. hence, it can be concluded that the most advanced form of metropolitan coordination in the netherlands was never planned as such, but is the unforeseen outcome of the coevolutionary process between the governance system and its environment. overall, the metropolitan governance arrangements proved to have remarkable self organizing capacities when under internal and external pressure. we can see that both strategies of exploitation and exploration were used in response to selection pressures. in terms of exploitation it can be observed that the structure of the arrangements proved to be an important constant element. once the relevant relationships have been selected there is a high degree of stability in these relationships. only after the referendum we can see that alternative system boundaries were explored to regain a fit with the environment. yet, it took new boundaries to stabilize the newly formed arrangement. in terms of process we can see initial adaptive moves by the first two arrangements in the sense that local authorities delegated authority to restore stability. however, these adaptive moves proved to be insufficient to deal with the external pressures of the national government and the process towards the city-province was initiated. this pushed the system in a pattern of exploitation to reach the planned end-state of a city province and eventually led to the loss of its adaptive capacity. after the referendum a voluntary coordination mechanism proved to provide a better fit. the content of the arrangements seems to show the most dynamics as new goals and tasks are continuously being adopted and explored, with governmental funding being an important incentive. 5. conclusions this article sets out to explain the emergence of metropolitan governance in coevolutionary terms. a longitudinal coevolutionary analysis of metropolitan governance was developed in order to understand the emergence of metropolitan governance systems as a complex process of reciprocal selection. in the case description we have presented the selection pressures exerted on the metropolitan governance systems in the amsterdam region between 1982 and 2009 and the changes made to the configuration of the governance systems in terms of content, structure and process. if we consider the full 25 years covered in the analysis, it appears that long periods of dynamic equilibrium are punctuated by short periods of instability, where considerable change to the governance system is possible. we ascribe the periods of dynamic equilibrium to the capacity of the system to accommodate the pressures in its existing routines. stability occurs when relatively little variety is accepted and/or generated, i.e. exploitation of the existing form of collective action. change is associated with article_14-13.indd 72 03/03/15 3:04 pm d. schipper et al. / the emergence of metropolitan governance 73 an increase of the variety that is considered, i.e. the exploration of different forms of governance. the various forms of collaboration in the amsterdam region are a result of this tension between exploration and exploitation. the analysis shows that metropolitan governance is largely driven by this process of reciprocal selection in order to find a better fit with its environment, as actors are under continuous pressure to change the means (structure and process), and ends (content) of the governance system. the change, or lack thereof, is therefore a function of increasing or decreasing fit between the two. the analysis showed that there is always a mix of exploitation and exploration, with the first being dominant in times of stasis and the latter during times of change. hence, there seems to be a relationship between adaptive capacity and variety, i.e. that variety is created when the system has lost its adaptive capacity and fit with the environment. however, keeping balance between exploitation and exploration isn’t only difficult because of the reinforcing nature of exploitation and exploration, but also because of the events and developments relegated outside the span of control leaving the metropolitan governance system only partially in control. hence, stable systems run the risk of losing their resilience in the face of external selection pressures creating variety, as witnessed by e.g. the referendum. successful configurations of structure, process and content will therefore last as long as fitness is maintained. this however does not imply that the governance systems have a complete lack of control. the actors were often capable of influencing the process. especially remarkable in this case were the relationship between the regions and the national government. twice, the threat of externally induced change by the national government triggered the region’s explorative capacity in an attempt to remain in control over its own future. in contrast, the case also showed that potential national government funding acted as a trigger for the region’s capacity to transform gains from exploration into objects of exploitation. here key actors played an important role in stimulating the self-organizing capacity of the metropolitan region and making selections in terms of content, structure and process. we have visualized the trajectory of the government arrangements in figure 2. each arrangement, as described in tables 2 to 7, has been classified according to the flexibility in its composition (vertical axis) and the measure of hierarchy (horizontal axis). note that the axis in the illustration don’t represent absolute values, but rather are an indication based on our own judgment. the first two arrangements (tables 2 and 3) are characterized by adaptive changes towards the city province in a dynamic process. soon the selection pressure on the system constrain the direction of change and possible variation (table 4). with the system no longer able to adapt to changing circumstances, it had to change radically to renew its fitness with the environment (table 5). a flexible and voluntary arrangement was deemed necessary, but proved to be detached from its environment because of its overemphasis on exploration. with selections made in terms of structure and content and the enabling principles of the selection pressures led to the stabilizing of the arrangement and further exploitation (tables 6 and 7). article_14-13.indd 73 03/03/15 3:04 pm 74 d. schipper et al. / the emergence of metropolitan governance it should be noted that the coevolutionary process described above has no fixed endstate in the long run, even though actors make purposeful decisions about it. as shown it is subject to non-ergodic chance events that have a considerable influence on the evolutionary trajectory. each decision or chance event contributes to the overall course of the coevolutionary process, but never fully determines it. the fact that the metropolitan governance has survived in various forms up until the current mra is a testimony to the adaptive capacity present in the systems. however, instability is a sign of a healthy system that has the capacity to regain fitness. this should provide consolation to those who busy themselves with the reorganization of metropolitan governance. references arthur, w. b. (1994). increasing returns and path dependence in the economy. ann arbor, mi: the university of michigan press. arthur, w. b., durlauf, s. n., & lane, d. a. (eds.). (1997). the economy as an evolving complex system ii. reading, ma: addison-wesley. bertolini, l. (2011). complex systems, evolutionary planning? in g. de roo & e. a. silva (eds.), a planners encounter with complexity (pp. 81–98). london, england: ashgate. boons, f. a. a., gerrits, l. m, & marks, p. (2011). collective decision making as an evolutionary process. in gerrits, l. m, p. marks, & f. a. a. boons (eds.), public administration in complexity [proceedings of the compact conference, rotterdam]. litchfield park, az: emergent publications, 368–385. figure 2. the evolution of metropolitan governance in the amsterdam metropolitan region (1982–2009). article_14-13.indd 74 03/03/15 3:04 pm d. schipper et al. / the emergence of metropolitan governance 75 bordewijk, p. (2000). onderhandelen binnen en tussen bestuurslagen: de stadsprovincie rotterdam [negotiation within and between levels of government: the county town of rotterdam]. in p. l. hupe, m. a. beukenholdt-ter mors, & h. l. klaassen (eds.), publiek onderhandelen: een vorm van eigentijds besturen. samsom, alphen aan den rijn, 61–89. brenner, n. (2002). decoding the newest metropolitan regionalism in the usa: a critical overview. cities, 19(1), 3–21. brenner, n. (2003). metropolitan institutional reform and the rescaling of state space in contemporary western europe. european urban and regional studies, 10(4), 297–324. buitelaar, e., lagendijk, a., & jacobs, w. (2007). a theory of institutional change: illustrated by dutch cityprovinces and dutch land policy. environment and planning a, 39(4), 891–908. byrne, d. (1998). complexity theory and the social sciences. london, england: routledge. byrne, d. (2005). complexity, configurations and cases. theory, culture & society, 22(5), 95–111. campbell, d. t. (1969). variation and selective retention in socio-cultural evolution. general systems, 14, 69–85. corning, p. a. (1983). the synergism hypothesis: a theory of progressive evolution. new york, ny: mcgraw-hill. daneke, g. a. (1996). a science of public administration? public administration review, 50, 383–392. david, p. a. (1985). clio and the economics of qwerty. economic history, 75(2), 332–336. dopfer, k. (2005). evolutionary economics: a theoretical framework. in k. dopfer (ed.), the evolutionary foundation of economics. cambridge, ma: cambridge university press. duit, a., & galaz, v. (2008). governing complexity—insights and emerging challenges. governance, 21, 311–335. faludi, a. (1994). coalition building and planning for dutch growth management: the role of the randstad concept. urban studies, 31(3), 485–507. feiock, r. c. (2008). metropolitan governance and institutional collective action. urban affairs review, 44(3), 356–377. foster, j., & hölz, w. (2004). introduction and overview. in j. foster & w. hölz (eds.), applied evolutionary economics and complex systems. cheltenham, england: edward elgar publishing. gerrits, l. m. (2008). the gentle art of coevolution: a complexity theory perspective on decision making over estuaries in germany, belgium and the netherlands. rotterdam, netherlands: erasmus university. gerrits, l. m. (2011). a coevolutionary revision of decision making processes: an analysis of port extensions in germany, belgium and the netherlands. public administration quarterly, 35(3), 309–339. gerrits, l. m. (2012). punching clouds: an introduction to the complexity of public decision-making. litchfield park, az: emergent publications. greener, i. (2002). theorising path dependency: how does history come to matter in organisations, and what can we do about it? york, england: the university of york. gual, m. a., & norgaard, r. b. (2010). bridging ecological and social systems coevolution: a review and proposal. ecological economics, 69(4), 707–717. hajer, m. a., van tatenhove, j. p. m., & laurent, c. (2004). nieuwe vormen van governance: een essay over nieuwe vormen van bestuur [new forms of governance: an essay on new forms of administration] (rivm rapport 500013004). haran, n. (2010). the power to collaborate: how judicious use of power accelerates the strategic capacity of regions in the netherlands. delft, netherlands: eburon. heinelt, h., & zimmermann, k. (2011). how can we explain diversity in metropolitan governance within a country? some reflections on recent developments in germany. international journal of urban and regional research, 35(6), 1175–1192. hodgson, g. m. (2003). the mystery of the routine: the darwinian destiny of an evolutionary theory of economic change. revue economique, 54(2), 355–384. innes, j. e., & booher, d. e. (2003). collaborative policymaking: governance through dialogue. in m. hajer & h. wagenaar (eds.), deliberative policy analysis: understanding governance in the network society. cambridge, ma: cambridge university press. article_14-13.indd 75 03/03/15 3:04 pm 76 d. schipper et al. / the emergence of metropolitan governance janssen-jansen, l. b. (2011). from amsterdam to amsterdam metropolitan area: a paradigm shift. international planning studies, 16(3), 257–272. kantor, p. (2008). varieties of city regionalism and the quest for political cooperation: a comparative perspective. urban research and practice, 1(2), 111–129. kaufman, h. (1991). time, chance and organizations: natural selection in a perilous environment (2nd ed.). chatham, nj: chatham house. kerr, p. (2002). saved from extinction: evolutionary theorising, politics and the state. british journal of politics and international relations, 4(2), 330–358. koppenjan, j. f. m., & klijn, e. h. (2004). managing uncertainties in networks. london, england: routledge. levelt, m., & janssen-jansen, l. (2013). the amsterdam metropolitan area challenge: opportunities for inclusive coproduction in city region governance. environment and planning c: government & policy, 31(3), 540–555. lewin, a. y., & volberda, h. w. (1999). prolegomena on coevolution: a framework for research on strategy and new organizational forms. organization science, 10(5), 519–534. marshall, s. (2009). cities, design & evolution. london, england: routledge. norgaard, r. b. (1984). coevolutionary development potential. land economics, 60(2), 160–173. norgaard, r. b. (1994). development betrayed; the end of progress and a coevolutioary revisioning of the future. new york, ny: routledge. norgaard, r. b. (1995). beyond materialism: a coevolutionary reinterpration. review of social economy, 53(4), 475–486. norgaard, r. b. (2010). ecosystem services: from eye-opening metaphor to complexity blinder [special issue: payments for ecosystem services]. ecological economics, 69(6), 1219–1227. pierson, p. (2000). increasing returns, path dependence and the study of politics. american political science review, 94(2), 251–267. room, g. (2011). complexity, institutions and public policy: agile decision-making in a turbulent world. cheltenham, england: edward elgar. salet, w. g. m. (2003). amsterdam and the north wing of the randstad. in w. g. m. salet (ed.), metropolitan governance and spatial planning. london, england: routledge. savitch, h. v., & vogel, r. k. (2009). regionalism and urban politics. in j. davies & d. imbroscio (ed.), theories of urban politics (2nd ed). thousand oaks, ca: sage. sementelli, a. (2007). distortions of progress: evolutionary theories and public administration. administration and society, 39(6), 740–760. teisman, g. r., gerrits, l. m. & buuren, m. w. van (2009). introduction. in g. r. teisman, m. w. van buuren & gerrits, l. m. (eds.), managing complex governance systems. dynamics, self-organisation and coevolution in public investments (pp. 1–16). new york / london: routledge. tetteroo, p. (2003). uniformiteit en pluriformiteit op decentraal niveau [uniformity and diversity at local level]. in a. j. van dijk (ed.), laag voor laag: aspecten van regionalisering (pp. 23–32). den haag, netherlands: rob. tomàs, m. (2012). exploring the metropolitan trap: the case of montreal. international journal of urban and regional research, 36(3), 554–567. van de ven, a. h., & poole, m. s. (1995). explaining development and change in organizations. the academy of management review, 20(3), 510–540. van den bergh, j. c. j. m., & kallis, g. (2009). evolution policy: papers on economics and evolution (max planck institute of economics, evolutionary economics group). reading, ma: addison-wesley. van der bol, n. (2010). managing metropolitan complexity: understanding metropolitan management issues from a public managers’ perspective. rotterdam, netherlands: erasmus university. van der lans, j. (2006). kleine geschiedenis van de stadsregio amsterdam. wormer, netherlands: inmerc bv. van der veer, j. (1997). omstreden stadsgrenzen: een eeuw besluitvorming over annexaties en regionale besturen rond amsterdam en eindhoven. delft, the netherlands: eburon. article_14-13.indd 76 03/03/15 3:04 pm d. schipper et al. / the emergence of metropolitan governance 77 voigt, r., & van spijker, w. (2003). spelen met betekenis: verhalen over succesvol vernieuwen bij de overheid [playing with meaning: stories of successful renewal in government]. assen, the netherlands: koninklijke van gorcum bv. volberda, h. w., & lewin, a. y. (2003). co-evolutionary dynamics within and between firms: from evolution to co-evolution. journal of management studies, 40(8), 2111–2136. wilson, e. o. (1975). sociobiology: the new synthesis. cambridge, ma: belknap press. witte, g. t. (2010, may). provincie is oorzaak van het probleem [province is causing the problem]. openbaar bestuur, 5, 33–36. article_14-13.indd 77 03/03/15 3:04 pm article_14-13.indd 78 03/03/15 3:04 pm complexity, governance & networks special issue: complexity, innovation and policy (2017) 2-6 introduction: on the coevolution of innovation and public policy doi: http://dx.doi.org/10.20377/cgn-38 2 university of bamberg press introduction: on the coevolution of innovation and public policy peter marks & lasse gerrits why a special issue on the coevolution of innovation and public policy? innovation and public policy are much researched subjects and evolutionary theories have offered a very helpful explanatory framework for these topics in many scientific disciplines (sanderson, 1990). naturally, societies benefit from innovation as a way of generating growth. however, to enable or even steer innovation through public policy is not a simple, straightforward task. misguided policies can waste considerable resources, including tax money, without gaining any traction. questions about policy and innovation do not only have practical relevance, they also tap into important theoretical and methodological debates. this is particular the case in evolutionary and complexity economics. perhaps somewhat hidden from the common foci in the scholarly debates of public administration and public policy, there is a lively debate within economics about the nature of economic systems and how this should be researched. the debate pitches neoclassical or traditional approaches against so-called heterodox economics. our community should pay ample attention to that debate because it is one where the case for complex systems and evolutionary approaches develops quickly and profoundly. an economic angle may seem at odds with the focus on public policy, but bear with us for a minute. of course, economics is a social science because it focuses on the behavior of people from a scarcity perspective. social scientists from e.g. public administration or public management may have a different approach to human behavior than economists do. but these differences aside, the common theme is behavior and its mechanisms in a dynamic society, and that is where economics and the social sciences can inform each other. indeed, society itself can benefit from understanding the coevolutionary relation between innovation and public policy because such understanding will enhance innovative capacity, which in turn enhances the welfare distribution in society. this special issue deals with questions such as how public policies can facilitate innovation in order to change that welfare distribution, and how the evolutionary economics perspective can inform public policies by looking at complementary mechanisms that facilitate innovation. the perspective selected for this special issue will be of importance for anyone interested in public policies that promote innovation. innovation and economics the theme for this special issue has a long intellectual history. in economics, innovation, or ‘creative destruction’, as it was first coined by joseph schumpeter in 1942, refers to the “[…] process of industrial mutation that incessantly revolutionizes the economic structure from within, incessantly destroying the old one, incessantly creating a new one." (schumpeter, 1942: 83) this innovation process replaces long-standing arrangements by new ones, whilst freeing resources to http://dx.doi.org/ complexity, governance & networks special issue: complexity, innovation and policy (2017) 2-6 introduction: on the coevolution of innovation and public policy doi: http://dx.doi.org/10.20377/cgn-38 3 university of bamberg press be relocated and deployed elsewhere to seek profit. although people may lose their jobs and companies may disappear, society will benefit from the innovations as it will be able to become more productive, thus raising welfare. that is, it is in the nature of capitalism that economic markets evolve naturally (conf. “spontaneous order” by hayek). the idea that innovation spurs economic growth by reshaping knowledge and technology is a step away from neoclassical economics that focus on productive factor accumulation (capital and labor) for economic growth. innovation then has impact on the macroeconomic performance as factor markets will function differently, but also impact long term growth. “at the microeconomic level, restructuring is characterized by countless decisions to create and destroy production arrangements. these decisions are often complex, involving multiple parties as well as strategic and technological considerations.” (cabellero, no year: p. 1) the analysis of decision making doesn’t focus on the individuals and firms but especially on the institutional context of research, governments and society. arguably product or process innovations take place at any moment in time, which means that a continuous restructuring of the economy would occur if no institutional barriers would be put into place. however, there are many such barriers to change, e.g. technological, economical, institutional, political, as well as with regard to specific policies such as labor safety or consumer protection. in evolutionary terminology, which relates strongly to the complexity sciences, one could speak of selection mechanisms that select certain innovations from the space of possibilities (variety) while non-selecting or stopping others. innovation then is propelling change in a context where certain elements are selected from a variety of options that are enabling or constraining the future space of possibilities. this is essentially an evolutionary process. from an evolutionary (economics) perspective “[…] it is argued that selection mechanisms bring to the fore techniques, organizational routines and products that are best adapted to their respective environmental contexts” (foster & metcalfe, 2001: 1). the introduction of this evolutionary perspective can be attributed to the works of boulding’s (1981) book evolutionary economics and to nelson & winter’s seminal book an evolutionary theory of economic change (1982). the evolutionary perspective puts much more emphasis on processes, changes and structural transformations (hodgson, 2002: xix). persistent questions cover how selection principles operate in economic contexts and how the variety upon which selection works comes into being (foster & metcalfe, 2001; dopfer, 2005). the evolutionary metaphor, once carefully reconstructed, provides the basis for the study of economic change and process (potts, 2000: 183). the continuously developing set of institutional arrangements and technological innovations implies that the economy is not necessarily in stasis, but is ‘out of equilibrium’. that is, the economy might show every-changing novel behavior that does not appear in a steady state. the lack of stasis is due to constant disruptions coming from agents adjusting to new situations. this presents a classical theme in the complexity sciences; i.e. agents reacting to changes caused by other agents. it is ever more complex as the agents are uncertain how other actors perceive the same situation and how they might react. “agents therefore inhabit a world that they must cognitively interpret—one that is complicated by the presence and actions of other agents and that is ever changing.” (arthur, durlauf & lane, 1997: 5) in this complexity perspective structure matters because all economic action involves recurring interactions among agents. such interactions define systems that conversely constrain and facilitate the interactions, as per holland’s thesis (1995). these multilevel interactions mean http://dx.doi.org/10.20377/cgn-38 complexity, governance & networks special issue: complexity, innovation and policy (2017) 2-6 introduction: on the coevolution of innovation and public policy doi: http://dx.doi.org/10.20377/cgn-38 4 university of bamberg press that the traditional neoclassical perspective of methodological individualism doesn’t hold anymore. the fact that these actions and interactions perform at different levels means that the reciprocal causation operates between the different levels of organization; in other words: the fundamental principle of organization is the idea that units at one level combine to produce units at the next higher level (arthur, durlauf & lane, 1997: 5-6) thus it is of interest to research how an economy emerges, grows and changes structurally over time represented by ideas about innovation, economic development, structural change, and to focus on the role of history, institutions, and governance in the economy. all in all, evolutionary and complexity economics offer a different way of understanding innovation and public policy in the economy. it questions traditional assumptions, in particular the ceteris paribus clause. time is an important driving factor, which implies that structures may constantly (re)form and that equilibria are temporal at best. traditional equilibrium analyses are no longer sufficient. the complexity and evolutionary perspective “[…] gives us a world closer to that of political economy than to neoclassical theory, a world that is organic, evolutionary, and historically contingent.” (arthur, 2013: 1) this issue the article by j.w. stoelhorst demonstrates that the evolutionary perspective, which combines the starting point of competition with the ideas of cooperative arrangements, offers explanatory value for economic progress. he builds his argument on the so-called naturalistic approach, which considers the various forms of human cooperation as products of gene-culture co-evolutionary processes and multi-level selection. it explains why human nature has evolved into being better than rational in collective action problems. moving away from the decentralized exchange and rational self-interested choice, humans solve the collective action problems through establishing functional groups like households and firms. it creates and economic landscape of multi-level competition; within and between these functional groups. here, organizations dominate markets; in contrast to economic theory’s traditional emphasis on markets over organizations. this means that a different perspective on designing institutions that improve our welfare is needed. jason potts, in the second article, focuses on that need, though from a different angle. his argument starts with the presentation of two more or less opposing views of science, economics and policy: an equilibrium approach and a complexity approach that focuses on rules. this means that there is an equilibrium-type of economics – focused on forces to move resources – and a complexity approach, that is focused on rules structuring knowledge. consequently, (economic) policies are either focusing on reallocating resources or on redesigning rules for collective action. he demonstrates how the innovation problem from the allocation perspective implies top-down policies reallocating resources towards innovation in order to maximize some social welfare function. however, from the complexity perspective the innovation problem is a collective action problem that can be resolved with effective rules to enable agents to cooperate to pool resources and information. these emergent coordinating rules form a solution to the innovation problem: they are, in effect, policy. http://dx.doi.org/ complexity, governance & networks special issue: complexity, innovation and policy (2017) 2-6 introduction: on the coevolution of innovation and public policy doi: http://dx.doi.org/10.20377/cgn-38 5 university of bamberg press the third article by koen frenken also argues that current innovation policies are based on two dominant frames; i.e. on market failure or systems failure. frenken deems these two frames of reference as too narrow and too limited in the light of contemporary societal challenges. innovation is strongly path-dependent, i.e. concerns strongly related activities instead of novel and unrelated activities. however, any economy will benefit more from a mixture of related and unrelated innovations, and requires innovation policies that facilitate both. this means that generic innovation policies can be relied upon to stimulate generic diversification, while at the same time stimulating unrelated diversification through more specific policies aimed at organizing the process of demand articulation. this would lead to clear translations of societal challenges into concrete objectives in order to build temporary innovation systems that can develop solutions bottom-up. in order to realize such innovation policies much can be learned by incorporating other disciplines, e.g. political sciences or institutional sociology. peter marks and lasse gerrits in the fourth article take a retrospective approach and argue that policies should be understood and evaluated in terms of the coevolution between social practice and (technological) innovation. they apply this perspective to the dutch high-speed railway case, which features a mixture of successful and less or even unsuccessful innovations in the three decades of development. lack of success can be explained in terms of misfits between innovations and social practices.. that is, the case demonstrates that different underlying coevolving technologies as well as social practice need to be in sync in the multilevel high-speed railway innovation to reach the desired policy goals. this complexity informed socio-technological innovation perspective contrasts with traditional policy evaluations that are geared to finding the one variable that supposedly controls for all the policy problems encountered. together, the articles in this special present a refreshing, alternative and complexityinformed look at innovation and innovation policy. we hope that this inspires scholars in public administration and public policy, as well as practitioners eager to find the keys to successful innovation policy. references arthur, w.b. (2013). complexity economics: a different framework for economic thought, sfi working paper 2013-04012 arthur, w. b., durlauf, s. n., & lane, d. a. (eds.). (1997). the economy as an evolving complex system ii. proceedings volume xxvii, santa fe institute studies in the science of complexity. reading, ma: addison-wesley. boulding, k.e. (1981). evolutionary economics. beverly hills, ca: sage publications. caballero, r. (no year). creative destruction, mit: working paper, found at http://economics.mit.edu/files/1785, visited december 10th 2016. dopfer, k. (ed.). (2005). the evolutionary foundations of economics. cambridge university press. foster, j. & metcalfe, j.s. (2001). frontiers of evolutionary economics: competition, self-orgnization and innovation policy, cheltenham, uk: edward elgar publishing. hodgson, g. m. (ed.). (2002). a modern reader in institutional and evolutionary economics: key concepts. cheltenham, uk: edward elgar publishing. holland, j. (1995). hidden order: how adaptation builds complexity. jackson: perseus books nelson, r.r. & winter, s.g. (1982). an evolutionary theory of economic change. cambridge, ma: harvard university press. potts, j. (2000). the new evolutionary microeconomics: complexity, competence and adaptive behaviour. cheltenham, uk: edward elgar publishing. http://dx.doi.org/ http://economics.mit.edu/files/1785 complexity, governance & networks special issue: complexity, innovation and policy (2017) 2-6 introduction: on the coevolution of innovation and public policy doi: http://dx.doi.org/10.20377/cgn-38 6 university of bamberg press sanderson, s. (1990). social evolutionism: a critical history. cambridge, ma: basil blackwell schumpeter, j. (1942). capitalism, socialism and democracy. new york, ny: harper & row. http://dx.doi.org/ monetary convergence across the economic community of west african states: observing convergence as patterns 50 university of bamberg press monetary convergence across the economic community of west african states: observing convergence as patterns author: david alemna the university of portsmouth, united kingdom email: david.alemna@port.ac.uk since its inauguration, the economic community of west african states has stressed its desire to advance regional integration through the establishment of a common single currency (the eco). this policy has been considered advantageous given the economic benefits derived from the existence of one of the oldest sub-regional monetary unions across frenchspeaking west african economies. for this reason, the west african monetary zone was created as a suggested second monetary zone consisting of english-speaking countries in the region in anticipation that in the long run, the two would converge. while empirical studies into the feasibility of achieving monetary integration in west africa have provided some understanding of causal notions and possible effects, very few studies embrace complexity theory or attempt to use complexity-related conceptual notions in the identification and interpretation of patterns produced in longitudinal applications. using both empirical and theoretical methods, this paper provides a unique longitudinal application of dynamic patterns synthesis as an exploratory tool for observing the potential complexities that the proposed single currency arrangement across west africa is likely to pose. the findings highlight multiple conjunctural causation in observing convergence and unpredictability across the monetary zone. these observations suggest more time is needed to achieve an established single currency. keywords: complexity; dynamic pattern synthesis; monetary convergence; ecowas introduction following the creation of the european monetary union, there has been a growing interest in regional monetary unification arrangements (nkwatoh, 2018; mati, civcir, and ozdeser, 2019). it is suggested that, in unified regional economies, member states can reduce risks amongst vulnerable economies, minimise wars, encourage intra-regional trade and allow complementarities, and entrench competitiveness across member states (mogaji, 2017). this is seen to accelerate economic development and further improve living standards through trade stimulation (abban, 2020). over time, member states would be able to attract investment and deal with asymmetric shocks (such as those resulting from unsustainable debt or fluctuations in export commodity prices), as opposed to a nationalistic approach that may leave a nation vulnerable (mundell, 1961). indeed, the imf (2005) indicated that the economic benefits of monetary cooperation are highlighted in the fact that 52 of its 184 members participate in such arrangements. in west africa, earlier attempts at currency cooperation date as far back as 1912 with the formation of the west african currency board under british colonial rule, which collapsed after decolonisation. following the establishment of the economic community of west african states (ecowas) in 1975, ecowas has stressed its desire to advance regional integration through the establishment of a common single currency (the eco). with the existence of an already established monetary union across french-speaking west african economies (union economique et monetaire ouest africaine—waemu 1), a two-track system was proposed (de grauwe, 2020). this led to the creation of the west african monetary zone (wamz) as a suggested second monetary zone consisting of english-speaking countries in the region in anticipation that in the long run, the two would converge. the two-track approach would suggest that monetary integration may not be linear across ecowas members. ecowas economies are broadly considered to have relatively weak and insufficiently diverse economies in certain cases. the region also has a legacy of political instability 1 the west african economic and monetary union (waemu) is also known as union économique et monétaire ouestafricaine (uemoa). it created the communauté française d’afrique (cfa) franc, which to date is still in use under the west african economic and monetary union (waemu). complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 mailto:david.alemna@port.ac.uk 51 university of bamberg press and poor governance (okafor, 2017). these dynamics are seen to impact inflation and monetary policy responses (carmignani, 2003). significant variations in existing currency regimes across wamz members also suggests a need for strong regulatory frameworks. these countries would have to undergo gradual structural and regulatory reforms in order to surrender their sovereignty and monetary autonomy, and eventually adopt the eco. in some instances, their currencies may lose the competitive advantage they may have derived from the current situation. after a series of delays in earlier attempts to adopt the eco (from 2005 to 2009 to 2015 to 2020 and now to 2027), it is clear that the complexities surrounding the west african monetary integration programme may seem more profound than previously anticipated. evidently, there is a need to reflect on previous experiences within the region to draw lessons for future efforts. while mogaji (2017) studied the evaluation of macroeconomic indicators and the dynamics for monetary integration of west africa—focusing specifically on wamz—those who attempt to assess ecowas convergence have focused on expansions in intra-trade relations and the possible effects of a currency union (see for instance, ezekwesili, 2011; abban, 2020; ilyas, et. al., 2021). alagidede, coleman, and cuestas (2012) applied fractional integration and cointegration econometric modelling to study the behaviour of inflation among wamz countries and discovered the existence of substantial heterogeneity. houssa (2008) utilised a dynamic factor model to observe the asymmetric shocks of monetary union in west africa and found that french-speaking countries experience more demand shocks. through the use of correlation analysis, fielding, lee, and shields (2004) found that the extent of macroeconomic integration across cfa franc members is encouraged by monetary integration. others have also applied optimum currency area to examine the preparedness of ecowas members to form a monetary union and found that such aspirations may be impossible (nkwatoh, 2018; mati, et. al., 2019). while these important studies have provided an understanding of causal notions and possible effects, very few studies embrace complexity theory or attempt to use such conceptual notions in the identification of patterns produced in longitudinal applications. however, a non-linear perspective that applies complexity related methodologies may be more fitting in observing how variables and cases interact over time (mcevoy and richards, 2006; zelli, gerrits, and möller, 2021). complexity theory provides more thorough insights and observation of trajectories over time. the patterns that evolve are also useful in understanding the nature of complexity within the monetary union. identifying these patterns over time also offers a degree of system predictability (order or chaos) and observed macro stability (convergence). to advance the observation of these complex patterns, this paper utilizes complexity theory as a methodological tool for understanding how the two monetary trajectories can be observed during five historical time points. monetary convergence across ecowas – a complexity perspective studies into the feasibility of achieving monetary convergence have revolved around the optimum/optimal currency area (oca) theory. this theory holds that a country anticipating membership into a monetary union should satisfy a set of criteria as a condition for assimilation (hsu, 2010; regmi, nikolsko-rzhevskyy and thornton, 2015). oca is grounded on the concept of sigma convergence. here, convergence is viewed as a decrease in the dispersion index of a variable across a group of cases over time (hammouda et. al., 2009). thereby demonstrating similar patterns of economic behaviour (for instance, in inflation expectancies and long-term interest rates). empirical research on the macroeconomic effects of convergence often focus their analysis on a ‘growing resemblance’ to identify processes of catching up (beta-convergence), growing similarities (sigmaconvergence), or movement towards exemplary models (delta-convergence) (heichel, et. at., 2005). such categorisation of cases into restricted parameters for observing case mobility can be limiting. although convergence may occur over time, case movement may be haphazard and irregular. in applying complexity theory, a social researcher sees society as dynamic, and not static. the predominant characteristics of this dynamism are the interactions that may transpire between and within systems. over time, these interactions may produce patterns (haynes, 2017). nevertheless, although in a state of perpetual chaos and disorder, patterns may also result over time in societies complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 52 university of bamberg press quest for order. in macro-political economics, these changes occur in inconsistent ways at the micro, meso, and macro levels. changes could be triggered by endogenous factors (e.g., changes in government spending) or as a result of external factors (e.g., changes in aggregate demand for export commodities) (see for instance stone, 2008). as such, there are multiple interactions between and within systems. in this complex system, “the causal categories become intertwined in such a way that no dualistic language of state plus dynamic laws can completely describe it” (rosen, 1987.p. 324). bovaird (2008) also describes a complex system as a succession of subsystems. these complex systems are seen to exist with their own rules of behaviour, have external forces to deal with, and interact with each other (klijn, 2008). from this perspective, multiple systems can be identified in the context of ecowas currency convergence. a country can be regarded as a system having its own domestic policy values, priorities, and goals, inter alia, which define its interactions with other systems. in this country, numerous systems exist and interact with each other at all levels. from a broader perspective, the country also becomes a component of a larger system (in this case, a regional block—ecowas) that interacts with other systems. this picture becomes even more complex when one considers interactions at the continental and global levels (cilliers, 1998; nanz & steffek, 2004; stone, 2008)2. in this research, multiple systems are seen in the form of countries and their monetary systems (howlett & ramesh, 2002). figure 1 (below) attempts to provide a graphical representation of the complex interactions observed in the ecowas monetary system. from this, it is evident that each system is bounded by a dotted line. this suggests that systems are semi-permeable and factors within a system may influence activities within another system. this emphasises the assumption that no smaller social system exists in isolation and as such, they may have an influence on each other—and are also influenced by their external environment (cilliers, 1998). figure 1. complex interactions in the ecowas monetary system (developed by author). indeed, highlighted in the literature on currency convergence, in order to reduce complexity in monetary transactions and eliminate exchange rate uncertainty, among other things, countries adopt monetary cooperation (de grauwe, 2000). here, governments attempt to create “order” by identifying certain standards they seek to achieve over time. this is done in anticipation that, in the long run, their monetary systems can become more alike and merge into one larger system. for this reason, monetary cooperation requires a common “single” currency, a common central bank, and a common monetary policy (plasmans et al., 2006)—thereby reducing the complexity in monetary transactions. 2 other systems also include intergovernmental cooperation, nongovernmental organizations, and market inflows that influence trade, labour, and capital flows into countries. p os it iv e an d n eg at iv e fe ed ba ck cou ntry b1 cou ntry c1 cou ntry a1 cou ntry d1 cou ntry b cou ntry c cou ntry a cou ntry d positiv e and negati ve feedba ck ecowas africa global wamz wa em uhuman andfinancial capital markets g lo ba l g ov er na nc e an d po lic ie s ex ter na liti es complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 53 university of bamberg press the implementation of a convergence criterion is seen to promote aspects of certainty and stability, and to limit instability. while attempting to create order, patterns may emerge as a result of the interactions between and within cases, as well as their externalities over time. in such instances, the task is to understand the various journeys the cases may take over time, the patterns that may emerge, and how similar or different cases become. for such observations, causality becomes unpredictable as policy outcomes may vary across similar cases or outcomes may appear similar in divergent cases (ragin, 1997; byrne, 2002; ragin, & rihoux, 2009, haynes, 2017). the interdisciplinary approach to the application of complexity has provided some conceptualisation of certain features that can be used to describe complex systems based on the properties they may exhibit (hmt, 2020). to contain the research within a manageable scope, focus is contextually placed on some fundamental temporal elements of complex systems that tend to be missed in the literature on ecowas convergence. observing complex dynamic patterns over time neoclassical models on long-run macroeconomic convergence have focused on a growing similarity across a range of aggregate scores. such methodological applications sometimes tend to be skewed by outliers or misrepresent changes in variable distribution across cases (haynes and haynes, 2016). it is for this reason that some scholars have highlighted the crucial role the spatial and time dimensions play in observing convergence (knill, 2005; heichel, et. al., 2005). for instance, in their assessment of convergence and heterogeneity across euro-based economies, haynes and haynes (2016) provide some limitations in statistical techniques used to assess convergence. these scholars have highlighted the complex notion of observed convergence and challenged traditional reductionist models that tend to ignore the context and case sensitive nature of this growing similarity. for some complexity scientists, the notion of a system moving from one stable state to another as a result of change is flawed. instead, complex adaptive systems are typically non-linear and highly sensitive to initial conditions (stacey, 1996). here, the context within which interactions occur within, and between, systems at specific time points form the initial conditions for the emergence of future order. while monetary convergence turns to observe an aggregate growth in similarity between currency regimes over time, an application of complexity theory enables the researcher to observe the case-sensitive nature of convergence over time. for instance, in cases where multiple causalities are repeatedly reinforced by experiences associated with circular causality, when “effects” are fed back to alter “causes,” a social researcher may observe the presence of feedback, feedback loops, or complex recurring patterns that may prompt logical paradoxes (smith, 2005). feedback can be explained as the process in which a part of the response of a system is fed back to a cause (martínez-garcía, and hernández-lemus, 2013). this could result in recursive “loops.” positive feedback increases dynamics or the divergence of a system to its original state, thus destabilizing the system. negative feedback on the other hand does the opposite (maturana, 1980). contextually, periods of stability of government policies towards the achievement of a convergence criteria may come with occasional (and often abrupt) changes/events3. these changes may be explained by the presence of positive and negative feedback processes resulting from interactions with other systems, or with the external environment. such interactions could disrupt convergence as national specific policy interventions may produce varying outcomes. in response to negative events, such as the ebola outbreak or the covid-19 pandemic, a country may forgo its ambition to meet a convergent standard in order to respond to domestic impacts. additionally, nationalistic approaches to economic recovery could impact convergence. if this is understood, recognising self-organising patterns may result in better understanding and control of complex systems, and aid in the allocation of resources to combat such impacts (baumgartner and jones, 2002; martínez-garcía, and hernándezlemus, 2013). given that complex systems can be sensitive to initial conditions, and as such, interactions within and between these complex systems could result in feedback loops, emergent complex patterns 3 nelson & winter (1982) categorizes three types of events to understand the subtle differences between pure chance and apparent chance. these are random, unpredictable, and deliberate. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 54 university of bamberg press may develop over time (smith and jenks, 2006). new, unexpected features may start to develop over time from interactions between and within systems (haynes, 2017). these patterns are not simple predictable outcomes, but complex patterns that may result from interactions within a given context and at a specific point in time. in the case of monetary convergence, these patterns should appear, provided that member states achieve the convergence criteria4. to observe these longitudinal interactions within the ecowas monetary system, this paper employs a new methodological tool that provides systematic comparative observations of variable and case patterns over time. methods and data since this article integrates aspects of macro-political economy with complexity theory in its analysis of monetary convergence, an application of dynamic pattern synthesis is utilised as an alternative methodological approach to the assessment of convergence. this is aimed at advancing the development of a new, combined longitudinal case-based configurational method. dynamic pattern synthesis can be used as an exploratory tool to observe complex patterns within a system over time. it combines the use of hierarchical cluster analysis (hca) with configurational modelling to examine similarities and differences between cases at different time points, as well as the extent to which such cases remain similar or different over time (haynes, 2017). hca is used as an exploratory tool to observe patterns of similarity and differences across cases. it provides a focus on countries that maintain cluster grouping over time while highlighting cases that may appear at the periphery of case clustering (outliers). in longitudinal analysis, dynamic pattern synthesis can also be used to identify the movement of cluster partnerships over time and the changes that may emerge based on the variables utilised in the modelling. the application of casebased configurational modelling also allows for careful observation of the trajectory of individual cases and the effects of variable influences on cluster groupings over time. in this study, dynamic pattern synthesis is applied to cases consisting of fifteen countries within the ecowas region in the period 2012-20165. waemu members consist of eight economies: benin, burkina faso, cote d’ivoire, guinea-bissau, mali, niger, senegal and togo. these countries use the cfa franc currency pegged to the euro. however, unlike the waemu, members of the wamz have their individual currency regimes—wamz consists of six west african states: cabo verde, gambia, ghana, guinea, liberia, nigeria, and sierra leone. the period under analysis (2012 2016) was also chosen to reflect the possible impact of an event (the ebola crisis). ecowas assesses monetary convergence on the attainment of specific macroeconomic indicators and structural adjustments that illustrate a threshold level of convergence to become members of the union6. this criterion contains two categories, primary and secondary (see table 1.). the primary criteria consist of mandatory standards that members must meet to be integrated into the union, while the secondary criteria are desirable standards, but not mandatory. in this study, these markers are used as ‘input descriptors’ to assess convergence: that is to say, certain features of a system that can make this system complex (zelli, et. al., 2021). these indicators highlight some relational qualities and timepoints that represent more than the sum of a system’s components, thereby defying reductionist techniques. data on each indicator was collected from the most recent ecowas convergence report (ecowas, 2017a). this report provided the most comprehensive dataset on the ecowas convergence situation during the period under discussion. the complete dataset can be found in appendix 1.4. indicators are measured as per the outlined standards in table 1 below. 4 it is from this perspective that methods like realist evaluation (pawson and tilley, 1997) and qualitative comparative analysis (rihoux & ragin, 2009), amongst many others, are employed to evaluate the influences of the context, intervention’s context, mechanism, and consistencies in outcomes. 5 the period of choice is based on the most recent available online published 2017 ecowas convergence report (see. economic community of west african states, 2017a). this dataset was also found to be the most comprehensive data on the convergence situation at that time. while attempts were made by the researcher to update this dataset, data on some cases could not obtain as some countries do not publicly release data in a timely manner. 6 for a detailed discussion on convergence criteria, bukowski (2006) provides a detailed overview of the economic theory behind this. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 55 university of bamberg press variable code meaning standards primary criteria bd ratio of budget deficit to nominal gdp ≤ 3% ir annual average inflation rate ≤ 10% of target ≤ 5% in 2019 cbf central bank financing of budget deficit ≤ 10% of tax revenues of n-1 ger gross external reserve ≥ 3months secondary criteria tpd ratio of total public debt to gdp ≤ 70% nerv nominal exchange rate variation ± 10% table 1. list of convergence criteria, standards, and variable codes hierarchical cluster analysis is applied using a combination of indicators on the primary and secondary convergence criteria (outlined in table 1). ward’s linkage was considered to be the most appropriate clustering proximity method as it produced fewer mathematical artefacts in the data. input variables were also standardised to z scores to lessen the impact of variables with greater variance. pearson’s correlation was also conducted to observe the possible influences of variables on each other over time (see appendix 1.2. for correlation plots using correlograms, see friendly, 2002). annex 1.3. provides some complementary descriptive diagrams highlighting trends across the convergence indicators. cluster dendrograms (figures 2 5 in the results section) offer a visual representation of the hca results at each time point and are used as the starting point for observing changes that may emerge over the period under analysis. in this hca presentation, cluster dendrograms are placed alongside each other to highlight emergent patterns across cases over time—cases that remain closely similar over one time point to another (i.e., contextually, these cases converge over t0 –> t1). the degree of similarity across clustering is also illustrated by the vertical and horizontal lines that join cases together. when these lines are represented by dashes or dots, this signifies higher dissimilarities. in contrast, tick/solid lines represent stronger similarities across cases. it is important to note that, unlike previous longitudinal applications of dynamic pattern synthesis which have focused on the use of three-time intervals (see for instance: haynes & haynes, 2016; haynes, 2018; taylor, haynes & darking, 2020), this paper compares hca dendrograms across five time points. following the observance of case interactions in cluster grouping over time, and applying the same dataset, configurational case-based modelling is used to identify the influence of associated variables on consistent and emergent clusters. this also allows for the observation of changes in variable patterns over time. configurational case-based modelling approaches such as qualitative comparative analysis (qca) have been used in systematic analysis to observe the ways in which the impacts (outcome) of an intervention can be evaluated by the observance of configurational factors (see for instance ragin, 2000; schneider & rohlfing, 2016; ragin & fiss, 2017; kaimann, 2017). this approach recognises varying contributory factors that can lead to change and allows for the observation of ‘multiple conjunctural causation’ (ragin, 1987; berg-schlosser, et. al., 2009; haynes, 2017). in this application, individual variables are converted into dichotomous variables (threshold scores) based on the achievement of standards set in the convergence criteria. for cases where the achievement of the convergence criteria is met, data is shaded in the colour of the pattern in the hca results (see tables 5 10 in annex 1.1.). likewise, in cases where this is demonstrated across a set of convergent cases, indicators are in “bold” and “italic.” the boolean convention of uppercase equals achieved criteria and lowercase equals did not achieve criteria (rihoux & ragin, 2009) is used in the analysis to highlight longitudinal trends and observe dynamic patterns across variables and cases over time. the results of the application of dynamic pattern synthesis are presented below. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 56 university of bamberg press analysis and results hierarchical cluster analysis: figure 2 (below) provides a comparative analysis of the convergence situation for the first time interval when the hca dendrograms for the years 2012 and 2013 are put together. from the left side (representing 2012 and the right side representing 2013), sierra leone, guinea, and liberia are clustered together. however, the dotted line linking sierra leone to guinea and liberia suggests these two countries share more similarities. below these cases, burkina faso, mali, and nigeria are clustered together. similarly, the dotted line linking burkina faso and mali to nigeria suggests more similarities between the two cases. at the next stage of clustering, the two lines (a tick/solid line for sierra leone, guinea, and liberia and then a dotted line for burkina faso, mali, and nigeria) linking the two cluster groups together suggests more similarities between sierra leone to guinea and liberia as compared to burkina faso, mali, and nigeria—in this sense nigeria can be seen as an outlier. this clustering sequence is repeated until all cases are grouped. in figure 2, two larger/broader cluster formations are also seen to appear in 2012 with a dotted line linking them (i.e., from sierra leone to nigeria and then from guinea bissau to ghana). this shows that cases demonstrate more similarities within cluster members as compared to cases outside cluster memberships. placing hca dendrograms alongside each other allows for the observation of case movement across clustering. for instance, in the top half of figure 2, the hca output shows that, while in 2012 sierra leone, guinea, and liberia shared some similarities, these similarities were stronger between guinea and liberia. however, in 2013 guinea starts to show more similarities with sierra leone and differences with liberia. this can aid in the observation of the stability (or otherwise) in the movement of cases over the period under analysis. in instances where cases remain clustered over time, emergent patterns begin to appear. for instance, between 2012 and 2013 three convergent clusters (emergent cases) can be observed (i.e., guinea bissau and cote d'ivoire; togo and senegal; and cabo verde, gambia, and ghana). these cases demonstrated strong similarities across the two time points as compared to the other cases. the same hca application is repeated across the dataset (i.e., for 2013 2014, 2014 2015, etc.). the dendrograms for the subsequent time points are represented in figures 3 to 5. for the purpose of this study, the hca analysis would place attention on the emergence of case patterns that appear during the period under discussion. table 2 (below) provides a simple presentation of the emergence of case patterns in order of appearance over the period under discussion. an additional column is added to highlight the currency used by each country. colour codes are also applied to highlight emergent case patterns. based on the journeys travelled, complex case patterns begin to appear. while some patterns remained consistent throughout the period under discussion (e.g., guinea bissau and cote d’ivoire), others faded over time (e.g., togo and senegal). in some cases, the dendrograms may suggest their differences grew with time (e.g., gambia, and ghana). in other cases, convergent case patterns are seen to emerge and then grow apart over time (e.g., guinea, sierra leone, and liberia). nevertheless, over time new patterns also begin to emerge (e.g., between 2015 2016 burkina faso and mali, and benin and niger). this highlights the evolution of the ecowas system and its self-organising nature. hca is used in this context to group cases on the basis of their similarities to observe case patterns over time. this is based on the actual values of macroeconomic indicators—not the achievement of the convergence criteria. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 57 university of bamberg press currency 2012 2013 2014 2015 2016 cfa franc guinea bissau guinea bissau guinea bissau guinea bissau guinea bissau cfa franc cote d'ivoire cote d'ivoire cote d'ivoire cote d'ivoire cote d'ivoire naira nigeria nigeria cfa franc togo togo cfa franc senegal senegal escudo cabo verde cabo verde dalasi gambia gambia gambia gambia cedi ghana ghana ghana ghana franc guinea guinea guinea leone sierra leone sierra leone sierra leone dollar liberia liberia liberia cfa franc burkina faso burkina faso cfa franc mali mali cfa franc benin benin cfa franc niger niger table 2. emergent case patterns complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 58 university of bamberg press configurational model: following the observance of case interactions in cluster grouping, and applying the same dataset, configurational case-based modelling is used to identify associated variable influences on consistent and emergent clusters at each time point. variable changes across cases are observed using the raw scores for each variable in the dataset. this is provided in tables 5 10 (see annex 1.1). complementarily, table 3 (below) provides a simplistic configurational model used to observe variable changes that may have influenced the emergence of case patterns over time. as shown in table 3 (below), cote d'ivoire and guinea bissau shared similar variable characteristics in the achievement of ir, cbf, ger, tpd, and nerv in 2012. although nigeria shared slight similarities in achieving cbf, ger, tpd, and nerv, it was not clustered with cote d'ivoire and guinea bissau (see figure 2). in this year, nigeria’s annual average inflation rate (ir) was at 12.20% as compared to the single-digit inflation rates experienced by cote d'ivoire and guinea bissau. from 2013, nigeria starts to show more similarities with cote d'ivoire and guinea bissau as they demonstrate similar variable characteristics for cbf, ger, tpd, and nerv. these similarities are further intensified as inflation rates stayed relatively stable in the 2013 2015 period. similarities peak in 2014 when all three countries met all the convergence criteria. despite remaining clustered together in 2015, dissimilarities start to appear as nigeria is unable to reduce its central bank financing of budget deficit. this was attributed to a decline in total revenue and grants (ecowas, 2016). in 2016, cote d'ivoire and guinea bissau showed strong similarities in achieving five of the six criteria. both countries were only unable to maintain their ratio of budget deficit to nominal gdp at a level equal to, or lower than, 3%. on the other hand, nigeria was able to achieve bd but underperformed with its inflation rates and nominal exchange rate variations. the increase in the general level of prices for nigeria could be explained by the depreciation of the naira in 2016 (ecowas, 2016). for these cases, some aspects of feedback can be observed in the emergence and unpredictability of the nigerian economy. table 3. configurational output. note: authors computation using excel. variable codes: primary criteria: bd ratio of budget deficit to nominal gdp; ir annual average inflation rate; cbf central bank financing of budget deficit; ger gross external reserve. secondary criteria: tpd ratio of total public debt to gdp; nerv nominal exchange rate variation. uppercase equals achieved criteria and lowercase equals did not achieve criteria. for togo and senegal, both cases shared similar variable characteristics in the achievement of bd, ir, cbf, ger, tpd, and nerv between the years 2012 2014. in 2012, differences start to emerge complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 59 university of bamberg press in the ratio of public debt to gdp across cases. while senegal’s tpd dropped by 6% between 2012 and 2013, togo’s increased by 1.3% (see table 6 in appendix 1.1.). this disparity becomes profound in 2014 when togo’s ratio of debt to gdp rose to 66%. in the subsequent years, this continued to rise moving from 76.8% in 2015 to 79.4% in 2016. according to the world bank-imf joint debt sustainability analysis (2019), togo’s high public debt is due to high deficits, contingent liabilities, and accumulated arrears amongst others. this has left little space to absorb external shocks. in this case, some adverse effects of the ebola outbreak (2013 2015) destabilised the economy, demonstrating some aspects of sensitivity to initial conditions and path dependency in the togolese economy— making it ineffective in its ability to absorb external shocks. highlighting the complexity within wamz, the configurational modelling of cabo verde, gambia, ghana, guinea, sierra leone, and liberia demonstrates complex interactions influencing the clustering of these economies. for cabo verde, gambia, and ghana, the hca results showed that these cases shared strong similarities in 2012 and 2013. this is partially observed in the achievement of bd, ir, ger, and nerv in 2012, and bd and ger in 2013. in the subsequent years, cabo verde starts to diverge after failing to meet the convergence criteria for only tpd. while ghana and gambia remain converged, they continue to show some similarities in bd, cbf, ger, tpd, and nerv. with the escudo pegged to the euro, cabo verde was able to reach all the primary criteria and stabilize its nominal exchange rate variation. however, the economy of cabo verde is driven by tourism and was unable to stabilise its ratio of total public debt to gdp due to the adverse effects of the ebola breakout. nyarko, et. al., (2015) indicated that, in africa, the ebola crisis affected the travel and tourism sector almost more than any other sector. this is also emphasised by rosselló, santana-gallego, and awan (2017), who looked at the impacts of infectious diseases on international tourism. evidently, cabo verde had earlier implemented an imf intervention (2010 2012) to manage domestic government debt, increase international reserves, and promote structural reforms to improve debt management and strengthen the financial sector (cabo verde, 2010). in 2014, gambia also experienced major shortages in rainfall, coupled with the ebola breakout across neighbouring countries resulting economic shocks drastically increased its central financing of budget deficit in the subsequent years. likewise, although the emergence of guinea, sierra leone, and liberia may also suggest partial convergence within the region, these countries are predominately members of the mano river union. nevertheless, the clustering of guinea, sierra leone, and liberia provides a noteworthy case of emergence resulting from an event (nelson and winter, 1982). the configurational model shows inconsistencies in meeting all the convergence criteria across the period under study. yet, these cases start to show strong similarities in the clustering between 2013 2015. during this period, these countries were hit by the ebola epidemic. cdc (2020) statistics suggest that they were the hardest hit countries with the highest death rates. here, the impact of an epidemic can be seen to have had adverse effects on the monetary system of these countries. after the ebola outbreak, guinea, sierra leone, and liberia do not cluster in 2016—to some extent highlighting nationalistic approaches to economic recovery. for burkina faso, mali, benin, and niger, these countries met all the convergence criteria throughout the period under study. a detailed look at the indicators (table 9 and table 10 in appendix 1.1.) would suggest that, over time, they become more similar. for instance, slight variations can be observed across the indicators for bd, ir, and tpd. overall, these patterns provide a path understanding of case variable movement over time—while developing some theories of social change. the configurational modelling also allows for the observation of ‘multiple conjunctural causation’ (ragin, 1987; berg-schlosser, et. al., 2009; haynes, 2017) in the optimum/optimal currency area (oca) theory—which suggests that in satisfying a set of criteria countries begin to show similar economic patterns over time (sigma convergence). this is slightly evident across cfa franc members. for instance, guinea bissau, cote d'ivoire, and nigeria (uses the naira) showed similarities in meeting the convergence criteria in 2013, but these similarities peaked in 2014. adverse effects are also observed when countries are unable to meet the convergence criteria (for instance, guinea bissau, cote d'ivoire, and nigeria in 2015 2016; and togo and senegal in 2014 2015). similarly, for cases where the achievement of all the convergence criteria was met throughout the period under discussion (i.e., burkina faso and mali, and benin and niger) cases begin to show more similarities over time— complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 60 university of bamberg press eventually leading to some form of ‘lock-in’ (unruh, 2000). these observations, to an extent, provide a demonstration of the optimum/optimal currency area (oca) theory. the clustering of cabo verde also provides some aspects of delta convergence. despite its legacy of debt issues, as well as some adverse impacts from the ebola outbreak on its tourism economy, cabo verde showed better resilience in meeting the primary criteria in the subsequent years as compared to gambia and ghana. this, to an extent, can be attributed to the fact that the escudo is pegged to the euro. interestingly, during these years (2014 2016), cabo verde showed more similarities with waemu members (cfa franc is also pegged to the euro). this also suggests aspects of delta convergence (a movement towards an example model—in this case the euro). nevertheless, some systemic spill overs can also be observed when policy actions within one subsystem affects another subsystem (howlett & ramesh, 2002; williams, 2009). this is seen in the form of global travel bans as a result of the ebola outbreak having an impact on cabo verde’s ability to meet a convergence criterion. other complex interactions are also observed in the form of feedback effects in response to events—as with the case of the ebola outbreak in guinea, sierra leone, and liberia; aspects of sensitivity to initial conditions in stable patterns observed between some cfa franc members, and path dependency in togo’s increasing public debt. longitudinal configurational model: in the longitudinal application of pattern synthesis, data can also be coded to further highlight trends in variables and case patterns over time. table 4 (below) shows a recoded configurational model highlighting the consistency/stability in achieving each criterion throughout the period under analysis. the model, to an extent, shows variations in patterns across and within the two monetary systems (reorder version highlighting disparities across the unions). from this, systemic instability can be observed within wamz—when case and variable patterns seem to be unstable over time. with the exception of cabo verde, patterns in variable trends and case patterns for wamz members is relatively unstable. very few countries were able to meet all the convergence criteria during the period under study. in these cases, emergent cluster patterns were not directly observed as a result of achieving the convergence criteria, but rather as a result of other impacts. such changes in case patterns are reflected in the instability of associated variable changes. observing stability in achieving the ecowas convergence criteria: 2012 2016 union country bd ir cbf ger tpd nerv variable pattern case pattern wamz gambia met met not met unstable unstable wamz ghana met wamz cabo verde met met met met not met met stable with some conditions unmet unstable wamz guinea met unstable unstable wamz sierra leone met met met wamz liberia met met met met wamz nigeria met met met unstable unstable waemu cote d'ivoire met met met met met largely stable stable waemu guinea bissau met met met met met waemu burkina faso met met met met met met stable initially unstable but convergence was achieved in 2015/2016 waemu mail met met met met met met waemu benin met met met met met met waemu niger met met met met met met waemu senegal met met met met met met largely stable initially stable but diverged after 2014 waemu togo met met met met met table 4. longitudinal configurational output (authors computation using excel). complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 61 university of bamberg press however, across waemu members, some form of stable dynamics can be observed—where variables and case patterns tend to remain stable over time. variable patterns seem to be largely stable with the achievement of ir, cbf, ger, and nerv across all members throughout the period under discussion. a few members show inconsistencies in the achievement of bd (cote d'ivoire and guinea bissau) and tpd (togo). in the waemu, members also tend to remain clustered over longer periods and in some cases, the same variables are associated with this clustering. for waemu members, the emergence of case patterns can be somewhat associated with the attainment of the convergence criteria or to the fact that these countries also have the pre-existing unified institutional and policy structures needed to ensure policy alignment. in the application of pattern synthesis, additional qualitative data can be used to interpret counterfactuals. conclusion: observing system change as patterns complexity theory provides a multidimensional perspective to the assessment of monetary convergence and offers a better understanding and observation of trajectories over time. case-based methods, such as dynamic pattern synthesis, allow for the observation of case changes as well as the variable movements associated with such changes. these patterns are usually unpredictable, resulting from interactions within a given context and at a specific point in time. over time, identifying these patterns offer some degree of system predictability based on the experiences of cases. in its longitudinal application, repeating dynamic pattern synthesis across the same dataset ensures that the method theorises social change, thereby assessing changes against the irreversible dimension of time. this allows for observation of multiple conjunctural causation and consideration of the impact of events across the monetary zone. combined with additional qualitative data, and knowledge of the context and cases, a researcher can interpret these patterns. for instance, in this study, the findings suggest more efforts need to be focused on convergence across wamz members. this is evident in the unstable nature of variables and case patterns across these economies. with exception of cabo verde, no wamz member was able to meet all the primary convergence criteria during the period under discussion. this highlights the need for stronger institutional and regulatory alignment/compliance across wamz members—or what buti and turrini (2015) would term ‘structural convergence.’ if this is targeted towards a focus on encouraging the implementation of structural reforms to enhance technical preparedness, it is more probable to create stable patterns within the region—as with the case of waemu members. the dps method also allows for the exploration of exceptional cases. for instance, in the study, it becomes evident that cabo verdes’ resilience in meeting all the primary criteria can be attributed to the fact that its currency is pegged to the euro. this could add to the debate on the use of fixed/floating exchange rates within the region to minimise risks amongst vulnerable economies (nathaniel, oladiran & oladiran, 2019). to strengthen regulatory alignment and encourage stable patterns, especially in cases of negative events (such as the covid-19 pandemic), national actors can consider introducing fixed exchange rates—as against the current floating mechanism used amongst most wamz members—to aid in improving alignment and accelerating convergence. however, some adverse effects of this can, to an extent, become evident in less diversified economies—as reflected in cabo verde’s dependence on tourism and growth in public debt. external events also impact the convergence situation during the period under discussion. in such instances, utilising complexity theory, to an extent, can aid in predicting possible impacts of social change on aspects of a system over time. evidently, while orthodox approaches may observe convergence as a ‘growing resemblance’—identifying processes of catching up, growing similarities, or movement towards exemplary models (heichel, et. at., 2005). this research demonstrates that, although convergence may occur over time, case and variable movement may be haphazard and irregular. these findings and methodological applications highlight some complexities observed within the ecowas monetary region. this exposes areas of weaknesses or policy impacts within the system and can assist in the development of policy responses to support integration (such as the post-covid19 economic recovery), and in the long run, minimise individual risk of asymmetry shocks within the complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 62 university of bamberg press region. other areas of possible application could focus on integration into the wider continental agreements (e.g., the african continental free trade area agreement). references abban, s. (2020). on the computation and essence of the nominal convergence criteria for africa currency union: ecowas in perspective. munich personal repec archive. alagidede, p., coleman, s., & cuestas, j. c. (2012). inflationary shocks and common economic trends: implications for west african monetary union membership. journal of policy modeling, 34(3), 460-475. baumgartner, f. r., & jones, b. d. (2002). positive and negative feedback in politics. policy dynamics, 3-28. boariu, a., & bilan, i. (2007). inflationary effects of budget deficit financing in contemporary economies. analele þtiinþifice ale universitãþii, “alexandru ioan cuza” din iaþitomul liv. bovaird, t. (2008). emergent strategic management and planning mechanisms in complex adaptive systems: the case of the uk best value initiative. public management review, 10(3), 319-340. bukowski, s. (2006). the maastricht convergence criteria and economic growth in the emu. quaderni del dipartimento di economia, finanza e statistica, 24, 1-19. buti, m. and turrini, a. (2015). three waves of convergence. can eurozone countries start growing together again? vox cepr’s policy portal. byrne, d. (2002). interpreting quantitative data. sage. byrne, d. s. (1998). complexity theory and the social sciences: an introduction. psychology press. cabo verde. (2010). letter of intent, memorandum of economic and financial policies, and technical memorandum of understanding. retrieved 3 august 2017, from. https://www.imf.org/external/np/cpid/default.aspx. carmignani, f. (2003). political instability, uncertainty and economics. journal of economic surveys, 17(1), 1-54. cairney, p. (2012). complexity theory in political science and public policy. political studies review, 10(3), 346-358. cdc (2020). 2014-2016 ebola outbreak in west africa | history | ebola (ebola virus disease) . retrieved 27 october 2020, from https://www.cdc.gov/vhf/ebola/history/2014-2016-outbreak/index.html. cilliers, p. (1998) complexity and postmodernism understanding complex systems. london routledge. cilliers, p. (2001). boundaries, hierarchies and networks in complex systems. international journal of innovation management, 5(02), 135-147. correia, a. f., cunha, v. g., heitor, f., maria, j. r., & saramago, l. (2007). exchange rate cooperation agreement between portugal and cabo verde: characterisation, developments and challenges after 20 years. future, 1. de grauwe, p. (2000). monetary policies in the presence of asymmetries. jcms: journal of common market studies, 38(4), 593-612. de grauwe, p. (2020). economics of the monetary union. oxford university press, usa. dolowitz, d. p., & marsh, d. (2000). learning from abroad: the role of policy transfer in contemporary policy‐ making. governance, 13(1), 5-23. drezner, d. w. (2001). globalization and policy convergence. international studies review, 3(1), 53-78. economic community of west african states. (2016). 2015 ecowas convergence report. abuja: ecowas. economic community of west african states. (2017). 2016 ecowas convergence report. abuja: ecowas. ezekwesili, c. e. (2011). can the monetary integration of ecowas improve intra-regional trade? fielding, d., lee, k., & shields, k. (2004). the characteristics of macroeconomic shocks in the cfa franc zone. journal of african economies, 13(4), 488-517. flier, b., bosch, f. a. v. d., & volberda, h. w. (2003). co‐evolution in strategic renewal behaviour of british, dutch and french financial incumbents: interaction of environmental selection, institutional effects and managerial intentionality. journal of management studies, 40(8), 2163-2187. foxon, t. j. (2011). a coevolutionary framework for analysing a transition to a sustainable low carbon economy. ecological economics, 70(12), 2258-2267. friedman, m. (1997). the role of monetary policy american economic review (1968) 58, march, pp. 1–17. in a macroeconomics reader (pp. 176-191). routledge. friendly, m. (2002). “corrgrams: exploratory displays for correlation matrices.” the american statistician 56 (4): 316–24. hammouda, h. b., karingi, s. n., njuguna, a. e., & jallab, m. s. (2009). why doesn't regional integration improve income convergence in africa? african development review, 21(2), 291-330. haynes, p. (2017). social synthesis: finding dynamic patterns in complex social systems. routledge. haynes, p. (2018). complex policy planning: the government strategic management of the social care market. routledge. haynes, p., & haynes, j. (2016). convergence and heterogeneity in euro based economies: stability and dynamics. economics. 4(3), 16. held, d. (1997). democracy and globalization. global governance: a review of multilateralism and international organizations, 3(3), 251-267. heichel, s., pape, j., & sommerer, t. (2005). is there convergence in convergence research? an overview of empirical studies on policy convergence. journal of european public policy, 12(5), 817-840. hm treasury (2020) the magenta book: supplementary guide. handling complexity in policy evaluation. london: hmt houssa, r. (2008). monetary union in west africa and asymmetric shocks: a dynamic structural factor model approach. journal of development economics, 85(1-2), 319-347. howlett, m., & ramesh, m. (2002). the policy effects of internationalization: a subsystem adjustment analysis of policy complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 https://www.imf.org/external/np/cpid/default.aspx https://www.cdc.gov/vhf/ebola/history/2014-2016-outbreak/index.html 63 university of bamberg press change. journal of comparative policy analysis, 4(1), 31-50. hsu, h.f. (2010). is a common currency area feasible for east asia? a multivariate structural vector autoregression approach. asian economic journal, 24(4), pp. 391-411. jones, t., & newburn, t. (2002). the transformation of policing? understanding current trends in policing systems. the british journal of criminology, 42(1), 129-146. kanga, d., murinde, v., & soumaré, i. (2020). capital, risk and profitability of waemu banks: does bank ownership matter? journal of banking & finance, 105814. klijn, e. h. (2008). governance and governance networks in europe: an assessment of ten years of research on the theme. public management review, 10(4), 505-525. knill, c. (2005). introduction: cross-national policy convergence: concepts, approaches and explanatory factors. journal of european public policy, 12(5), 764-774. laidler, d. (1971). some evidence of demand for money. journal of political economics, vol.74. martínez-garcía, m., & hernández-lemus, e. (2013). health systems as complex systems. american journal of operations research, vol. 3 no. 1a, 2013, pp. 113-126. doi: 10.4236/ajor.2013.31a011. mati, s., civcir, i., & ozdeser, h. (2019). ecowas common currency: how prepared are its members? investigación económica, 78(308), 89-119. maturana, h. (1980). introduction to autopoiesis and cognition: the realization of the living london; d. mcevoy, p., & richards, d. (2006). a critical realist rationale for using a combination of quantitative and qualitative methods. journal of research in nursing, 11(1), 66-78. mogaji, p. k. (2017). evaluation of macroeconomic indicators and dynamics for monetary integration of west africa: the case of the wamz. mundell, r. a. (1961). a theory of optimum currency areas. the american economic review, 51(4), 657-665. nanz, p., & steffek, j. (2004). global governance, participation and the public sphere. government and opposition, 39(2), 314 335. nathaniel, o. o., oladiran, o. i., & oladiran, a. t. (2019). impact of exchange rate regimes on economic integration in the ecowas (1980-2017). african journal of economic review, 7(2), 42-59. nkwatoh, l. s. (2018). does ecowas macroeconomic convergence criteria satisfy an optimum currency area? journal of economics and management sciences, 1(2), p61-p61. nyarko, y., goldfrank, l., ogedegbe, g., soghoian, s., aikins, a. d. g., & nyu-ug-kbth, g. e. w. (2015). preparing for ebola virus disease in west african countries not yet affected: perspectives from ghanaian health professionals. globalization and health, 11(1), 7. obi, k. o., & uzodigwe, a. a. (2015). dynamic impact of money supply on inflation: evidence from ecowas member states. okafor, g. (2017). the impact of political instability on the economic growth of ecowas member countries. defence and peace economics, 28(2), 208-229. oladipo, s. o., & akinbobola, t. o. (2011). budget deficit and inflation in nigeria: a causal relationship. journal of emerging trends in economics and management sciences, 2(1), 1-8. osberg, d., biesta, g., & cilliers, p. (2008). from representation to emergence: complexity’s challenge to the epistemology of schooling. educational philosophy and theory, 40(1), 213–227. pierson, p. (2000). increasing returns, path dependence, and the study of politics. american political science review, 251-267. pierson, p. (2004). positive feedback and path dependence. p. pierson, politics in time. history, institutions, and social analysis, 17-53. plasmans, j. e., engwerda, j., van aarle, b., di bartolomeo, g., & michalak, t. (2006). dynamic modeling of monetary and fiscal cooperation among nations (vol. 8). springer science & business media. ragin, c. (1987). the comparative method: moving beyond qualitative and quantitative methods. berkeley: university of california. ragin, c. c. (1997). turning the tables: how case-oriented methods challenge variable-oriented methods. comparative social research, 16(1), 27-42. ragin, c. c., & becker, h. s. (eds.). (1992). what is a case?: exploring the foundations of social inquiry. cambridge university press. ragin, c. c., & rihoux, b. (eds.). (2009). configurational comparative methods: qualitative comparative analysis (qca) and related techniques. sage. regmi, k., nikolsko-rzhevskyy, a., and thornton, r. (2015). to be or not to be: an optimum currency area for south asia? journal of policy modeling, 37(6), pp. 930-944. rihoux, b., rezsöhazy, i., & bol, d. (2011). qualitative comparative analysis (qca) in public policy analysis: an extensive review. german policy studies, 7(3), 9-82. rose, r. (1991). what is lesson-drawing?. journal of public policy, 3-30. rosen, r. (1987). some epistemological issues in physics and biology. quantum implications: essays in honour of david bohm, 314-327. rosselló, j., santana-gallego, m., & awan, w. (2017). infectious disease risk and international tourism demand. health policy and planning, 32(4), 538-548. sheppard, d. k., and duck, n. w.,(1978). a proposal for the control of the uk money supply. the economic journal, 88(349), 1-17. smith, j., & jenks, c. (2006). qualitative complexity: ecology, cognitive processes and the re-emergence of structures in post humanist social theory. routledge. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 64 university of bamberg press smith, l. b. (2005). cognition as a dynamic system: principles from embodiment. developmental review, 25(3-4), 278-298. solomon, m., & de wet, w. a. (2004). the effect of a budget deficit on inflation: the case of tanzania. south african journal of economic and management sciences, 7(1), 100-116. stacey, r. d. (1996). complexity and creativity in organizations. berrett-koehler publishers. stone, d. (2008). global public policy, transnational policy communities, and their networks. policy studies journal, 36(1), 19 38. taylor, l., haynes, p., & darking, m. (2020). english local government finance in transition: towards the ‘marketization of income’. public management review, 1-26. unruh, g. c. (2000). understanding carbon lock-in. energy policy, 28(12), 817-830. williams, r. a. (2009). exogenous shocks in subsystem adjustment and policy change: the credit crunch and canadian banking regulation. journal of public policy, 29(1), 29-53. weible, c. m. (2014). advancing policy process research. theories of the policy process, 3, 391-408. world bank; international monetary fund. (2019). togo joint world bank-imf debt sustainability analysis. world bank, washington, dc. © world bank. https://openknowledge.worldbank.org/handle/10986/32564 license: cc by 3.0 igo. zelli, f., gerrits, l., & möller, i. (2021). global governance in complex times: exploring new concepts and theories on institutional complexity. complexity, governance & networks, 6(1), 1-13. annex annex 1.1. configurational model for emergent clusters note: for cases where the achievement of the convergence criteria is met, data is shaded in the colour of the pattern in the hca results. likewise, in cases where this is demonstrated across a set of convergent cases, indicators are in “bold” and “italic.” variable codes: primary criteria: bd ratio of budget deficit to nominal gdp; ir annual average inflation rate; cbf central bank financing of budget deficit; ger gross external reserve. secondary criteria: tpd ratio of total public debt to gdp; nerv nominal exchange rate variation. bd_2012 ir_2012 cbf_2012 ger_2012 tpd_2012 nerv_2012 nigeria -1.40 12.20 0 8.50 12.60 0.70 cote d'ivoire 3.20 1.30 0 5.30 34.20 -4.80 guinea bissau 2.10 2.10 0 5.30 52.40 -4.80 bd_2013 ir_2013 cbf_2013 ger_2013 tpd_2013 nerv_2013 nigeria -1.3 8.5 0 8.9 12.4 2.1 cote d'ivoire 2.2 2.6 0 4.7 34 4.1 guinea bissau 3.4 0.7 0 4.7 52.6 4.1 bd_2014 ir_2014 cbf_2014 ger_2014 tpd_2014 nerv_2014 nigeria -1 8 0 6 12.5 -1.9 cote d'ivoire 2.2 0.4 0 5 36.9 0.1 guinea bissau 2.6 -1 0 5 53.3 0.1 bd_2015 ir_2015 cbf_2015 ger_2015 tpd_2015 nerv_2015 nigeria -1.5 9 13.1 8.2 12.6 -1.9 cote d'ivoire 2.9 1.2 0 5 40.8 -9.3 guinea bissau 2.7 1.4 0 5 46.8 -9.3 bd_2016 ir_2016 cbf_2016 ger_2016 tpd_2016 nerv_2016 nigeria -2.2 15.7 0 5.8 17.1 -23.5 cote d'ivoire 3.9 0.7 0 4.4 42.1 0.5 guinea bissau 4 1.5 0 4.4 46.1 0.5 table 5. configurational model for nigeria, cote d’ivoire, and guinea bissau complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 65 university of bamberg press bd_2012 ir_201 2 cbf_201 2 ger_201 2 tpd_201 2 nerv_201 2 senegal -5.80 1.40 0 5.30 36.70 -4.80 togo -5.80 2.60 0 5.30 44.00 -4.80 bd_2013 ir_201 3 cbf_201 3 ger_201 3 tpd_201 3 nerv_201 3 senegal -5.5 0.7 0 4.7 30.7 4.1 togo -4.6 1.8 0 4.7 45.3 4.1 bd_2014 ir_201 4 cbf_201 4 ger_201 4 tpd_201 4 nerv_201 4 senegal -5.2 1.1 0 5 35.4 0.1 togo -3.4 0.2 0 5 66.9 0.1 bd_2015 ir_201 5 cbf_201 5 ger_201 5 tpd_201 5 nerv_201 5 senegal -4.8 0.1 0 5 29.1 -9.3 togo -6.3 1.8 0 5 76.8 -9.3 bd_2016 ir_201 6 cbf_201 6 ger_201 6 tpd_201 6 nerv_201 6 senegal -4.2 0.8 0 5.79 55.7 0.5 togo -8.5 0.9 0 4.4 79.4 0.5 table 6. configurational model for senegal and togo bd_2012 ir_2012 cbf_2012 ger_2012 tpd_2012 nerv_2012 ghana -5.70 9.10 25.40 3.00 47.80 -4.40 cabo verde -12.40 2.50 0 4.00 91.10 -4.00 gambia -4.60 4.30 0.40 4.80 78.00 -4.50 bd_2013 ir_2013 cbf_2013 ger_2013 tpd_2013 nerv_2013 ghana -8.6 11.7 12.3 3.1 56.8 -7.4 cabo verde -8.8 1.5 0 4.9 102.5 4.1 gambia -8.7 5.7 0 4.6 88.1 -10.3 bd_2014 ir_2014 cbf_2014 ger_2014 tpd_2014 nerv_2014 ghana -6.4 17 13.7 3 70.2 -31.5 cabo verde -7.2 -0.2 0 5.4 115 0.1 gambia -9.6 6.9 40.8 3.7 104.1 -16.5 bd_2015 ir_2015 cbf_2015 ger_2015 tpd_2015 nerv_2015 ghana -4.8 17.2 4.1 2.6 73.2 -15.7 cabo verde -3.9 0.1 0 6.4 126.1 -9.3 gambia -6.3 6.8 41.5 2.5 101.1 4.9 bd_2016 ir_2016 cbf_2016 ger_2016 tpd_2016 nerv_2016 ghana -10.9 17.5 0 2.8 73.1 -4.2 cabo verde -3.5 -1.4 0 6.6 128.6 0.5 gambia -9.5 7.9 33.1 2.4 117.3 -3.3 table 7. configurational model for cluster ghana, cabo verde, and gambia complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 66 university of bamberg press bd_2012 ir_2012 cbf_2012 ger_2012 tpd_2012 nerv_2012 liberia 7.50 6.90 0 2.80 34.10 1.30 guinea 3.20 15.20 0 2.40 42.20 -2.50 sierra leone -5.20 12.90 37.70 3.40 36.70 3.30 bd_2013 ir_2013 cbf_2013 ger_2013 tpd_2013 nerv_2013 liberia -0.5 7.6 0 2.8 30.5 -4.1 guinea -2 11.9 0.01 2.9 44.5 2.1 sierra leone -1.6 10.4 1.7 3.2 30.8 1.1 bd_2014 ir_2014 cbf_2014 ger_2014 tpd_2014 nerv_2014 liberia 0.2 9.8 0 2.5 37.9 -9 guinea -3.56 9.7 0 3.2 73.5 -1.5 sierra leone -3.3 7.2 7.2 3.6 35.4 -4 bd_2015 ir_2015 cbf_2015 ger_2015 tpd_2015 nerv_2015 liberia 1.6 7.8 0 2.3 32 7.2 guinea -6.9 8.2 0.26 2.2 43.3 2.2 sierra leone -4.1 8.1 -0.7 3.8 29.1 -3.1 bd_2016 ir_2016 cbf_2016 ger_2016 tpd_2016 nerv_2016 liberia 2.2 8.8 0 3.3 36.7 -8.4 guinea 0.1 8.2 0.01 1.4 43.1 -16.4 sierra leone -6.4 10.8 33.1 4.7 55.7 -19.1 table 8. configurational model for liberia, guinea, and sierra leone bd_2012 ir_2012 cbf_2012 ger_2012 tpd_2012 nerv_2012 benin -0.40 6.80 0 5.30 26.80 -4.80 niger -1.10 0.50 0 5.30 18.80 -4.80 bd_2013 ir_2013 cbf_2013 ger_2013 tpd_2013 nerv_2013 benin -2.6 1 0 4.7 25.4 4.1 niger -2.6 2.3 0 4.7 23.1 4.1 bd_2014 ir_2014 cbf_2014 ger_2014 tpd_2014 nerv_2014 benin -1.9 -1.1 0 5 30.9 0.1 niger -8.1 0.9 0 5 25.6 0.1 bd_2015 ir_2015 cbf_2015 ger_2015 tpd_2015 nerv_2015 benin -8 0.3 0 5 42.4 -9.3 niger -9 1 0 5 36 -9.3 bd_2016 ir_2016 cbf_2016 ger_2016 tpd_2016 nerv_2016 benin -6.2 -0.8 0 4.4 49.4 0.5 niger -6.1 0.2 0 4.4 39.7 0.5 table 9. configurational model for benin and niger complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 67 university of bamberg press bd_2012 ir_2012 cbf_2012 ger_2012 tpd_2012 nerv_2012 burkina faso -3.10 3.80 0 5.30 27.96 4.78 mali -0.10 5.30 0 5.30 24.30 4.80 bd_2013 ir_2013 cbf_2013 ger_2013 tpd_2013 nerv_2013 burkina faso -3.58 0.5 0 4.7 28.58 4.12 mali -2.2 0.6 0 4.7 26 4.1 bd_2014 ir_2014 cbf_2014 ger_2014 tpd_2014 nerv_2014 burkina faso -1.85 -0.3 0 5 30.79 0.09 mali -3.8 0.9 0 5 27.1 0.1 bd_2015 ir_2015 cbf_2015 ger_2015 tpd_2015 nerv_2015 burkina faso -2.03 0.9 0 5 32.75 9.33 mali -1.8 1.5 0 5 30.8 9.3 bd_2016 ir_2016 cbf_2016 ger_2016 tpd_2016 nerv_2016 burkina faso -3.14 -0.2 0 4.4 34.23 0.5 mali -3.9 1.8 0 4.4 36 0.5 table 10. configurational model for burkina faso and mali annex 1.2. correlation plot for 2012 to 2015 note: correlograms display correlation coefficients for variable pairs. pairs with more intense colours have more extreme correlations. for insignificant correlations (not significantly different from 0), these are represented by a white box. for more details, see friendly (2002). variable codes: primary criteria: bd – ratio of budget deficit to nominal gdp; ir – annual average inflation rate; cbf – central bank financing of budget deficit; ger – gross external reserve. secondary criteria: tpd – ratio of total public debt to gdp; nerv – nominal exchange rate variation. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 68 university of bamberg press annex 1.2. correlation plot for 2016 (correlograms – see friendly, 2002) (continued) complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 69 university of bamberg press annex 1.3. trends in convergence indicators complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 70 university of bamberg press complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 71 university of bamberg press annex 1.3. trends in convergence indicators (continued) complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 72 university of bamberg press annex 1.3. trends in convergence indicators (continued) complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 73 university of bamberg press complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 74 university of bamberg press complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 75 university of bamberg press appendix 1.4. ecowas convergence situation 2012 2016 dataset7 variable codes: primary criteria: bd ratio of budget deficit to nominal gdp; ir annual average inflation rate; cbf central bank financing of budget deficit; ger gross external reserve. secondary criteria: tpd ratio of total public debt to gdp; nerv nominal exchange rate variation. 7 source: economic community of west african states. (2017). 2016 ecowas convergence report. abuja: ecowas. country b d _ 2 0 1 2 ir _ 2 0 1 2 c b f _ 2 0 1 2 g e r _ 2 0 1 2 t p d _ 2 0 1 2 n e r v _ 2 0 1 2 benin 0.40 6.80 0 5.30 26.80 4.80 burkina faso 3.10 3.80 0 5.30 27.96 4.78 cabo verde -12.40 2.50 0 4.00 91.10 4.00 cote d'ivoire 3.20 1.30 0 5.30 34.20 4.80 gambia 4.60 4.30 0.40 4.80 78.00 4.50 ghana 5.70 9.10 25.40 3.00 47.80 4.40 guinea 3.20 15.20 0 2.40 42.20 2.50 guinea bissau 2.10 2.10 0 5.30 52.40 4.80 liberia 7.50 6.90 0 2.80 34.10 1.30 mail 0.10 5.30 0 5.30 24.30 4.80 niger 1.10 0.50 0 5.30 18.80 4.80 nigeria 1.40 12.20 0 8.50 12.60 0.70 senegal 5.80 1.40 0 5.30 36.70 4.80 sierra leone 5.20 12.90 37.70 3.40 36.70 3.30 togo 5.80 2.60 0 5.30 44.00 4.80 complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 76 university of bamberg press appendix 1.4. ecowas convergence situation 2012 2016 dataset (continued)8 country bd_2013 ir_2013 cbf_2013 ger_2013 tpd_2013 nerv_2013 benin -2.6 1 0 4.7 25.4 4.1 burkina faso -3.58 0.5 0 4.7 28.58 4.12 cabo verde -8.8 1.5 0 4.9 102.5 4.1 cote d'ivoire 2.2 2.6 0 4.7 34 4.1 gambia -8.7 5.7 0 4.6 88.1 -10.3 ghana -8.6 11.7 12.3 3.1 56.8 -7.4 guinea -2 11.9 0.01 2.9 44.5 2.1 guinea bissau 3.4 0.7 0 4.7 52.6 4.1 liberia -0.5 7.6 0 2.8 30.5 -4.1 mail -2.2 0.6 0 4.7 26 4.1 niger -2.6 2.3 0 4.7 23.1 4.1 nigeria -1.3 8.5 0 8.9 12.4 2.1 senegal -5.5 0.7 0 4.7 30.7 4.1 sierra leone -1.6 10.4 1.7 3.2 30.8 1.1 togo -4.6 1.8 0 4.7 45.3 4.1 variable codes: primary criteria: bd ratio of budget deficit to nominal gdp; ir annual average inflation rate; cbf central bank financing of budget deficit; ger gross external reserve. secondary criteria: tpd ratio of total public debt to gdp; nerv nominal exchange rate variation. 8 source: economic community of west african states. (2017). 2016 ecowas convergence report. abuja: ecowas. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 77 university of bamberg press appendix 1.4. ecowas convergence situation 2012 2016 dataset (continued) c o u n tr y b d _ 2 0 1 4 ir _ 2 0 1 4 c b f _ 2 0 1 4 g e r _ 2 0 1 4 t p d _ 2 0 1 4 n e r v _ 2 0 1 4 b d _ 2 0 1 5 ir _ 2 0 1 5 c b f _ 2 0 1 5 g e r _ 2 0 1 5 t p d _ 2 0 1 5 n e r v _ 2 0 1 5 benin -1.9 -1.1 0 5 30.9 0.1 -8 0.3 0 5 42.4 -9.3 burkina faso -1.85 -0.3 0 5 30.79 0.09 -2.03 0.9 0 5 32.75 9.33 cabo verde -7.2 -0.2 0 5.4 115 0.1 -3.9 0.1 0 6.4 126.1 -9.3 cote d'ivoire 2.2 0.4 0 5 36.9 0.1 2.9 1.2 0 5 40.8 -9.3 gambia -9.6 6.9 40.8 3.7 104.1 -16.5 -6.3 6.8 41.5 2.5 101.1 4.9 ghana -6.4 17 13.7 3 70.2 -31.5 -4.8 17.2 4.1 2.6 73.2 -15.7 guinea -3.56 9.7 0 3.2 73.5 -1.5 -6.9 8.2 0.26 2.2 43.3 2.2 guinea bissau 2.6 -1 0 5 53.3 0.1 2.7 1.4 0 5 46.8 -9.3 liberia 0.2 9.8 0 2.5 37.9 -9 1.6 7.8 0 2.3 32 7.2 mail -3.8 0.9 0 5 27.1 0.1 -1.8 1.5 0 5 30.8 9.3 niger -8.1 0.9 0 5 25.6 0.1 -9 1 0 5 36 -9.3 nigeria -1 8 0 6 12.5 -1.9 -1.5 9 13.1 8.2 12.6 -1.9 senegal -5.2 1.1 0 5 35.4 0.1 -4.8 0.1 0 5 29.1 -9.3 sierra leone -3.3 7.2 7.2 3.6 35.4 -4 -4.1 8.1 -0.7 3.8 29.1 -3.1 togo -3.4 0.2 0 5 66.9 0.1 -6.3 1.8 0 5 76.8 -9.3 variable codes: primary criteria: bd ratio of budget deficit to nominal gdp; ir annual average inflation rate; cbf central bank financing of budget deficit; ger gross external reserve. secondary criteria: tpd ratio of total public debt to gdp; nerv nominal exchange rate variation. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 78 university of bamberg press appendix 1.4. ecowas convergence situation 2012 2016 dataset (continued) variable codes: primary criteria: bd ratio of budget deficit to nominal gdp; ir annual average inflation rate; cbf central bank financing of budget deficit; ger gross external reserve. secondary criteria: tpd ratio of total public debt to gdp; nerv nominal exchange rate variation. country b d _ 2 0 1 6 ir _ 2 0 1 6 c b f _ 2 0 1 6 g e r _ 2 0 1 6 t p d _ 2 0 1 6 n e r v _ 2 0 1 6 benin -6.2 -0.8 0 4.4 49.4 0.5 burkina faso -3.14 -0.2 0 4.4 34.23 0.5 cabo verde -3.5 -1.4 0 6.6 128.6 0.5 cote d'ivoire 3.9 0.7 0 4.4 42.1 0.5 gambia -9.5 7.9 33.1 2.4 117.3 -3.3 ghana -10.9 17.5 2.8 73.1 -4.2 guinea 0.1 8.2 0.01 1.4 43.1 -16.4 guinea bissau 4 1.5 0 4.4 46.1 0.5 liberia 2.2 8.8 0 3.3 36.7 -8.4 mail -3.9 1.8 0 4.4 36 0.5 niger -6.1 0.2 0 4.4 39.7 0.5 nigeria -2.2 15.7 0 5.8 17.1 -23.5 senegal -4.2 0.8 0 5.79 55.7 0.5 sierra leone -6.4 10.8 33.1 4.7 55.7 -19.1 togo -8.5 0.9 0 4.4 79.4 0.5 complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 50-78 doi: http://dx.doi.org/10.20377/cgn108 measuring risks of organizational failure in contract exchange structures complexity, governance & networks-vol 02, issue 01 (2015) 45–64 45 doi: 10.7564/13-cgn01 measuring risks of organizational failure in contract exchange structures julia l. carboni indiana university purdue university indianapolis and iu lilly family school of philanthropy e-mail: jcarboni@iupui.edu governmental agencies increasingly contract out service delivery, but have little understanding of the systemic risk associated with dependence on contractors in service delivery networks. in this study, affiliation network concepts are used to develop a structural index of government dependence on actors in service delivery networks without joint service delivery. networks include direct links to government funders and indirect links among contracted programs based on shared parent organizations. understanding the structure of governance arrangements has practical implications for governance, in terms of understanding government dependence on a particular contractor and the risk associated with organizational failure of those contractors. governance structures may also influence individual incentives to perform. this study makes two contributions to the governance literature. first, the study offers conceptualization of contracted programs as networked structures of exchange even where joint production does not occur. second, an index is proposed to capture the position of individual organizations in complex networks of exchange with government. this index can be used to determine the dependence of government on a specific actor in the service delivery network along with the risk to the network should a particular organization fail. this index can also be incorporated into statistical models of contract performance. keywords: contracting, networked exchange, affiliation networks, governance structures. 1. introduction cross sector collaboration is increasingly used as a governance tool to implement public policies, particularly to deliver public services (ansell & gash, 2008; bryson, crosby, & stone, 2006; emerson, nabatchi, & balogh, 2012). governance theories commonly emphasize both management and organization of service delivery can influence performance (hill & lynn, 2004; lynn, heinrich, & hill, 2000, 2001; milward & provan, 2000). however, the extant literature does not consider how government dependencies on contractors in service delivery networks may place the entire service delivery system at risk for failure (see carboni & milward, 2012, for an exception). instead, scholarship 53559_13-1.indd 45 18/08/15 5:31 pm 46 j. l. carboni / measuring risks of organizational failure in contract exchange structures emphasizes network processes, structure and client outcomes of joint service delivery networks without considering risks for network failures (see agranoff, 2008; agranoff & mcguire, 2003; isett, mergel, leroux, mischen, & rethemeyer, 2011; isett & provan, 2005; milward & provan, 2000; milward, provan, fish, isett, & huang, 2009; popp, milward, mackean, casebeer, & lindstrom, 2014; provan & milward 1995). the contract management literature emphasizes contract design and management in dyadic relationships between government funders and principals but fails to account for the larger structure of exchange in which contracts are embedded (brown & potoski, 2003; brown, potoski & van slyke, 2007; fernandez, 2007; malatesta & smith, 2014; van slyke, 2007). scholarship on service delivery networks and contract management are largely independent of each other. in some cases, networked structures of exchange without joint production arise. these arrangements do not necessitate joint service delivery but exchanges are linked by contracts with a common funder in pursuit of a common public policy goal. a government funder may contract with multiple providers to produce the same service without interacting with each other. this arrangement takes on features of a service delivery network where the interactions between the funder and a particular contractor may influence other interactions in the network. in particular, the presence of substitutable alternative contractors for government in the networked structure of exchange may provide a buffer for government because government can replace them with a similar contractor if necessary. if a governmental agency contracts with a single organization to produce multiple programs for a single client population, the agency increases its dependence on that particular organization. assigning programs to multiple organizations reduces the dependence of the agency on a single organization. if a contractor the agency depends on fails, clients may be left without services and the functioning of the network may be jeopardized (carboni & milward, 2012). for example, if a large social service contractor fails, clients would be left without services if other contractors in the network could not absorb them. this jeopardizes government ability to provide legally mandated services. to understand the risk of dependence associated with contracting in service delivery networks, we must be able to measure the network structure in a way that captures complex interdependencies among elements of the system, including the direct connections among government funders and contracted programs and interdependencies among contractors that are not directly related to exchange. this might include separate programs for the same funder being administered by the same parent organization, thus creating a connection among programs even though if they are operated separately. this underlying affiliation changes the structure of dependence for governmental agencies. instead of relying on two autonomous organizations to provide two independent programs, it is reliant on a single organization to provide two independent programs. relationships among contracted programs may be obscure. for example, a parent organization might run several contracted programs with different names, making it less clear the programs have a connection. understanding the structure of affiliations in these exchange networks will increase our understanding of government dependence and risks to service production 53559_13-1.indd 46 18/08/15 5:31 pm j. l. carboni / measuring risks of organizational failure in contract exchange structures 47 if organizations fail. for example, if an organization that administers 30 programs fails, there will be greater disruption to service than if an organization responsible for one program fails. to date, scholars have not proposed a satisfactory way to simultaneously operationalize and measure direct connections and interdependencies among service delivery programs in service delivery networks. the primary contribution of this paper is to develop an index that accounts for both direct and indirect connections among programs in service delivery networks simultaneously. in the next section, i characterize contract arrangements as networked structures of exchange, followed by a discussion of affiliation networks. affiliation networks are a special type of network that can account for links among multiple types of actors. then i develop an index that reflects the dependence of governmental agencies on specific organizations in a network. this measure indicates how large the effect of failure would be on the network. i apply the measure to develop an index of risk for a state funded service delivery network and discuss its implications for organizational failure in that service delivery network. finally, i discuss the implications of this index for governance and along with additional applications of the index. the index sheds light on the structure of dependence among government funders and contractors and can be incorporated into models that predict contract performance. 2. contracts as networked structures of exchange studies in the network and collaboration literature typically focus on collaborative aspects of joint production and use properties of networks as variables of interest (see agranoff & mcguire, 2003; ansell & gash, 2008; bryson, crosby, & stone, 2006; huang & provan, 2007; isett et al., 2011; isett & provan, 2005; milward & provan, 2000; milward, provan, fish, isett, & huang, 2009; provan, huang, & milward, 2009; provan & milward, 1995; o’leary, gerard, & bingham, 2006). conversely, the contract literature emphasizes the management of dyadic interactions to induce contractor performance specific to contracts, but pays little attention to the structure of wider contract arrangements or potential for organizational failure. because governmental agencies often contract for services in suboptimal situations (e.g. thin markets) and for reasons other than efficiency (ferris & graddy, 1991; smith, 1996; smith & smyth, 1996), the contract management literature emphasizes how to design, administer, and manage individual contracts to ensure performance (brown & potoski, 2003a, 2003b; brown, potoski, & van slyke, 2007, 2010; brudney, fernandez, ryu, & wright, 2005; fernandez, 2007; hefetz & warner, 2004; hodge, 2000; romzek & johnson, 2002, 2005). when a governmental agency contracts with multiple parties to produce the same service for a client population, it is effectively creating a networked structure of interdependent contract exchanges pursuing the same public policy goal, even if joint service production does not occur. networks are formally established and goal directed systems of interdependent exchange (provan, fish & sydow, 2007). in service delivery networks, actors (government funders and contractors) are bound to the network by formal contracts; they draw from 53559_13-1.indd 47 18/08/15 5:31 pm 48 j. l. carboni / measuring risks of organizational failure in contract exchange structures the same resource pool and pursue the same public policy goal (e.g. mental health treatment) as the governmental agency. in joint production arrangements, producers collaborate to serve individual clients. in arrangements without joint production, a single program provides the full spectrum of services for a client. in both cases, exchanges are linked by pursuit of a common public policy goal. in service delivery networks without joint production, governmental agencies rely on multiple contractors that work independently to produce services. in these systems, contractors are able to provide a full spectrum of services independently but compete for government clients. competition may be active or passive. when government creates networked structures of exchange with multiple producers, the presence of substitutable alternatives creates competitive markets even if actors do not actively compete. competition reduces the level of dependence on a specific contractor and protects the governmental agency from some risk if a contractor fails because competition creates substitutable alternatives for government. for example, a state government may contract with multiple private prison companies, each providing a full spectrum of services and not interacting with each other. in that case, the failure of a single contractor administering a jail may affect the overall ability of the network to house prisoners. the impact would depend on jail size and availability of alternatives jails to place inmates. the failure of a large jail may leave prisoners without services and prevent the state government from providing legally mandated services especially if other options do not exist. from an exchange theory perspective, as the number of substitutable producers increases, the dependence of a governmental agency on a single exchange partner decreases (emerson, 1972; kelley & thibaut, 1978). for example, if a governmental agency relies on five programs to administer its foster care programs, it creates substitutable alternatives for itself that may buffer the service delivery network from shocks associated with organizational failure. however, we increasingly see organizations producing multiple programs for government. common affiliations (e.g., legally owned by same parent organizations) among programs add complexity to networked structure of exchange. for example, providence service corporation, a national human service provider, contracts with the same government funder to produce multiple programs without joint production in the same geographic area. in maine, two separate providence service corporation subsidiaries receive funding from the same entity, the maine department of health and human services, to provide services for the same pool of at-risk youth. this common affiliation between programs with providence service corporation changes the structure of exchange and the number of substitutable alternatives for the maine department of health and human services. see figure 1 for a visual representation of these relationships. what appear to be two independent service delivery programs are actually two programs administered by the same parent entity, thereby increasing the dependence of maine on providence service corporation even though the state has established two independent programs. should providence service corporation fail, maine would lose both programs even though they are independent of each other. 53559_13-1.indd 48 18/08/15 5:31 pm j. l. carboni / measuring risks of organizational failure in contract exchange structures 49 when service delivery networks are segmented into sub-networks, the complexity of the network is further increased. sub-networks may be differentiated by a specific need or characteristic in the population served by the contract network, thereby decreasing the number of substitutable alternatives in the network. for example, in juvenile justice programs, a state may fund 100 contracted programs to reduce recidivism through juvenile rehabilitation but these programs may be segmented based on juvenile treatment needs with some programs to specifically treat sex offenders or substance abusers. contractors directly compete for resources such as juveniles within sub-networks. contractors do not compete directly for resources among sub-networks and are therefore not substitutable alternatives. for example, specialized programs that are created to treat mentally ill juvenile delinquents would not compete for juveniles with specialized programs created to treat sex offender juvenile delinquents. however, exchanges among sub-networks are indirectly linked by virtue of being attached to the same government funder and pursuit of the same public policy goal to reduce juvenile recidivism. common affiliations with parent organizations may exist across sub-networks, potentially increasing complexity in the dependence of a government agency on an organization. the failure of an organization with programs in multiple specialties may cause more disruption than the failure of an organization with a program in a single specialty. figure 1. underlying common affiliation of service delivery program. 53559_13-1.indd 49 18/08/15 5:31 pm 50 j. l. carboni / measuring risks of organizational failure in contract exchange structures in the next section, i use affiliation networks to simultaneously operationalize and measure multiple interdependencies among service delivery programs. interdependencies include direct links to the government funder through sub-networks and to parent organizations. accounting for both simultaneously allows us to understand the true nature of government dependence on organizations in the service delivery network. i then apply this measure to an existing contract system to demonstrate that some organizational failures would be more disruptive to a network than others would. for the sake of brevity and clarity, i limit the discussions in this study to service delivery networks that are government monopsonies directly administered by a government funder (i.e. a direct relationship between funder and contractor). 3. common affiliations and the structure of exchange in networks affiliation networks are multi-mode networks with two or more distinct types of entities called modes. modes may be actors, events, or distinct sets of organizations. connections among actors are established through common connections to another mode (e.g. – parent organization or government funded specialty) rather than a direct tie to another actor in the same mode (e.g. – program to program) as is typical in most network studies. work on affiliation networks is based on the acknowledgement that actors are members of multiple collectivities, or social circles (simmel, 1955). the way those social circles intersect has implications for actors’ identities and behaviors. in terms of contracts, affiliations may shape dependence of government on a particular organization and vice versa. affiliations may also shape incentives for contractor performance. where government is highly dependent on organizations to provide services, those organizations may have with less incentive to perform. affiliation networks allow researchers to focus on how actors are connected through events, how events are connected through actors or consider multiple modes simultaneously (breiger, 1974). in contrast, most network analyses involve one-mode networks where structural variables (e.g., network position) are measured on a single mode of actors with actor-to-actor ties (wasserman & faust, 1994). it is critical to examine common affiliations of programs in and among subnetworks in order to understanding the risks individual organizational failures pose to the network for example, figure 2 depicts a simple service delivery network where there are multiple programs (square nodes) attached to a government funder (square nodes) but not to each other, suggesting that there is a lack of joint production among producers. the structure in this figure does not capture the indirect connections that occur through programs’ common affiliation with another mode (e.g., with a parent organization). when programs have a type of common affiliation with each other, the structure changes as shown in figure 3. in figure 3, programs (square nodes) are not engaged in joint production but they are indirectly connected to each other through a common affiliation to another mode—a common parent organization (triangle nodes)—indicating that programs 53559_13-1.indd 50 18/08/15 5:31 pm j. l. carboni / measuring risks of organizational failure in contract exchange structures 51 are not be fully independent of each other. this relationship cannot be characterized as a form of exchange, but it affects the structure of exchange relationships in that it reduces substitutable alternatives. for example, if a service delivery network has eight programs administered by four parent organizations, the number of independent exchange partners is reduced and government choice is more constrained than if it contracted for eight programs administered by eight separate parent organizations. furthermore, suppose all eight programs are administered by one parent organization. if one parent organization administers all programs, production monopolies are simply shifted to the private sector and competition is nonexistent. an organizational failure (i.e., the failure of the parent organization) in this situation would be catastrophic for the government agency, as it is dependent on one organization and has no other immediate substitutable alternatives. affiliation networks help us understand these types of interdependencies. once interdependencies are understood, they can be incorporated into indices. indices developed from affiliation networks can give us a better understanding of the true nature of governmental agency dependence on a particular organization and the risk of disruption to a system should an organization fail. in the next section, i discuss the measurement of common affiliations across segmented markets and propose indexes of organizational importance for parent organizations. the affiliation networks considered contain three modesprograms, parent organizations and specialties that segment the market. figure 2. simple service delivery network. 53559_13-1.indd 51 18/08/15 5:31 pm 52 j. l. carboni / measuring risks of organizational failure in contract exchange structures 4. measurement of common affiliations in the previous section, i described properties of affiliations networks. in this section, i discuss measurement of multi-mode networks and develop an index that reflects how dependent government is on specific organizations in a network. measuring affiliation multi-mode affiliation networks is more complex than measuring standard one-mode networks. network measures typically require a square matrix because they are based on the number and/or length of paths between actors. affiliation networks typically contain two modes. two-mode matrices can easily be transformed into a one-mode matrix by multiplying the matrix by its inverse. in a three-mode matrix, transforming the matrix is more complex. simply converting a three-mode matrix to a one-mode matrix through multiplication would require two modes be treated as a single mode (two modes would be represented as columns or rows), obscuring relationships between those modes. for example, in a three-mode matrix with programs, parent organizations and program specialties, programs could represent rows and parent organizations and sub-networks would represent columns. converting the matrix into a one-mode matrix would treat parent organizations and specialties as a single mode rather than as two distinct modes. figure 3. service delivery network with affiliations. 53559_13-1.indd 52 18/08/15 5:31 pm j. l. carboni / measuring risks of organizational failure in contract exchange structures 53 another option to analyze three modes would be to examine two separate two-mode matrices. for example, one matrix would contain programs and parent organizations while the second matrix would contain programs and specialties. this option is problematic because examining two-mode matrices separately does not represent interdependencies among all three modes; separate two-mode matrices consider service delivery program affiliations with parent organizations and specialties and thus does not accurately capture the true nature of dependence in the networked structure of exchange. to overcome problems with measuring affiliation networks with more than two types of modes, fararo and doreian (1984) present a generalized version of duality (breiger, 1974) that includes three distinct modes based on the following axioms. in their approach, there are three types of nodes (called modes) and links only exist between modes. limiting network to ties among modes allows for examination of overlaps and common membership without the “noise” of actor-to-actor relationships within modes and is well suited to service delivery networks without joint production. the tripartite matrix allows the researcher to examine three modes simultaneously, rather than examining separate twomode matrices. see figure 4 for a visual representation of a three-mode network. this network includes three type of modesprograms (circles), parent organizations (squares), and figure 4. visual representation of a three-mode affiliation network. 53559_13-1.indd 53 18/08/15 5:31 pm 54 j. l. carboni / measuring risks of organizational failure in contract exchange structures government funded specialties (triangles). links or ties exist among modes but not within modes. there are program to specialty links and program to parent organization links. measures for three-mode matrices are rare and largely descriptive (see cornwell, curry, & schwirian, 2003, for a descriptive example). to overcome the problems of using more than two modes without losing analytical leverage of considering three modes simultaneously, i developed a weighted proximity estimate based on a valued two-mode matrix that incorporates all three modes of data. in the valued affiliation matrix, two modes are reflected as columns and rows. the corresponding cell values between rows and columns reflect the number of third-mode actors with attachment to both actors in rows and columns. for example, if two programs were connected to parent organization 1 and specialty 1, there would be a two in the cell that corresponds with organization 1 and specialty 1. see figure 5 for a visual network representation of a valued two-mode matrix. in figure 5, parent organizations and program specialties are represented. links between the modes are based on the number of program affiliated with both. as line thickness increases, the number of common programs among organizations and specialties also increases. following the example given figure 5. visual representation of a valued two-mode affiliation network. 53559_13-1.indd 54 18/08/15 5:31 pm j. l. carboni / measuring risks of organizational failure in contract exchange structures 55 above, organization 1 and specialty 1 are linked by 2 common programs. the 2.0 on the line between organization 1 and specialty 1 indicates this affiliation. following the creation of the valued two-mode affiliation matrix, i created a onemode adjacency matrix by multiplying the two-mode matrix by its inverse using the cross products formula in ucinet (borgatti, everett, & freeman 2002). cross products are derived for each pair of actors represented in the columns and rows by multiplying common participation in events between two actors and summing the result. the resulting adjacency matrix represents a single mode. for example, the adjacency matrix developed from a valued two-mode affiliation matrix with parent organizations and specialties would have either parent organizations or specialties as both rows and columns. in the case of parent organizations as the mode represented in the one-mode adjacency matrix, the cell value represents the breadth and depth of specialties parent organizations have in common. see figure 6 for a visual representation of the adjacency matrix. in figure 6, only parent organizations are represented in the figure. where organizations are connected, it indicates the parent organizations exist within and across the same program specialties. as line thickness increases, the number of common programs specialties also increases. as size of node increases, parent organizations are more central to service delivery network functioning, or government is more dependent on them. from this one mode matrix, i created an index that reflects dependent government is on individual organizations for service delivery in the network. the index captures the breadth and depth of the organization’s presence in the network relative to figure 6. visual representation of a one-mode adjacency network. 53559_13-1.indd 55 18/08/15 5:31 pm 56 j. l. carboni / measuring risks of organizational failure in contract exchange structures importance of other organizations. the index contains the row average scores for each organization in the one-mode matrix. this index provides a singular measure that tells us which organizations are most important to the network in terms of their place in the network structure. this measure can be incorporated into multilevel models to assess the effect of government dependence on organizations for some outcome (i.e., performance). in the next section, i apply this index using data from a state funded service delivery network. 5. application to demonstrate the utility of this index i apply it to an empirical case: a state funded and managed network of juvenile justice residential treatment programs. the purpose of this application is to demonstrate how the index works and to demonstrate that governmental agencies depend on organizations for public policy implementation in varying degrees. the data used in application are from a service delivery network of 123 state funded juvenile justice residential services programs for juvenile delinquents in florida in 2008. programs are delivered without joint production and were administered by 30 parent organizations and segmented into eight sub-networks based on program treatment specialty and juvenile gender. the program specialties were for: developmentally disabledmale; general populationmale; mental healthmale; sex offender-male; substance abuser-male; general populationfemale; mental health-female; and substance abuser-female. summary data on the distribution of programs across organizations and specialties is included in table 1. the table includes the number of programs in each specialty by parent organization. a visual representation of this information is given in figure 7. residential services programs provide a continuum of care for juveniles committed to the care and custody of the state. in addition to rehabilitation, programs provide education and vocational services. programs vary in size and scope but they are likely to be small, group home settings rather than large lock down detention programs. the residential services program network has multiple producers that are not engaged in joint service production and the state is the sole buyer of services. though programs receive some federal funds, all funding is routed through the state and directly administered to programs by the state. parent organizations are awarded the right to administer individual programs. the state decides where juveniles are placed. programs are indirectly linked to each other through common affiliation with parent organizations and through common affiliation with sub-network specialties established by the state. each program is affiliated with one specialty and one parent organization but specialties and parent organizations do not have direct affiliations. the number of programs per sub-network varies, as does the number of programs administered by each parent organization as demonstrated in table 1. to recap, in this affiliation network there are three modes: programs, parent organizations, and specialties (sub-network based on program treatment specialty and juvenile gender). 53559_13-1.indd 56 18/08/15 5:31 pm j. l. carboni / measuring risks of organizational failure in contract exchange structures 57 table 1 distribution of programs across parent organizations and specialties. developmentally disabled (male) general (male) mental health (male) sex offender (male) substance abuse (male) general (female) mental health (female) substance abuse (female) total organization 1 0 6 0 0 0 1 0 0 7 organization 2 0 2 0 0 1 0 0 0 3 organization 3 1 2 0 0 0 0 0 1 4 organization 4 0 1 0 0 0 0 0 0 1 organization 5 0 7 0 0 0 1 0 0 8 organization 6 0 1 0 0 0 0 0 0 1 organization 7 0 8 2 2 2 0 0 0 14 organization 8 0 0 0 0 1 0 0 0 1 organization 9 0 0 0 0 0 2 0 1 3 organization 10 0 0 1 0 0 0 0 0 1 organization 11 0 4 0 0 0 1 0 0 5 organization 12 0 0 0 0 0 0 1 0 1 organization 13 0 1 0 0 0 0 0 0 1 organization 14 0 2 0 0 0 1 0 0 3 organization 15 0 2 0 0 0 1 0 0 3 organization 16 0 1 0 0 0 0 0 0 1 organization 17 0 3 0 0 0 0 0 0 3 organization 18 0 2 0 0 1 0 0 0 3 organization 19 0 1 0 0 0 0 0 0 1 organization 20 0 1 0 0 0 0 0 0 1 organization 21 0 0 0 0 0 1 0 0 1 organization 22 0 1 0 0 0 1 0 0 2 organization 23 0 7 3 1 1 1 3 0 16 organization 24 0 2 1 1 1 0 1 0 6 organization 25 0 2 0 0 0 0 0 0 2 organization 26 0 1 0 4 0 0 0 0 5 organization 27 1 7 1 0 1 1 0 0 11 organization 28 0 1 0 1 1 0 2 0 5 organization 29 0 1 0 0 0 2 0 0 3 organization 30 0 6 0 0 0 1 0 0 7 total 2 72 8 9 9 14 7 2 123 see figure 7 for a visual representation of the three-mode network. in figure 7, shapes represent the modes (circle5program; square5parent organization; triangle5specialty). the matrix used to derive figure 7 is included in table 1. using affiliations from this threemode network, a two-mode valued affiliation matrix is derived as described in the previous section. 53559_13-1.indd 57 18/08/15 5:31 pm 58 j. l. carboni / measuring risks of organizational failure in contract exchange structures see figure 8 for a visual representation of the two-mode valued matrix. for the two-mode network map in figure 8, squares represent specialties and are labeled with the specialty1. circles represent parent organizations and are labeled with their names. the thickness of the line represents the number of programs an organization has in a specialty2. the thicker the line is, the more the number of programs a parent organization has in a specialty. using the two-mode valued affiliation matrix, a one-mode adjacency matrix was derived in ucinet (borgatti et al., 2002) using the cross products of the rows. cross products are derived for each pair of actors by multiplying common participation in events between two actors and summing the result. the sum represents the number of specialties and degree of specialties (measured as the number of programs in a specialty) parent organizations have in common. see figure 9 for a visual representation of the one-mode adjacency matrix. for the one-mode network maps, circles represent parent organizations. lines between parent organizations indicate they have programs in the same specialty. node size represents how important organizations are to the network. as node size increases, government is more dependent on an organization. nodes are sized according to the dependence index described in the prior section. once matrices were squared (converted to one-mode), row 1 abbreviationsdd 5 developmentally disabled; gen 5 general; mh 5 mental health; so 5 sex offender; sa 5 substance abuse 2 market share must be controlled for in a statistical model. this measure reflects presence of programs within and across specialties. figure 7. visual representation of the three-mode network. key shape 5 mode (circle 5 program; square 5 parent organization; triangle 5 specialty) 53559_13-1.indd 58 18/08/15 5:31 pm j. l. carboni / measuring risks of organizational failure in contract exchange structures 59 figure 8. visual representation of the two-mode valued affiliation network. key color/ shape 5 mode (circle 5 organization; square 5 specialty) line thickness 5 number of common programs figure 9. visual representation of the one-mode adjacency matrix. key square 5 organization larger nodes compete more intensely within and across sub-networks 53559_13-1.indd 59 18/08/15 5:31 pm 60 j. l. carboni / measuring risks of organizational failure in contract exchange structures averages were computed for each parent organization. the row average is a measure of direct competition with other parent organizations for each parent organization. this index reflects how dependent government is on each organization. the score in the index accounts for the parent organization’s presence within an across sub-networks relative to other parent organizations in the network. as organizations have a greater presence within and across specialties (measured by the number of programs they administer within and across specialties), organizations have a greater score and higher importance to network functioning. the scores for parent organizations are included in table 2. as scores get larger, government is more dependent on an organization. the failure of organizations with high scores would have a larger disruption to the network than failure of organizations with lower scores. for example, the state is much more dependent on organization 7 for service production than on organization 12. organization 7 administers 14 separate programs across six specialties, while organization 12 administers one program in a single specialty. the failure of organization 7 would be much more disruptive to service delivery in the larger network than the failure of organization 12. again, this measure goes beyond the count of programs a parent organization administers or specialties in which it has a presence. organizations become more important to the entire network as they provide a greater number of programs across more specialties and more important to the specialty as they provide more programs in a specialty. for example, the failure of organization 7 would be slightly more disruptive to the entire network than that of organization 23, even though organization 23 administers more total programs. the distribution of organization 7’s programs across the entire network means its failure would have great impact. it has more than one program across four specialties, while organization 23 only has more than one program across three specialties. in other cases, a single organization has such a dominant presence in one specialty that the organization’s removal would become catastrophic for the network. the index simultaneously captures breadth and depth of parent organization presence in the networked structure of exchange. 6. conclusions in many cases, governmental agencies must contract in sub-optimal conditions, which violate assumptions of competitive markets. as government contracting increases, it will become more important to understand how the structure of exchange in networks influences the availability of substitutable alternatives for governmental agencies. the stability of government-created structures is partially dependent on the nature of alternatives for the government and the risk of an organization’s failure. beyond understanding the structure of alternatives, governments should investigate the likelihood that exchange partners might fail, particularly when that exchange partner is critical to service delivery. disruption to service delivery could be catastrophic if central organizations fail. this is 53559_13-1.indd 60 18/08/15 5:31 pm j. l. carboni / measuring risks of organizational failure in contract exchange structures 61 table 2 parent organization index scores (fy08) organization index score organization 1 13.633 organization 2 4.933 organization 3 4.733 organization 4 2.367 organization 5 15.600 organization 6 2.367 organization 7 18.400 organization 8 0.267 organization 9 0.833 organization 10 0.233 organization 11 9.500 organization 12 0.200 organization 13 2.367 organization 14 5.100 organization 15 5.100 organization 16 2.367 organization 17 6.900 organization 18 4.933 organization 19 2.367 organization 20 2.367 organization 21 0.433 organization 22 2.800 organization 23 17.033 organization 24 5.633 organization 25 4.667 organization 26 3.033 organization 27 16.133 organization 28 3.233 organization 29 3.167 organization 30 13.633 particularly true where existing programs do not have the capacity to absorb clients immediately. in juvenile justice or prison settings, governments cannot wait for new programs to get up and running due to legal obligations to maintain incarceration. even in less restrictive situations, the failure of an important organization has the potential to disrupt service production for the entire network. 53559_13-1.indd 61 18/08/15 5:31 pm 62 j. l. carboni / measuring risks of organizational failure in contract exchange structures in addition to having implications for system stability, the structure of exchange and the embeddedness of actors within service delivery networks may influence network performance. by choosing to exchange with multiple actors, government funders may be able to leverage their dependence on exchange partners and decrease the risks associated organizational failures. incentives to perform are likely to be influenced by how actors are embedded in a particular exchange structure. competition among providers, rather than privatization, produces benefits for government (domberger, 1994; donahue, 1989). as government increases the number of substitutable alternatives, individual contractors will have more incentive to perform well to maintain the flow of resources to their program. it is likely that as government increases dependence on a single organization, that organization will have less incentive to perform beyond contractually mandated goals. according to exchange theory, when contractors are replaceable, they may be more likely to perform according to the terms of their exchange partner, the government funder, for fear of not being selected or of being replaced (emerson, 1972). measuring dependence and performance over time would shed light on this issue. in this paper, i emphasized the organization of governance arrangements and the potential effect of organization on governance outcomes, a topic not widely considered. at a broad level, this study follows the recent tradition of examining the changing role of government and its increasing use of private agents to produce goods and services traditionally produced by government with two contributions. first, i developed the idea that contracts are embedded in larger networked system of exchange even if joint production does not occur. understanding the structures of these networks has the potential to influence contract performance incentives that are independent of contract management practices. second, i offered a way to measure the complex networks of exchange with direct links to a government funder and common affiliations among contractors. i demonstrated the networked structure of providers is more complex than a simple count of programs or parent organizations, particularly where networks are segmented by client types. the index developed here may be used to understand how important individual actors are for network functioning. it can also be incorporated into statistical models on contract performance. incorporating structural measures into statistical models will improve ability to predict individual contract performance and understand risks associated with organizational failure in service delivery networks. finally, while i focused on networks without joint production, the index i propose may be extended to consider government contracting across multiple networks with different resource bases and networks with joint production. incorporating structural variables into contract performance models may shed light on the interplay or complementarity between structural predictors of performance and management predictors of performance. previous studies suggest that structure and management are important governance tools, but using both may be redundant (rowley, behrens, & krackhardt, 2000). understanding the conditions under which each is likely to work will provide managers with additional tools in deciding where to employ additional time and resource intensive relational strategies and where to rely more heavily on structure to induce performance. for example, when 53559_13-1.indd 62 18/08/15 5:31 pm j. l. carboni / measuring risks of organizational failure in contract exchange structures 63 contractors are likely to perform because of their structural position, it may be unnecessary to spend additional resources developing relationships. where contractors are more likely to underperform because of their position (e.g. governmental agency is dependent on them), developing more resource intensive relationships may promote desired performance. references agranoff, r. (2008). enhancing performance through public sector networks: mobilizing human capital in communities of practice. public performance & management review, 31(3), 320–347. agranoff, r., & mcguire, m. (2003). collaborative public management: new strategies for local government. washington, dc: georgetown university press. ansell, c., & gash, a. (2008). collaborative governance in theory and practice. journal of public administration research and theory, 18(4), 43–571. breiger, r. l. (1974). the duality of persons and groups. social forces, 53, 181–190. borgatti, s. p., everett, m. g., & freeman, l. c. (2002). ucinet for windows: software for social network analysis. harvard, ma: analytic technologies. brown, t., & potoski, m. (2003). managing contract performance: a transaction costs approach. journal of policy analysis and management, 22(2), 275–297. brown, t., & potoski, m., & van slyke, d. (2007). trust and contract completeness in the public sector. local government studies, 33(4), 607–623. brown, t., & potoski, m., & van slyke, d. (2010). contracting for complex products. journal of public administration research and theory, 20(suppl. 1), i41–i58. brudney, j., fernandez, s., ryu, j., & wright, d. (2005). exploring and explaining contracting out: patterns among the states. journal of public administration research and theory, 15(3), 393–419. bryson, j. m., crosby, b. c., & stone, m. m. (2006). the design and implementation of cross-sector collaborations: propositions from the literature. public administration review, 66 (suppl. 1), 44–55. carboni, j. l., & milward, h. b. (2012). governance, privatization and systemic risk in the disarticulated state. public administration review, 72(suppl. 1), 34–44. cornwell, b., curry, t. j., & schwirian, k. p. (2003). revisiting norton long’s ecology of games: a network approach. city and community, 2(2), 121–142. domberger, s. (1994). public sector contracting: does it work? the australian economic review, 27(3), 91–96. donahue, j. (1989). the privatization decision: public ends, private means. new york, ny: basic books. emerson, r. m. (1972). exchange theory. in j. berger, m. zelditch, & b. anderson (ed.), sociological theories in progress (vol. 2, pp. 38–87). boston, ma: houghton-mifflin. fararo, t., & doreian, p. (1984). tripartite structural analysis: generalizing the breiger-wilson formalism. social networks, 6(2), 141–175. fernandez, s. (2007). what works best when contracting for services? an analysis of contracting performance at the local level. public administration, 85(4), 1119–1140. ferris, j., & graddy, e. (1991). production costs, transaction costs and local government contractor choice. economic inquiry, 29(3), 541–554. hefetz, a., & warner, m. (2004). privatization and its reverse: explaining the dynamics of the government contracting process. journal of public administration research and theory, 14(2), 171–190. hill, c. j., & lynn, l. (2004). governance and public management: an introduction. journal of policy analysis and management, 23(1), 3–11. hodge, g. (2000). privatization: an international review of performance. boulder, co: westview press. huang, k., & provan, k. g. (2007). structural embeddedness and organizational social outcomes in a centrally governed mental health services network. public management review, 9(2), 169–189. isett, k. r., mergel, i. a., leroux, k., mischen, p. a., & rethemeyer. r. k. (2011). networks in public administration research: understanding where we are and where we need to go. journal of public administration research and theory, 21(suppl. 1), 157–173. 53559_13-1.indd 63 18/08/15 5:31 pm 64 j. l. carboni / measuring risks of organizational failure in contract exchange structures isett, k. r., & provan, k. g. (2005). the evolution of dyadic interorganizational relationships in a network of publicly funded nonprofit agencies. journal of public administration research and theory, 15(1), 149–165. kelley, h. h., & thibaut, j. w. (1978). interpersonal relations: a theory of interdependence. new york, ny: wiley and sons. lynn, l., heinrich, c., & hill, c. (2000). studying governance and public management: challenges and prospects. journal of public administration research and theory, 10(2), 233–261. lynn, l., heinrich, c., & hill, c. (2001). improving governance: a new logic for empirical research. washington, dc: georgetown university press. malatesta, d. m., & smith, c. r. (2014). designing contracts for complex services. public administration, 92(3), 531–548. milward, h. b., & provan, k. g. (2000). governing the hollow state. journal of public administration research and theory, 10(2), 359–380. milward, h. b., provan, k. g., fish, a., isett, k. r., & huang, k. (2009). governance and collaboration: an evolutionary study of two mental health networks. journal of public administration research and theory, 20(suppl. 1), i125–i141. o’leary, r., gerard, c., & bingham, l. (2006). introduction to the symposium on collaborative public management. public administration review, 66(suppl. 1), 6–9. popp, j., milward, h. b., mackean, g., casebeer, a., & lindstrom, r. (2014). inter-organizational networks: a review of the literature to inform practice. ibm center for the business of government. retrieved from http:// www.businessofgovernment.org/report/inter-organizational-networks-review-literature-inform-practice provan, k. g., fish, a., & sydow, j. (2007). interorganizational networks at the network level: a review of the empirical literature on whole networks. journal of management, 33(3), 479–516. provan, k. g., huang, k., & milward, h. b. (2009). the evolution of structural embeddedness and organizational social outcomes in a centrally governed health and human services network. journal of public administration research and theory, 19(4), 873–893. provan, k. g., & milward, h. b. (1995). a preliminary theory of network effectiveness: a comparative study of four mental health systems. administrative science quarterly, 40(1), 1–33. romzek, b., & johnson, j. (2002). effective contract implementation and management: a preliminary model. journal of public administration research and theory, 12(3), 423–453. romzek, b., & johnson, j. (2005). state social services contracting: exploring the determinants of effective contract accountability. public administration review, 65(4), 436–449. rowley, t., behrens, d., & krackhardt, d. (2000). redundant governance structures: an analysis of structural and relational embeddedness in the steel and semiconductor industries. strategic management journal, 21(3), 369–386. simmel, g. (1955). conflict in the web of group affiliations. new york, ny: free press. smith, s. r. (1996). transforming public services: contracting for social and health services in the united states. public administration, 74(1), 113–127. smith, s. r., & smyth, j. (1996). contracting for services in a decentralized system. journal of public administration research and theory, 6(2), 277–296. van slyke, d. (2007). agents or stewards: using theory to understand the government-nonprofit social service contracting relationship. journal of public administration research and theory, 17(2), 157–187. wasserman, s., & faust, k. (1994). social network analysis: methods and applications. oxford, england: cambridge university press. 53559_13-1.indd 64 18/08/15 5:31 pm the shape of watershed governance: locating the boundaries of multiplex networks complexity, governance & networks-vol 02, issue 01 (2015) 65–82 65 doi: 10.7564/15-cgn25 the shape of watershed governance: locating the boundaries of multiplex networks steve scheinert*, christopher koliba, stephanie hurley, sarah coleman, and asim zia university of vermont e-mail: steve.scheinert@uvm.edu; e-mail: ckoliba@uvm.edu; e-mail: shurley1@uvm.edu; e-mail: scoleman@uvm.edu; e-mail: azia@uvm.edu governance networks are both nested and interconnected systems. identifying internal boundaries within governance networks, such as those governance structures that influence and are influenced by large and diverse watersheds such as the lake champlain basin, is necessary for differentiating between multiple functional subnetworks. internal network boundaries exist between functional subnetworks when the networks have divergent structures (weible & sabatier, 2005). a qualitative case study of lake champlain basin watershed governance networks identified several key overlapping subnetworks in which organizations interact in a variety of ways (koliba, reynolds, zia, & scheinert, 2015). an online survey of institutional actors was used to identify which actors were connected in five different functional subnetworks. structural comparisons are made by analyzing the correlation between the subnetworks based on the quadratic assignment procedure (qap) and network macrostructure. results show that the information sharing, technical assistance, and project collaboration subnetworks formed one grouping, while the reporting and financial resource sharing subnetworks formed another grouping. the results demonstrated that this triangulated comparison was necessary to reach valid conclusions on the structural variation between the subnetworks on a multiplex network when subnetworks were structurally similar. keywords: multiplex networks, network boundaries, governance networks boundaries define the space of operations for actors in a complex system and influence the shifting coalitions that can exist within the system. koliba, meek, and zia (2010) explain the importance of boundaries within governance networks, finding that, “internal boundaries will likely be influenced by the nature of the multiplex ties formed between actors in the network” (p. 169). multiplex ties are network ties that encode multiple types of interaction within a single network tie. coding multiple meanings in a single multiplex tie renders reliable interpretation of analysis based on multiplex ties problematic (butts, 2008, 2009). disaggregating multiplex ties into their separate meanings generates * corresponding author. 62194-15-25.indd 65 18/08/15 10:30 am 66 s. scheinert et al. / the shape of watershed governance functional subnetworks, where all of the ties in each subnetwork encode only one type of interaction. internal boundaries may exist between functional subnetworks when those subnetworks have different structures (weible & sabatier, 2005). the location of internal boundaries determines which functional subnetworks should be analyzed as a single network or analyzed individually. the presence and location of boundaries also influences how managers can ensure that services are delivered and policies are implemented (provan & lemaire, 2012). structural analysis of the functional subnetworks of a governance network will reveal where boundaries exist between those functional subnetworks. identifying a network’s internal boundaries is not the same as identifying its external boundaries. approaches for defining external network boundaries can identify the outer boundary, the “network boundary,” as seen in figure 1 (butts, 2008, 2009; lauman, marsden, & prensky, 1989). the external boundaries of the multiplex network also define the external boundaries of the subnetworks; every node that is included in any subnetwork is included in the multiplex network and all of the other subnetworks. since external bounding approaches can only set this external boundary, a different approach is needed to identify internal boundaries. weible and sabatier (2005) offer a method for documenting the existence of boundaries within specific systems. they use a method of structural comparison, based in the argument that the subnetworks with the most similar structures are the most closely related ones. figure 1 depicts three structurally different subnetworks, and shows where boundaries exist between them. any structurally similar networks depicted would lack a boundary between them. figure 1. network concept definitions in context 62194-15-25.indd 66 18/08/15 10:30 am s. scheinert et al. / the shape of watershed governance 67 1. boundaries within networked systems a thorough structural analysis of a multiplex governance network can reveal the internal boundaries between the functional subnetworks of a governance network. however, research findings indicate that, in a multiplex network, the existence of a tie between two nodes in one subnetwork increases the likelihood that a tie will exist between those same two nodes in the other subnetworks (provan, fish, & sydow, 2007; weible & sabatier, 2005), indicating that the subnetworks’ dyadic structures will be highly correlated. for the purpose of placing internal boundaries, this expected correlation prevents from even quadratic assignment procedure (qap), which weible and sabatier (2005) apply as their sole analysis for structural comparison, from demonstrating sufficiently strong conclusions about structural variation. instead, for identifying internal boundaries, a more extensive structural analysis is needed to identify smaller, consistent structural variations. we examine this hypothesis through a case study of the empirical watershed governance networks that are present in portions of the lake champlain basin. 1.1. networks in watershed governance and climate change policy a major theme in the studies on governance networks is about how these networks can and should be managed (klijn, steijn, & edelenbos, 2010; mcguire, 2002). these studies often focus on two issues: the role of trust between the actors in a network and how that trust can be built (hajer & versteeg, 2005; klijn, edelenbos, & steijn, 2010; klijn & skelcher, 2007; sorenson & torfing, 2005, 2009) and the impact of network structure and design on network outputs (agranoff, 2004; provan & kenis, 2008; provan & lemaire, 2012; provan & milward, 2001). in this literature, the governance network is treated largely as a proverbial “black box.” instead of linking differing structures to differing management strategies, universal approaches are promoted. managers are instructed to use certain behaviors (milward & provan, 2006) or to build the networks with a certain network structure or pattern of relationships (provan & milward, 2001), to seek a minimum network density (hirschi, 2010), or to manage through a certain type of organization, such as a governmental program or non-governmental organization (ngo) (provan & kenis, 2008). further studies are needed regarding the structures that existing governance networks and their subnetworks take, and what implications those structures have for policy making and service delivery. where the general governance network literature tends to examine conceptual or theoretical networks, the literature on watershed governance networks is able to examine specific governance networks (imperial, 2005; schneider, scholz, lubell, mindruta, & edwardsen, 2003; weible & sabatier, 2005). the research on watershed networks includes a wide body of literature on the benefits and requirements of cooperation and coordination in networks (hirschi, 2010; imperial, 2005; jost & jacob, 2004; lubell & fulton, 2008; schneider et al., 2003; scholz, berardo, & kile, 2008; weible & sabatier, 2005), planning 62194-15-25.indd 67 18/08/15 10:30 am 68 s. scheinert et al. / the shape of watershed governance (dutcher & blythe, 2012; koontz & johnson, 2004; lienert, schnetzer, & ingold, 2013), knowledge diffusion and learning (betsill & bulkeley, 2004; cash et al., 2003; newig, guenther, & pahl-wostl, 2010; vignola, mcdaniels, & scholz, 2013), and system scaling (cohen & davidson, 2011; norman & bakker, 2009; vignola et al., 2013), all of which are harnessed to build theories that apply to all governance networks. what the watershed governance literature continues to miss is a focus on implementation networks and the network reactions to policy decisions (rykkja, neby, & hope, 2014). 1.2. the role of functional ties in forming internal boundaries butts (2009) determines that rigorous network analysis can only rest on networks that use a clear and consistent definition of what the included ties represent. these tie types each define a separate functional subnetwork within the multiplex network. in a rigorous analysis, separate subnetworks must be analyzed independently. the subnetworks in figure 1, for example, arise as separate networks for analysis, each with its own unique type of interaction and pattern of network ties. examples of different types of ties in an institutional network include sharing information, providing reports, and collaborating on projects. the effectiveness of any network relies on its actors properly matching their strategies and structures to the network’s environment, and dyadic and governance arrangements (baltazar & brooks, 2007; koliba et al., 2010; ostrom, 2012; provan & lemaire, 2012; provan & milward, 2001) and of forming an accurate situational awareness through coordination and collaboration (hutchins, 1995; luokkala & virranttaus, 2014; van de walle & turoff, 2008). each instance of a network environment, structure, and governance arrangement generates a new and different operational environment. baltazar and brooks (2007) interpret this variation across operational environments as the formation of boundaries between systems, with organizations requiring different strategies for different environments. since each subnetwork with a different structure is potentially a different operational environment, a method is needed to determine which subnetworks do represent different environments. 1.3. methods for defining external network boundaries finding the proper approach to define the external boundaries of a network is one of the most persistent challenges to any empirical network study (butts, 2008, 2009). a wealth of literature examines the procedures researchers can apply to pre-determine external network boundaries. two methods dominate this literature. the first is a “deductive approach,” where the researcher defines a class of nodes that will be included in the study’s network and then identifies the actors that fit this definition. examples include the individuals of a social network (lienert et al., 2013), the words of a semantic network (diesner & carley, 2011), or the organizations of an institutional network (comfort, oh, 62194-15-25.indd 68 18/08/15 10:30 am s. scheinert et al. / the shape of watershed governance 69 ertan, & scheinert, 2010; koliba et al., 2010). while there are many ways to define a target group of nodes for inclusion, the key aspect of the deductive approach is that the researcher determines the complete list of included nodes in the network prior to collecting data on network ties (betsill & bulkeley, 2004; jost & jacob, 2004; lubell & fulton, 2008; vignola et al., 2013). the second approach is “inductive,” where a set of documents or a small group of informants is selected and the actors mentioned in the documents or by the informants are included in the network (comfort et al., 2010; johnson, 1990; pustejovsky & spillane, 2009; scheinert & comfort, 2014). the deductive method is generally used when boundaries are easily defined and membership in either the node class or network is easily determined. the inductive method is used when boundaries cannot be defined or are unknown in advance. 1.4. structural analysis for locating internal network boundaries the most relevant literature on the definition and role of internal network boundaries focuses on the identification of communities in networks (duch & arenas, 2005; fortunato, 2010). subgroup detection, including clique analysis (burt, 1978), newman’s grouping algorithms (newman, 2006), structural equivalence (burt, 1987; sailer, 1978), and fuzzy overlapping groups (davis & carley, 2008; gregory, 2011), among many other methods, identifies nodes that fit in certain groups. the difference between these methods is that subgroup detection uses a network’s structure to carve that network into subgroups while the external boundaries approaches define the size of the network. weible and sabatier (2005) offer the strongest example of how to build a structural comparison of institutional networks. they compare the structures of groupings of organizations within the subnetworks of a multiplex network for setting policy regarding marine protected areas. they define network links as allies, coordination, advice/ information, and shared beliefs regarding a specific policy, the california marine life protection act, networks. they compare the position of existent dyads with a correlational analysis using quadratic assignment procedure (qap). qap correlation analysis provides a general indicator of when a tie between two given nodes in one network is likely to indicate a tie between those same two nodes in another network. this is a powerful process for comparing subnetworks, but it is not the only one that can and should be considered. subtle differences in tie locations can lead to meaningful differences in network structure; changing just a few ties can be the difference between a lattice network and a small world network (telesford, joyce, hayasaka, burdette, & laurienti, 2011; watts & strogatz, 1998), which is a small enough difference that it might not reflect in qap correlation results. additional, supporting results are needed to augment the qap correlation results. the network analyst must seek results through the triangulation of descriptive analytic techniques whose combined results yield an overall conclusion. to address this need for consistency across a range of results, in this study we applied several analyses 62194-15-25.indd 69 18/08/15 10:30 am 70 s. scheinert et al. / the shape of watershed governance of the subnetworks’ structures to identify consistent and inconsistent structure: subnetwork dyadic correlations and subnetwork macrostructures. internal boundaries will exist between subnetworks with different structures and not exist between subnetworks with similar structures. since networks may take many different forms, it is more important to identify consistencies across a set of subnetworks. the results will support the placement of internal boundaries when measurements are differ across subnetworks. subnetwork dyadic correlation. in this study we applied qap correlation analysis, as used by weible and sabatier (2005). when a high correlation exists between two networks, coded as matrices, there is a greater likelihood that the presence of a tie in one network accurately predicts the presence or absence of a tie between the same two nodes in the other network. high positive correlations indicate similar patterns of observed and unobserved ties while negative correlations indicate an opposite pattern of structural holes and ties. similar networks will share high, positive correlation coefficients. subnetwork macrostructures network macrostructural analyses include small world networks (watts & strogatz, 1998), scale free networks (barabasi & albert, 1999), and core-periphery networks (wasserman & faust, 1994). the public administration literature does not yet regularly apply network macrostructural concepts for describing and analyzing governance networks. each macrostructural analysis defines a theoretical structure that networks can take, including identifying a test for determining if a given network fits that definition. this study identifies small world networks using the coefficient proposed by telesford et al. (2011): w = lrand l − c clatt (1) where l is the observed average path length, c is the observed clustering coefficient, lrand is the average path length of a random network, and clatt is the clustering coefficient of a lattice network. small world networks produce a coefficient at or near zero. in a scale free network, the distribution of degree centrality scores, which count the number of ties that each nodes has (wasserman & faust, 1994), takes the form of a power-law distribution when plotted in a histogram. core-periphery networks can be determined in ucinet (borgatti, everett, & freeman, 2002), which provides an algorithm that identifies core membership and indicates how closely the network matches a core-periphery structure (borgatti, everett, & johnson, 2013). ucinet’s categorical core-periphery algorithm operates similarly to block modeling approaches, identifying and sorting nodes into two blocks (borgatti, everett, & freeman, 2002). the variation patterns found in macrostructural analyses augment dyadic correlation results. similar small world coefficient values, histogram shapes, and core-periphery correlations and core membership are used to determine the location of internal boundaries. if these results can corroborate the results of the qap analysis, then both sets of results will indicate the location of an internal boundary. 62194-15-25.indd 70 18/08/15 10:30 am s. scheinert et al. / the shape of watershed governance 71 2. method 2.1. case study focus the lake champlain basin (lcb) was chosen as a case study for its large and established watershed governance network (osherenko, 2014). lake champlain is a large, freshwater lake that sits astride the borders between vermont and new york in the united states and quebec, canada. the basin is the site of farming communities, particularly in vermont and quebec, and has struggled to meet its water quality requirements under the clean water act. the arrangements made to govern the lcb and its sub-basins (dutcher & blythe, 2012; osherenko, 2014) require interaction and collaboration across state and national borders in order to expand governance structures throughout the watershed. vermont, with limited state budgets and producing the largest share of pollution, particularly from non-point sources (lake champlain basin program [lcbp], 2012; osherenko, 2014), has relied on public-private partnerships for pursuing its water quality goals. this reliance has generated a large and engaged network of stakeholders. the extensive history of multi-stakeholder interactions related to water quality in the vermont section of the lcb makes it an effective case study for examining the structures of a mature governance network and for demonstrating a method of identifying the boundaries between the functional subnetworks of a multiplex governance network. 2.2. data our research used network surveys to acquire data regarding watershed-scale governance in the lake champlain basin. as part of their work, koliba, reynolds, et al. (2015) used an inductive approach to define network boundaries (comfort et al., 2010; scheinert & comfort, 2014), where water resource management documents are used to identify a wide swath of organizations that play a role pertaining to water quality outcomes in the lake champlain basin. the current study approaches a targeted set of representatives from these organizations with an online survey that recorded interactions between organizations with a focus on two lcb sub-basins: the winooski river watershed and the missisquoi river watershed. these watersheds represent vermont’s primary region of development and one of vermont’s two primary agricultural environments, respectively. land uses in these watersheds contribute to the two leading sources of non-point source pollution in vermont into the lcb, urban stormwater runoff and agricultural runoff (lcbp, 2012). we designed this study’s survey as a whole population survey, rather than a samplebased survey. we established an initial alter roster, based on the organizational list that koliba, reynolds, et al. (2015) found. we augmented this list through a range of stakeholder interactions between july, 2012 and may, 2014, including eighteen stakeholder interviews, four focus groups, and two large-scale mediated modeling workshops. the focus groups included approximately two dozen different individual stakeholders representing nearly the same number of organizations while more than 100 stakeholders attended each mediated modeling workshop. this process produced a list of 187 organizations, including ngos, governmental programs at both the state and federal level, regional and municipal 62194-15-25.indd 71 18/08/15 10:30 am 72 s. scheinert et al. / the shape of watershed governance governing entities, and private businesses, along with five collective groups which contain organizations that could not be reached for response.1 we then reviewed this list in consultation with stakeholders to ensure consistency, accuracy, and completeness. during stakeholder consultations, we identified five key types of interaction between organizations: information sharing, technical assistance provision, reporting,2 financial resource sharing, and project coordination and collaboration. researchers identified and approached specified representatives from each organization through email, phone, and in-person contact to recruit participants. each survey participant was presented with a list of all the other organizations in the network and asked to indicate whether they interact with that organization “frequently,” “infrequently,” or not at all for all five different types of interactions. to account for the limits in identifying the initial list of organizations in a network survey, which marsden (2005) discussed, we asked respondents to identify organizations that were not included in the original sample. requests for additional organizations increased the total number to 198 organizations that could be contacted and to a total set of 204 potential alter organizations, including those groups of organizations that could not be reached for a response. as respondents suggested additions, these new additions were added to the survey’s alter rosters and contacted to participate. this method combines the benefits of a pre-determined alter list with the benefits of snowball sampling for establishing an externally bounded governance network (marsden, 2005; scott, 1991). to aid respondents in addressing this extensive list of potential alters the list was divided into four groupings based on organizational type and geographic field of operation. these groupings were governmental programs, regional actors and ngos, organizations with interests in the winooski watershed, and organizations with interests in the missisquoi watershed. we initially asked respondents if their organization interacted with any members of a grouping. if a respondent indicated that their organization did not, then the respondent was not asked about interactions with the members of that group. table 1 contains the survey response rates by grouping. responses provided the data to generate five separate subnetworks, each representing one type of interaction and a sixth network that is the mathematical union of the five subnetworks and representing the full multiplex watershed governance network in both the winooski and missisquoi watersheds of the lake champlain basin. network data are sensitive to missing data (wasserman & faust, 1994). to minimize the error introduced by these missing data, we binarized and symmetrized the network matrices.3 once binarized and symmetrized, the representative of only one organization in the dyad must indicate that the link exists for the link to be included in the network, providing two possible observations of each link. following binarization and symmetrization, the only part of the network that remains unobserved are the dyads that include any two 1 collective groups were used to record interactions with large, medium, and small farms in each the survey’s targeted watersheds and a catch-all organization for interactions with vermont’s agency of agricultural, farms, and markets (vtaafm) that were not covered by the named vtaafm programs included in the survey. 2 for determining reporting structure, the network was designated as the respondent’s “report to” network and framed as a respondent’s “direct reports.” 3 binarization and symmetrization were performed such such that: aij = aji = max(aij, aji), where a is the adjacency matrix and aij and aji are the related cells of the adjacency matrix. aij = aji = max(aij, aji) 62194-15-25.indd 72 18/08/15 10:30 am s. scheinert et al. / the shape of watershed governance 73 nodes both of which did not respond. the observed percentage of the network is reported in table 1’s observation rate column.4 low observation rates for certain groups would undermine confidence if the analysis only used data from the members of those groups. since the analysis always uses the full organization list the high observation rate for the complete network gives confidence in accuracy of the analytic results. 3. network analysis results5 the following results indicate where boundaries fall between the functional subnetworks within the multiplex governance network for the lake champlain basin. the results show two sets of substantially similar subnetworks. the variation in results is subtle, but present. these results suggest that internal boundaries separate two groups of subnetworks. one group contains the reporting and financial resource sharing subnetworks while the other group contains the information sharing, technical assistance provision, and project coordination and collaboration subnetworks. 3.1. subnetwork dyadic correlation the results from the qap correlation analysis (table 2) confirm the necessity of running additional analyses. the subnetworks are all highly correlated, preventing firm conclusions about structural differences. nevertheless, there is a pattern in the results that 4 observation rate is obtained using the number of organizations that completed responses in each organizational group in the network such that: density = 1 − # unobserved dyads total dyads (2) density = 1 − (nrn)(nrn−1) (n)(n−1) (3) where nrn stands for the number of non-responding nodes and n stands for the total number of nodes. in the second, more specific version of this formula, both the numerator and denominator should be divided by 2, to account for the non-directionality introduced by symmetrization. since both terms are divided by 2, this step becomes a multiplicative identity and so is omitted. 5 this study uses two different network analysis programs: *ora (carley, 2001–2011) and ucinet (borgatti, everett, and freeman, 2002). a note with each result will indicate which program was used to obtain that result. table 1 network survey response rates by organizational grouping organizational group number of contacts completed responses response rate (%) observation rate (%) governmental programs 56 26 46.4 71.75 regional actors and ngos 50 26 52.0 73.47 winooski watershed 52 11 21.2 38.16 missisquoi watershed 40 12 30.0 51.54 total 198 75 37.9 60.26 62194-15-25.indd 73 18/08/15 10:30 am 74 s. scheinert et al. / the shape of watershed governance table 2 qap correlation matrix reporting project coordination financial resources information sharing technical assistance multiplex network reporting 1 sig. – project coordination 0.618 1 sig. (0.000) – financial resources 0.700 0.719 1 sig. (0.000) (0.000) – information sharing 0.542 0.735 0.595 1 sig. (0.000) (0.000) (0.000) – technical assistance 0.602 0.788 0.670 0.765 1 sig. (0.000) (0.000) (0.000) (0.000) – union 0.526 0.760 0.586 0.940 0.809 1 sig. (0.000) (0.000) (0.000) (0.000) (0.000) – provides some guidance on where internal boundaries may lie. the weakest correlation coefficients are in the relationships where either the reporting or financial resource sharing subnetworks are compared with either the information sharing, technical assistance, or coordination subnetworks. the reporting and financial resource sharing subnetworks are relatively highly correlated, as are any pairings of the information sharing, project coordination, and technical assistance subnetworks. this pattern suggests that this system has two separate sets of related functional subnetworks. one set includes the information sharing, coordination and collaboration, and technical assistance subnetworks. the second set includes the reporting and financial resource sharing subnetworks. the correlation coefficients are all high, and high correlations prevents firm conclusions about the placement of boundaries and supports the conclusion that observing a tie in one network increases the likelihood of observing that same tie in the other subnetworks (provan fish, & sydow, 2007; weible & sabatier, 2005). 3.2. subnetwork macrostructures figure 2 contains the histograms that can be used to determine if each functional subnetwork is a scale free network. all five subnetworks and the union network fit the definition of a scale free network, displaying a power law distribution for their centrality scores. the reporting subnetwork fits a consistent and unambiguous power law; each successive step in the graph is lower than the preceding step until the graph tails off with single nodes marking the highest levels. the graphs for information sharing, coordination, technical assistance, and for the union network contain a consistent complicating characteristic. in each, a secondary hump emerges at around 40% of the total possible ties. 62194-15-25.indd 74 18/08/15 10:30 am s. scheinert et al. / the shape of watershed governance 75 figure 2. scale free network analysis: histograms of normalized degree centrality in all subnetworks 62194-15-25.indd 75 18/08/15 10:30 am 76 s. scheinert et al. / the shape of watershed governance this pattern is clearest in the information sharing network and least clear in the coordination network but remains present. the financial resource sharing subnetwork has a similar small hump, though it is at a lower centrality score, about 20%. this pattern of differences supports linking the information sharing, technical assistance, and coordination subnetworks, while not supporting the links between the reporting and financial resource sharing subnetworks that emerged in the qap analysis. the results for the small world network analysis ( table 3) show the clearest divides of any analysis yet, suggesting three groupings. in this case, the information sharing and union networks stand out from both the grouping of the technical assistance and coordination subnetworks and the group of the reporting and financial resource sharing subnetworks. all subnetworks can likely be considered as small world networks. while telesford et al. (2011) suggest a range of -0.5 , ω , 0.5 on a scale between 21 and 1, while allowing for variation in the range in different network contexts. this suggested range would likely make a value of 0.3, as observed for the reporting and financial resource sharing subnetworks, a borderline result. with positive coefficients of less than 0.2, the information sharing, technical assistance, and coordination subnetworks are small world networks. the results of the core-periphery analysis (table 4) suggest similar internal boundary locations and begin to offer some basis for why the reporting and financial resource sharing subnetworks differ from the other subnetworks. though all of the subnetworks are core-periphery networks, the reporting and financial resource sharing subnetworks do not correlate as highly with a theoretical core-periphery network as do the other three table 3 small world coefficient calculation (ucinet) reporting project coordination information sharing financial resource sharing technical assistance multiplex network node count 204 204 204 204 204 204 density 0.041 0.080 0.117 0.050 0.090 0.130 average degree 8.10 16.19 23.74 9.94 18.19 26.47 clustering coefficient6 0.606 0.581 0.611 0.603 0.566 0.614 average distance 2.196 2.112 1.985 2.133 2.068 1.954 clustering coefficient (lattice) 0.643 0.700 0.717 0.667 0.706 0.720 ave. distance (erdos-renyi) 2.738 2.171 1.936 2.538 2.088 1.899 clustering ratio (lattice) 0.94 0.83 0.85 0.90 0.80 0.85 distance ratio 1.25 1.03 0.98 1.19 1.01 0.97 telesford small world coefficient 0.30 0.20 0.12 0.29 0.21 0.12 6 though the definition of the clustering coefficient is consistent across ucinet and *ora, the two programs consistently produce different values for this measure. all comparisons of the value are made using a consistent program to avoid potentially erroneous conclusions being made as a result of comparing differently scaled results. small world coefficients were calculated using clustering coefficients calculated in ucinet. 62194-15-25.indd 76 18/08/15 10:30 am s. scheinert et al. / the shape of watershed governance 77 subnetworks. the reporting and financial resource sharing subnetworks also have the largest number of government agencies in their cores, calculated as a percentage of the nodes in their core, showing that the cores for the reporting and financial resource sharing subnetworks differ, in a similar, substantive, way, from the cores of the other three subnetworks. 4. empirically-identified network boundaries as was expected due to the influence that presence of ties in one subnetwork has across other subnetworks (provan, fish, & sydow, 2007; weible & sabatier, 2005), these subnetworks are substantially similar, showing high correlation coefficients between all five subnetworks. we can confirm that ties of different types are highly correlated. we can also conclude that our additional structural analyses are necessary to find internal boundaries within governance networks. the results of this thorough structural analysis indicate that internal boundaries do exist between the functional subnetworks of the water quality governance network in the lake champlain basin. similar to how a factor analysis in statistics allows for viewing different variables as contributing to the same underlying conceptual factors, our analysis allows for viewing certain functional network ties as part of the same underlying subnetworks. in this governance network, the reporting and financial resource sharing subnetworks show quantitative differences from the information sharing, technical assistance, and project coordination and collaboration subnetworks. the qap correlation coefficients between the members of these two groupings of subnetworks are higher than those between subnetworks in different groupings. the subnetworks of one group are a better fit to the scale free network model and poorer fits to the small world and core-periphery network models. the cores of these same subnetworks also contain a higher percentage of governmental programs than the information sharing, technical assistance, and coordination and collaboration subnetworks. the same sets of subnetworks show similar patterns for the degree to which they fit macrostructural concepts and show similar deviations from the definitions when the subnetworks do not fit the definitions exactly, such as in the scale free analysis. each of these results could be interpreted separately and offer different conclusions about table 4 core-periphery analysis results network core-periphery correlation # core members gov in core %gov in core reporting 0.568 30 20 66.7% project coordination 0.610 47 27 57.4% information sharing 0.699 49 23 46.9% financial resource sharing 0.535 39 23 59.0% technical assistance 0.659 47 25 53.2% union 0.700 52 25 48.1% 62194-15-25.indd 77 18/08/15 10:30 am 78 s. scheinert et al. / the shape of watershed governance the network. the consistency in the patterns of variation, even while that pattern is nuanced, is what allows these results to be interpreted as identifying the boundaries between the functional subnetworks of a governance network. regardless of the amount of latitude organizations have in choosing their own ties across all the subnetworks, they will have the least latitude in the reporting subnetwork. this is the subnetwork that will be the most affected by mandated relationships, as much of the reporting regime exists between legally-defined and established governmental programs within the nested hierarchies of state and federal government (osherenko, 2014). further, the core of the financial resource sharing subnetwork is dominated by these same government programs, which routinely require evidence of how funds are used by those who receive those funds. this requirement links the reporting and financial resource sharing to each other, while also separating them from the sections of the network where organizations are freer to act. the financial resources and reporting subnetworks are the subnetworks through which money flows and performance is monitored. organizations active in this subnetwork can be expected to be the most effective in implementing policy and completing projects as they will have the most resources (scheinert & comfort, 2014). the reporting and financial resource sharing subnetworks also have the least correlation with information sharing, which is the most important subnetwork for supporting the operation of all other subnetworks (kenis & knoke, 2002). a greater correlation between the financial resource sharing and information sharing subnetworks would indicate that organizations are finding resources from a greater variety of sources, and so more aggressively matching the funds in the network to those who can best use the funds. improving this connection would effectively increase funding in the system by improving the efficiency by which funds are used. 5. conclusion currently, in the lake champlain basin, internal boundaries within the watershed governance network separate the financial resource sharing and reporting subnetworks from the information sharing, technical assistance, and collaboration subnetworks. the results of this study show that there is some separation of information sharing from the governance network’s other subnetworks, a division which could expand or contract over time. however, the current data are cross-sectional data, so they cannot provide any guidance on this change. the two sets of subnetworks supported by these data show an internal consistency. as discussed above, the combination of the financial resource sharing and reporting network takes the form of a semi-official network where formal accountability is important. activities like information sharing and project collaboration, though not without forms of accountability, are typically less hierarchical activities. nevertheless, access points for financial resources are limited and focused on government agencies, which demand accountability for the use of their funds. the need for resources to support technical assistance and collaboration forces actors to work in different parts of the watershed governance network. 62194-15-25.indd 78 18/08/15 10:30 am s. scheinert et al. / the shape of watershed governance 79 other governance networks may not have this distinction as strongly as the watershed governance network in the lcb does. each governance network must be examined individually for its own internal boundaries to be recognized. this research suggests a new approach for how this analysis can be done. this research shows that structural analysis of the subnetworks of a multiplex network can reveal small but important differences in the structures of these networks. these small differences represent important variations that but are too small to be reliably recognized by any one network analytic technique, variations that change the relationships between functional tie types that researchers apply and the operational pictures that managers face. instead, triangulation of techniques is required to give managers and researchers the information they both need to build situational awareness. the scale free network analyses do show a different pattern for the financial resource sharing subnetwork and the reporting subnetwork. additionally, where the information sharing, technical assistance, and coordination subnetworks have nearly identically-sized cores, the reporting and financial resource sharing subnetworks differ in core size by nearly ten organizations. in both cases, the financial resource sharing subnetwork is closer in structure to the subnetworks in the other grouping than it is the reporting subnetwork, though the financial resource sharing is still more similar to the reporting subnetwork than to the subnetworks of the other grouping. while the patterns so far discussed are still observed and so the boundaries are still supported, this break from consistency limits the certainty of these results and suggests that the boundaries may be porous or shifting. further research using this governance network as a case study should either use the two groups that we argue for or use three groups, one which includes information sharing, technical assistance, and coordination, and then two others that treat reporting and financial resource sharing individually. governance networks encode multiple types of interactions and so generate functional subnetworks. few studies to date have made thorough use of multiplex ties, preferring to select and analyze only one type of network or another. there is good reason for adhering to this limit (butts, 2008, 2009). even in exploring the relationship between functional subnetworks, this study must still adhere to this limit in defining its initial set of subnetworks. adhering to this limit does prevent researchers from examining questions that cross the boundaries between types of network interactions. without a way to explore the boundaries between functional subnetworks, no language can be developed for describing these boundaries, testing for their existence, or interpreting their meanings and impacts within a given multiplex networks. being able to address questions about the relationships between the functional subnetworks of a governance network will allow researchers to improve our understanding of governance networks generally and it will allow managers in governance networks to improve their situational awareness and so manage more effectively. 6. acknowledgement this document was developed by support provided by vermont epscor with funds from the national science foundation grant eps-1101317. 62194-15-25.indd 79 18/08/15 10:30 am 80 s. scheinert et al. / the shape of watershed governance references agranoff, r. (2004). leveraging networks: a guide for public managers working across organizations. in j. m. kamensky & t. j. burlin (eds.), collaboration: using networks and partnerships (pp. 61–102). lanham, md: rowman & littlefield. baltazar, r., & brooks, m. r. (2007). chapter 17 port governance, devolution and the matching framework: a configuration theory approach. research in transportation economics, 17, 379–403. doi:10.1016/ s0739-8859(06)17017-1 barabasi, a-l., & albert, r. (1999). emergence of scaling in random networks. science, 286(5439), 509–512. doi:10.1126/science.286.5439.509 betsill, m. m., & bulkeley, h. (2004). transnational networks and global environmental governance: the cities for climate protection program. international studies quarterly, 48(2), 471–493. doi:10.1111/j.0020-8833.2004.00310.x borgatti, s. p., everett, m. g., & freeman, l. c. (2002). ucinet for windows: software for social network analysis [computer software]. harvard, ma: analytic technologies. borgatti, s. p., everett, m. g., & johnson, j. c. (2013). analyzing social networks. los angeles, ca: sage. burt, r. s. (1978). cohesion versus structural equivalence as a basis for network subgroups. sociological methods & research, 7(2), 189–212. doi:10.1177/004912417800700205 burt, r. s. (1987). social contagion and innovation: cohesion versus structural equivalence. american journal of sociology, 92(6), 1287–1335. retrieved from http://www.jstor.org/stable/2779839 butts, c. t. (2008). social network analysis: a methodological introduction. asian journal of social psychology, 11(1), 13–41. doi:10.1111/j.1467-839x.2007.00241.x butts, c. t. (2009). revisiting the foundations of network analysis. science, 325, 414. doi:10.1126/ science.1171022 carley, k. m. (2001–2011). *ora [computer software]. pittsburgh, pa: center for computational analysis of social and organizational systems (casos), carnegie mellon university (cmu). cash, d. w., clark, w. c., alcock, f., dickson, n. m., eckley, n., guston, d. h., . . . mitchell, r. b. (2003). knowledge systems for sustainable development. proceedings of the national academy of sciences (pnas), 100(14), 8086–8091. doi:10.1073/pnas.1231332100 cohen, a., & davidson, s. (2011). the watershed approach: challenges, antecedents, and the transition from technical tool to governance unit. water alternatives, 4(1), 1–14. retrieved from http://www.water alternatives.org/index.php/alldoc/articles/vol4/v4issue1/123-a4-1-1 comfort, l. k., oh, n., ertan, g., & scheinert, s. (2010). designing adaptive systems for disaster mitigation and response: the role of structure. in l. k. comfort, a. boin, & c. c. demchak (eds.), designing resilience: preparing for extreme events (pp. 33–61). pittsburgh, pa: university of pittsburgh press. davis, g. b., & carley, k. m. (2008). clearing the fog: fuzzy, overlapping groups for social networks. social networks, 30, 201–212. doi:10.1016/j.socnet.2008.03.001 diesner, j., & carley, k. m. (2011). semantic networks. in g. barnett (ed.), encyclopedia of social networking (pp. 766–769). los angeles: sage publications. duch, j., & arenas, a. (2005). community detection in complex networks using extremal optimization. physical review e, 72(2 pt. 2), 027104. dutcher, d. d., & blythe, d. j. (2012). water pollution in the green mountain state: a case study of law, science, and culture in the management of public water resources. vermont journal of environmental law, 13. retrieved from http://vjel.vermontlaw.edu/publications/water-pollution-in-the-green-mountain-state-a-casestudy-of-law-cience-and-culture-in-the-management-of-public-water-resources/ fortunato, s. (2010). community detection in graphs. physics reports, 486, 75–174. doi:10.1016/j.physrep. 2009.11.002 gregory, s. (2011). fuzzy overlapping communities in networks. journal of statistical mechanics: theory and experiment, p02017. doi:10.1088/1742-5468/2011/02/p02017 hajer, m., & versteeg, w. (2005). symposium: governance networks performing governance through networks. european political science, 4, 340–347. doi:10.1057/palgrave.eps.2210034 62194-15-25.indd 80 18/08/15 10:30 am s. scheinert et al. / the shape of watershed governance 81 hirschi, c. (2010). strengthening regional cohesion: collaborative networks and sustainable development in swiss rural areas. ecology and society, 15(5). retrieved from http://www.ecologyandsociety.org/vol15/ iss4/art16/hutchins, e. (1995). cognition in the wild. cambridge, ma: mit press. imperial, m. t. (2005). using collaboration as a governance strategy: lessons from six watershed management programs. administration & society, 37(3), 281–320. doi:10.1177/0095399705276111 johnson, j. c. (1990). selecting ethnographic informants. newbury park, ca: sage. jost, g. f., & jacob, k. (2004). the climate change policy network in germany. european environment, 14, 1–15. doi:10.1002/eet.337 kenis, p., & knoke, d. (2002). how organizational field networks shape interorganizational tie-formation rates. the academy of management review, 27(2), 275–293. doi:10.5465/amr.2002.6588029 klijn, e-h., edelenbos, j., & steijn, b. (2010). trust in governance networks: its impacts on outcomes. administration & society, 42(2), 193–221. doi:10.1177/0095399710362716 klijn, e-h., & skelcher, c. (2007). democracy and governance networks: compatible or not? public administration, 85(3), 587–608. doi:10.1111/j.1467-9299.2007.00662.x klijn, e-h., steijn, b., & edelenbos, j. (2010). the impact of network management on outcomes in governance networks. public administration, 88(4), 1063–1082. doi:10.1111/j.1467-9299.2010.01826.x koliba, c., meek, j. w., & zia, a. (2010). governance networks in public administration and public policy. boca raton, fl: crc press. koliba, c., reynolds, a., zia, a., & scheinert, s. (2015). isomorphic properties of network governance: comparing two watershed governance initiatives in the lake champlain basin using institutional network analysis. complexity, governance and networks, 1(2). doi:10.7564/14-cgn12 koontz, t. m., & johnson, e. m. (2004). one size does not fit all: matching breadth of stakeholder participation to water shed group accomplishments. policy sciences, 37(2), 185–204. doi:10.1023/b:olic.0000048532. 94150.07 lake champlain basin program. (2012). state of the lake and ecosystem indicators report—2012. grand isle, vt: author. retrieved from http://sol.lcbp.org lauman, e. o., marsden, p. v., & prensky, d. (1989). the boundary specification problem in network analysis. in l. c. freeman, d. r. white, & a. k. romney (eds.), research methods in social network analysis (pp. 61–87). fairfax, va: george mason university press. lienert, j., schnetzer, f., & ingold, k. (2013). stakeholder analysis combined with social network analysis provides fine-grained insights into water infrastructure planning processes. journal of environmental management, 125, 134–148. doi:10.1016/j.jenvman.2013.03.052 lubell, m., & fulton, a. (2008). local policy networks and agricultural watershed management. journal of public administration research and theory, 18(4), 673–696. doi:10.1093/jopart/mum031 luokkala, p., & virrantaus, k. (2014). developing information systems to support situational awareness and interaction in time-pressuring crisis situations. safety science, 63, 191–203. doi:10.1016/j.ssci.2013.11.014 marsden, p. v. (2005). recent developments in network measurement. in p. j. carrington, j. scott, & s. wasserman (eds.), models and methods in social network analysis (pp. 8–20). new york, ny: cambridge university press. mcguire, m. (2002). managing networks: propositions on what managers do and why they do it. public administration review, 62(5), 599–609. doi:10.1111/1540-6210.00240 milward, h. b., & provan, k. g. (2006). a manager’s guide to choosing and using collaborative networks. ibm center for the business of government. retrieved from http://www.businessofgovernment.org/ report/managers-guide-choosing-and-using-collaborative-networks newig, j., guenther d., & pahl-wostl, c. (2010). synapses in the network: learning in governance networks in the context of environmental management. ecology and society, 15(4). retrieved from http://www .ecologyandsociety.org/vol15/iss4/art24/ newman, m. e. j. (2006). modularity and community structure in networks. proceedings of the national academy of sciences (pnas), 103(23), 8577–8582. doi:10.1073/pnas.0601602103 norman, e. s., & bakker k. (2009). transgressing scales: water governance across the canada–u.s. borderland. annals of the association of american geographers, 99(1), 99–117. doi:10.1080/00045600802317218 62194-15-25.indd 81 18/08/15 10:30 am 82 s. scheinert et al. / the shape of watershed governance osherenko, g. (2014). understanding the failure to reduce phosphorous loading in lake champlain: lessons for governance. vermont journal of environmental law, 15. retrieved from http://vjel.vermontlaw.edu/ publications/?volumes=volume-15 ostrom, e. (2012, march). enhancing the evolution of institutions for collective action. in the evolution institute, social evolution forum. retrieved from https://evolution-institute.org/focus-article/elinor ostrom-enhancing-the-evolution-of-institutions-for-collective-action/ provan, k. g., fish a., & sydow, j. (2007). interorganizational networks at the network level: a review of the empirical literature on whole networks. journal of management, 33, 479–516. doi:10.1177/0149206307302554 provan, k. g., & kenis, p. (2008). modes of network governance: structure, management, and effectiveness. journal of public administration research and theory, 18(2), 229–252. doi:10.1093/jopart/mum015 provan, k. g., & lemaire, r. h. (2012). core concepts and key ideas for understanding public sector organizational networks: using research to inform scholarship and practice. public administration review, 72(5), 638–648. doi:10.111/j.1540-6210.2012.02595.x provan, k. g., & milward, h. b. (2001). do networks really work? a framework for evaluating public sector organizational networks. public administration review, 61(4), 414–423. doi:10.1111/0033-3352.00045 pustejovsky, j. e., & spillane, j. p. (2009). question-order effects in social network name generators. social networks, 31, 221–229. doi:10.1016/j.socnet.2009.06.001 rykkja, l. h., neby, s., & hope, k. l. (2014). implementation and governance: current and future research on climate change policies. public policy and administration, 29(2), 106–130. doi:10.1177/0952076713510344 sailer, l. d. (1978). structural equivalence: meaning and definition, computation and application. social networks, 1, 73–90. doi:10.1016/0378-8733(78)90014-x scheinert, s., & comfort, l. k. (2014). finding resilient networks: measuring resilience in post-extreme event reconstruction missions. in n. kapucu & t. liou (eds.), disasters and development (pp. 181–199). new york, ny: springer. schneider, m., scholz, j., lubell, m., mindruta, d., & edwardsen, m. (2003). building consensual institutions: networks and the national estuary program. american journal of political science, 47(1), 143–158. doi:10.1111/1540-5907.00010 scholz, j. t., berardo, r., & kile, b. (2008). do networks solve collective action problems? credibility, search, and collaboration. journal of politics, 70(2), 393–406. doi:10.1017/s0022381608080389 scott, j. (1991). social network analysis: a handbook. london, england: sage. sorenson, e., & torfing, j. (2005). the democratic anchorage of governance networks. scandinavian political studies, 28(3), 195–218. doi:10.1111/j.1467-9477.2005.00129.x sorenson, e., & torfing, j. (2009). making governance networks effective and democratic through metagovernance. public administration, 87(2), 234–258. doi:10.1111/j.1467-9299.2009.01753.x telesford, q. k., joyce, k. e., hayasaka, s., burdette, j. h., & laurienti, p. j. (2011). the ubiquity of smallworld networks. brain connectivity, 1(5), 367–375. doi:10.1089/brain.2011.0038 van de walle, b., & turoff, m. (2008). decision support for emergency situations. information systems and e-business management, 6(3), 295–316. doi:10.1007/s10257-008-0087-z vignola, r., mcdaniels, t. l., & scholz, r. w. (2013). governance structures for ecosystem-based adaptation: using policy-network analysis to identify key organizations for bridging information across scales and policy areas. environmental science and policy, 31, 71–84. doi:10.1007/s10257-008-0087-z wasserman, s., & faust, k. (1994). social network analysis: methods and applications. new york, ny: cambridge university press. watts d. j., & strogatz s. h. (1998). collective dynamics of “smallworld” networks. nature, 393, 440–442. doi:10.1038/30918 weible, c. m., & sabatier, p. a. (2005). comparing policy networks: marine protected areas in california. policy studies journal, 33(2), 181–201. doi:10.1111/j.1541-0072.2005.00101.x 62194-15-25.indd 82 18/08/15 10:30 am case-based complexity: within-case time variation and temporal casing 29 university of bamberg press case-based complexity: within-case time variation and temporal casing authors: lasse gerrits, robin a. chang, sofia pagliarin institute for housing and urban development studies of the erasmus university rotterdam, the netherlands; technical university dortmund, germany; erasmus university rotterdam, the netherlands email: gerrits@ihl.nl we observe that time is central to most social dynamics and yet remains poorly understood. the complexity sciences have contributed a wide range of concepts and tools to investigate and recast time in social systems. however, the dominant focus on quantitative models and quantification of data in the complexity sciences also prohibits a deeper understanding of time. as such, there is a need to fuse alternative notions of time with how it is commonly understood and measured in the complexity sciences. to this end, we juxtapose diverse notions of time from the social sciences and comment upon how this contrasts with notions in the complexity sciences. we will demonstrate how (qualitative) temporal casing can more appropriately capture social and causal complexity through within-case time variation. we use examples from research into megaprojects to demonstrate how temporal casing plays out in empirical analysis. keywords: time; case study research; methods; complex causality 1 on time and case-based complexity it is one thing to say that social reality is complex. it is, however, quite another to pinpoint said complexity in such a way that it can be conceptually articulated and mapped in empirical social research. for social science, it is in the probabilistic following-up of events through time that complex causality becomes apparent as multiple and inherently uncertain future directions for observations or cases (prigogine, 1997; for a re-elaboration, see gerrits & pagliarin, 2020). in other words, the future is highly unpredictable because every event can be followed by several possible subsequent events with a different degree of plausibility. hence, it comes without a surprise that early scholars cast minds to the question of how to deal with time as the locus of causal and social complexity. compte conceptualised social change in terms of evolutionary stages centred on the specific development of western countries, while simmel’s analysis of modern urban life highlighted the differences between the non-urban, pre-modern ordinary man and the typical homme blasé of the “metropolis” (simmel, 1903 ed. 1995). many scholars have found ways to express time in their research, primarily through quantification, distribution of phenomena over regular intervals, additional narratives, and through abstract conceptualisations. yet, “time” manages to elude social scientists when it comes to ontology, epistemology, and research methods: fixed time points may fail to capture shifts or the clustering of events in episodes. alternatively, stages may feel arbitrary, while quantification tends to black box considerable social complexity, and qualitative approximations of change and stability seem clumsy and lack the rigour needed to move beyond truisms. the challenges and limitations for achieving conceptual dexterity in accounting for time in social research might never have a final and uncontested answer, but this should not discourage scholars from engaging with the topic. rather, we might find a range of theoretical concepts in the complexity sciences that enhance our understanding of time in the social realm, including emergence, non-linearity, tipping points and punctuated equilibrium – as phenomena playing out over time and to be understood only through time. these concepts also highlight the irregular, multifinal, subjective, and volatile nature of time. there is no doubt that the complexity sciences continue to enrich the conceptual toolkit of the social sciences. however, and inevitably, they are limited in some respects. as complexity relies predominantly on quantification and systems of equations, there is still much work to do in accounting complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 mailto:gerrits@ihl.nl 30 university of bamberg press for the qualitative aspects of time, which characterise social research. the goal of this paper is to propose a case-based conceptualisation of time that acknowledges the irregularity and equifinality of durations as units of time. to this end, we juxtapose diverse notions of time from the social sciences and comment upon the limitations of how time is understood in the complexity sciences. this sets up explications for how case-based research can more appropriately capture social and causal complexity through within-case time variation. this is achieved via temporal casing. we will identify various methodological operations to guide empirical research by re-elaborating on a well-known analysis in megaprojects management (flyvbjerg, bruzelius, and rothengatter 2003). we believe there is much to be gained from the dialogue between case-based research and complexity and will derive methodological implications from such a dialogue. in our analysis, we observe that time is usually treated as sui generis or as a given and uniform measure—structured and aligned with—but external to social processes (elchardus, 1988). some accounts skirt temporal dimensions via other concepts such as structure (nowotny, 1992). many, if not most approaches to time, rest on conventional measurements rather than on following the dynamics of how time passes. these conventions are deeply ingrained in the practice of research, for example the use of clock time or calendar time to structure narratives or to distribute panel data, as well as simulations that use iterations spaced on a uniform timescale (e.g., one day or month’s “time ticks”). conventions exist for a reason: that is, they serve as workable proxies to account for time in empirical research. as they stem from a more or less stable consensus within academic communities in different fields, these conventions also impose a regimentation on data. nevertheless, such conventions may not reflect the actual nature of social processes that data supposedly represent. what is needed, then, is greater reflexivity in how researchers address time in their work (adam 2003). in view of this, we argue for an understanding of temporal units as embedded in social processes; these units of time reflect irregularity in duration and equifinality (see ragin, 2008). this approach to understanding and analysing time has implications on the comparative study of sequences of events and phenomena; it also opens the possibility to conceptualise time that is made, not only given. we root these claims in the critical examination of how time is structured and set in empirical research. instances of these include how specific temporal horizons and norms define the research question at hand, as well as a reconsideration of the methods used (altomonte, 2016; aeon & aguinis, 2017). the kind of processes we focus on can be studied in their own right—e.g., via process tracing (beach & pedersen, 2013) and event-sequence analysis (see abbott, 1995) or comparatively (bengtsson & ruonavaara, 2011; pagliarin & gerrits, 2020; ragin, 2008). of key importance here is the notion that complex systems nest within other systems, and have systems nested within them (byrne, 1998). this perspective implies that any case takes place within a variable context, thus explaining why a casual observation in one context does not necessarily hold in another. it then appears methodologically plausible that empirical research may also adopt a case-based approach to explain and differentiate between contexts. beyond the differentiation (and at times, opposition) between methodological standpoints (i.e., the focus on rich individual histories of within-case research and on general patterns of cross-case research), we claim that within-case complexity of social processes can be an entry point to reach cross-case complexity. an essential dimension of this complexity is time and how it is methodologically characterised; the subsequent sections will address. we first present an overview of some of the main scholarly alternatives to time and space, which is followed by a discussion on how time and space are accounted for in the complexity sciences. next, we present the main properties of temporal casing to account for within-case variation. we then re-cast the notion of temporal casing in the context of managing megaprojects to demonstrate how such a refocusing alters the ways in which the complexity of developments may be understood. this paper does not present a conventional case study but rather opens a conversation about time and complexity. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 31 university of bamberg press 2.1 conceptualisations of space and time 2.1.1 a brief incursion into the debates about space the claim that “we must get to know […] invisible time and recognise its fundamental role in the constitution of the lives and institutions of those we study” (adam, 1995, pp.41–42) marks the starting point for our argument. in developing this argument, we first look to the critical re-considering of space and its units of analysis. surprising as this may seem, it shows how researchers’ choices socially construct definitions of space, which may aggregate as consensus and reify as conventions. this may inspire a shift in how we currently conceptualise and account for time. it also compels us to take stock of the various ways through which time is considered in social reality by examining the transition from a pre-industrial to an industrial society. this transition introduced not only the clock, but also technological and social standardisation as well as systematisation; these processes have since imposed a regime on the social world—irrespective of its appropriateness. time zones or standard working hours are examples of and outcomes from such standardisations and systematisations. the ones mentioned here are but a few key developments that have left us unwittingly inattentive to considering time beyond notions of linearity or regularities. in parallel, the conceptualisation and measurement of spatial units has engaged scholars of diverse backgrounds, ranging from statisticians and geographers to economists and sociologists. from their engagement grows increasing acceptance and awareness for how spatial units and spatial representation influence findings on spatial phenomena (kitchin & dodge, 2007). international comparative research showed the fallacy of using similar statistical classifications (often based on conventional administrative units) in different contexts. spatial units may be called the same term but represent quite differently when comparing countries. for instance, the definition of what a village, a town, or a city means varies substantially across european countries (le gléau, pumain, and saint-julien 1997). also, data comparability across the different classifications of urban areas around the globe (e.g., between europe and africa) can call into question the current rhetoric of the “urban age” as a key characteristic of the 21st century (brenner & schmid, 2014). as a result, the definition (epistemology) and measurement (methodology) of spatial phenomena have crucial consequences for the way space is represented and understood. both definitions and measurements rest on conventions and consensus among scholars. more importantly, researchers inform conventions and consensus through the choices they make regarding their focus of analysis. again, by referring to le gléau et al. (1997) to illustrate this point, the delimitation of comparable metropolitan areas in europe has resulted in two different spatial units (through consensus) by combining the approaches (through conventions) of the french statistical institute (insee) and the us census. the latter follows the application of functional urban regions (furs) proposed by hall and hay (1980). as approximations of reality, spatial units can be multiple and equally correct. the consequence of their consensual and conventional character is that scholars change, update1, and continuously open the definition of spatial units for debate (see roca cladera, 2003) as alternative classifications are proposed—e.g., the complexitybased, fractal approach to the delimitation of metropolitan areas as proposed by (tannier & thomas, 2013). in this regard, researchers need to ask themselves: which “space” is their “area of interest”? which scale(s) best represents spatial phenomena? or, to what extent are administrative boundaries sufficient in capturing spatial relations and processes? spatial measurements and maps, which are “a subset of all potential mappings” (kitchin & dodge, 2007, p. 337), result from socially constructed processes composed by a diverse set of practices. in sum, defining spatial units is an epistemological process enabling us to represent spatial phenomena, which 1 the two spatial units for metropolitan areas identified by le gléau et al. (1997) have been discontinued and have been replaced by the nomenclature of territorial units for statistics (nuts) and by the urban atlas classification (eurostat 2020). complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 32 university of bamberg press relies on methodological choices regarding the measurement and selection of several indicators (e.g., population), including temporal data that can be either static (one time point) or dynamic (multiple time points, e.g., commuting flows, see below). 2.1.2 various conceptualisations of time this brief discussion of the ambiguity that comes with delineating something as ostensibly clear as a physical space shows that spatial units are, even at best, provisional results of epistemological and methodological processes within the scientific community (see e.g. latour, 1987). further, researchers’ practices have brought conventions and consensus to bear on these processes. comprehending this compels scholars to interrogate their own ideas and practices that delineate (spatial and temporal) units of analysis as the basis of their investigations. by translating these epistemological and methodological observations on how we conceptualise space to how we conceptualise time, we can also reasonably ask, “which ‘time’ is this case?” to structure our response to this question, we first discus the historic process that encouraged unmindful and continued repackaging of temporal qualities and patterns. following this, we turn to scholarship from sociology, geography, anthropology, philosophy, economics, and psychology that have progressed methods and theories to account for time or temporality in society. investigations on time have moved from initially interrogating the social use of time, to focusing on the disparity between how we relate temporal meanings to social reality and history (see table 1). these strands have developed independently from each other and consider time primarily in axiomatic manners (adam, 2003; elchardus, 1988). we begin with durkheim—a deliberator of things social, including the nature of time (bergmann, 1992; cheng, 2017; zerubavel, 1985). according to zerubavel (1985), durkheim’s pioneering and sociological impressions on religion uncovered foundational devices and behaviors that gave rise to, and institutionalised temporal regularity. the linear, progressive, non-repeatable idea of time developed by modern societies, typically represented by an arrow, emerged in opposition to the circular and cyclical conceptualisation of time from pre-modern societies (eliade, 1965). the calendar, together with the clock (simmel, 1903 ed. 1995), has become the epitomical technology anchoring a “time grid” in western societies’ understanding of time (adam 1995, p. 20). permeating studies of time that underline unidirectional orientations towards the futures, linear and regimented conceptualisation of time highlight irreversible and path-dependent events, upon which we rely to track and study time (bergmann, 1992; van tienoven, 2019). in essence, our understanding of time implies a constant motion for which there is no obvious finality, albeit pressed on by present perceptions and continuous emergence of temporal realities (van tienoven, 2019). as already highlighted by simmel (1903 ed. 1995), modern time technologies have been key in supporting this directed conceptualisation of time. modern time technologies are functional to the planning and controlling of time and thus essential for how we fundamentally understand time and its flows (adam, 2006). put differently, these technologies are at the crux of societies’ continuous desire for, and ability to track, measure, anticipate and influence how other social processes engage with, or occupy time or “reckon” time (adam, 1995, p. 26). as an example, time geographic heuristics show that the representation of individual paths by line are shaped by technologies as coupling constraints or by spatialised forces (authority constraints; see (hägerstrand, 1970; ellegård, 2018). while time geographic heuristics make the abstract conceptualisation of time through space possible (e.g., measuring individual paths through points representing spaces), it also reproduces conventional ways of thinking about time and space. namely, as linear patterns that representing time as intervals of movement in space or as durational points of activity in place. an outcome of these developments is an “artefactual” sense of time, which “constitutes our most taken-for-granted reality” (adam, 1995 pp. 25–26). this sense of time departs in a “direction of ever greater complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 33 university of bamberg press abstraction, emancipating itself from local particularities.” durkheim (1915) highlighted this as embodied in local and religious sovereignty through his work. , this departure and change in understanding time initiated the naturalisation of a limited way to account for time—reinforcing linear and regulated conceptualisations of time and time-based institutions; these have further enforced temporal conventions “as given and unalterable” (adam, 2003, p.60; van tienoven, 2019). our abilities to investigate and explicitly articulate considerations of time could be considered modern but are far from sophisticated. from a bird's eye perspective, it appears that current and conventional ways to account for time are a part of “social time” rooted in religious developments that created and ingrained an “impersonal set of indispensable guidelines for daily life that transcend the individual” (van tienoven 2019, p. 976). this impersonal and collective conceptualisation of time has then provided the background for conceiving time in (post-)industrial societies. post-industrialisation, urbanisation, globalisation, and even climate change are contexts through which we are developing finer-grained temporal awareness (adam, 2003; pahl, sheppard, boomsma, and groves,. 2014). studies in the aforementioned contexts slowly patch the pre-existing and historically accepted understandings of time that distanced and numbed personal as well as complex accounts for time. recall how processes of industrialisation administered and streamed collective and eventually conventional understandings of time. these effects were most keenly perceived through fordism and taylorism, which standardised concepts for how we structure(d) labour time; currently these dominating conventions are subject to increasing changes in valuation and structuration as a result of post-fordism and networked flexibilisation of time and activities (adam, 1995, 2003). in regulating how societies have measured time, these processes not only created clear structuring devices (shifts, work periods etc.), they were key in the systematisation and universalisation of temporal conventions (madanipour, 2017). how else could we now have need for time zones and daylight savings time, if not to accommodate the expansion and coordination required for the innovations of industrialisation? not only did these layered developments scaffold the means to track time cycles—both social and natural (such as those governing agricultural seasons and natural biorhythms), but they also bolstered a quantitative subdivision of time (i.e. hours, minutes, and seconds, etc.) that has become unquestionably superimposed onto phenomena (madanipour, 2017). through this mechanistic lens, even understandings for natural cycles have become amounts and rates, for which the full effects of these changes are still unclear or unpredictable (adam, 1995). in sketching out this legacy in accounting for time, we root our disagreements in the enduring preferences to quantify time, which most notably sprung from the chronological ordering or sequencing of time to support production-line processes. while initial interests were for efficiency and control, this capacity has shifted focus to leverage the exchange value of time for labour; put simply, time is no longer about labour but has become a commodity (adam, 1995, pp. 25–26). with the aims to mark uniform units of work represented by shifts or interrupted by pauses, we have come to receive time as an almost prescriptive framework of “temporal norms” assumed through economically set tempos, sequences, and durations. clear illustrations of the extent for which social lives are regimented by “clock-like” time are the strictly punctuated activities that not only follow a certain order, but also produce inescapable routines (goffman, 1961 ed. 2007). time regimentation also pervades the quintessential form of work-free time: the holiday. for example, minca (2009) examines the paradox of regulated activities and routines that characterise leisure time in some of the most common spaces for leisure, such as cruises, campsites, holiday villages, and hotels. less so appreciated are the naturally rooted understandings of time governed by biological or environmental cycles of time-keeping (elchardus, 1988). indeed, social time contours and possibly constrains a fully objective understanding and study of time beyond our collective and internalised experiences with time (zerubavel, 1982). in transport and mobility studies, scholars differentiate between perceived and objective measures of travel time. as self-reported trip duration constitutes the main data source for surveys and travel diaries, scholars are however aware that such data are based on travelers’ complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 34 university of bamberg press memories and perceptions (perceived measure) that might not correspond to the real duration of travel time (objective measure (delclòs-alió, marquet, and miralles-guasch 2017)). the increasing availability and pervasiveness of smartphones with built-in gps-tracking greatly helps us to estimate time misperceptions while also identifying possible factors that generate discrepancies. for instance, delclòsalió et al. (2017) showed that longer trips do not only present greater misperceptions, but also an underperceived travel time. kolarova et al. (2019) investigated how the subjective perception of travel time changes as an effect of the multiple ways through which digitalisation and automation of private mobility decreased the need for active driving while, at the same time, increased the potential for performing other activities while travelling. millonig et al. (2012) explored the influence of different entertainment offerings on the perception of waiting times at transit stations among a sample of public transport passengers. the difference between perceived and objective measures of time relates to another more subtle quality of time: range in velocity. tempos are particularly slow for when we wait (wexler 2015) or are perceived to progress faster when time is commodified or tightly structured as has been the case after the industrial revolution (rosen, 2012). in short, we are increasingly competent at distinguishing comparable regularity in duration and interval (rhythm), synchronicity (attunement), or simultaneity (density) of how events and phenomenon flow in time (levine, 2006). moreover, we are also progressively reliant on the “temporal location” of events that we infer through other event-based markers (provonost, 1989; zerubavel, 1985). again, time geographic heuristics such as the bundling of individual and linear paths emphasise how we sequentially anchor or mark spatial and temporal locations in relation to each other ( ellegård , 2018, p. 7. put differently, sometimes it is not so much that time comes to an end for an activity (intrinsic/intentional delimiting of duration) as it is that the time for another event is beginning (external/inflicted foreclosing of duration). for instance, the blurred boundaries that many experience in working from home are disadvantageous for those who operate better with external markers or sources of timing embodied as “zeitgebern or time-givers” (parkes & thrift, 1979, p.356). when individuals have a weak sense of temporal awareness or do not have to self-regulate time, then a stronger reliance develops through which necessary activities, such as the transition of a commute to an office building, help dictate when certain rituals, such as morning preparations, conclude while others, such as work, begin. these illustrate concatenation of activities or fluid temporal boundaries. all this underlines the prevalence of certain conceptualisations of time that also feedback into and shape our experiences of time. the independent development of time conceptualisations (elchardus, 1988; adam, 2003) on the social use of time (linearity, clock-time, global time coordination, regimentation of work and leisure time) and the subjective perceptions of time (meanings related to time according to specific activities) have effects that ripple and swell through all manners of practices. these practices include those involved in epistemological and methodological processes, which produce unquestioned definitions and measurements of time. these practices, in aggregate and particularly in research, also leave us with convenient consensus and less than accurate conventions that the predominantly positivist worldview only reinforces. by allowing science to lean on particular understandings of time, we are also yoking scientific investigation and reporting to structures we must remember not to take for granted. a key illustration of our argument can be found in time series analysis. this form of analysis deals with the application of statistical methods to data whose observations are not independent and identically distributed, but that are autocorrelated and covary as they are a sample of adjacent points in time (shumway & stoffer, 2017). a time series is formed by a collection of assumed random variables (columns) indexed according to the order in which they are obtained in time and labelled as t=1,2, ... n, where each observation (rows) is conceptualised as an array of time points. key concepts in time series analysis are sampling length and rate. on the one hand, particular discipline-specific time intervals (sampling length) often define time points. time intervals can be, for instance, expressed in years and quarters (e.g., economics), days or hours (e.g., finance), or seconds or decimals of seconds (e.g., speech analysis), but also through cycles or periodic behaviour (e.g., annual or multiple-year cycles in climate research) or as complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 35 university of bamberg press specified in experimental settings (e.g., stimulus given every period of x seconds or minutes). on the other hand, sampling rate refers to the frequency of data sampling within a specific time interval (sampling length), for instance 20 observations every quarter, 5 observations every hour, or a single observation every 3 seconds. time series data can be continuous or composed of discrete parameters depending on whether the sampling rate is higher or lower. the point is that the conceptualisation and assessment of time in certain disciplines must draw and construct its definition from external and social conventions. broadly speaking, even in the highly formalised approach to time series analysis, time is not given, but is “constructed”: it needs to be adapted in terms of sampling length and rate to the phenomenon at hand in order for researchers identify patterns. in time series analysis too, we can recognise conventions, consensus, and research practices because the definition of sampling length and rate can vary and is discipline specific. furthermore, the sampling rate is particularly important in time series because data behaviour can appear completely different if different sampling rates are used. timeseries scholars know that one vs. six-month sampling rates can show different patterns (e.g., peaks) or dynamics (e.g., seasonality). an inappropriate sampling rate is associated with the serious problem of “aliasing,” meaning that the phenomenon under analysis is falsely represented (see lentz, 1990). to avoid this, adequate sampling in time series involves both time intervals with a sufficient length and an appropriate amount of data points (sampling rate). again, what “sufficient” and “appropriate” mean depends on the conventions, consensus and research practices related to a certain object of analysis. we can, however, recognise a certain degree of “qualitative” reasoning that characterises time series analysis beyond its “objective” definition and measurement of time. naturally, and by carefully avoiding postmodernist drifts, we are not the first to question such conventions. conventions, consensus, and research practices are not inherently wrong, but they have to be acknowledged as characteristics of epistemology and methodology as processes of “doing science” (latour, 1987). as we have seen from the conceptualisations of space, debates about the epistemology and methodology of time open up a healthy dialogue among scholars by revealing limitations or gaps between theoretical and empirical work and alternative conceptualisations (cheng, 2017). time geographers already explicitly acknowledge the lack of analytical rigor in addressing time, particularly with capacities to measure, compute, and compare data coming out of innovations such as location-aware technologies or location-based services (miller, 2005). they also highlighted the selective tendencies to make use of approaches that are incongruent with, and amplify analytical conventions (shen, chai, and kwan, 2015, p.90; ellegård, 2018, pp.7–8). these inclinations reinforce how researchers characterise time by delineating movements or boundaries in space. those delineations are essentially steps of casing (ragin & becker, 1992) to which we will return to further below. first, we need to turn our attention to the complexity sciences and the ways it attempts to deal with time. 2.2 the complexity sciences and time with roots in the so-called “hard” sciences, the complexity sciences have clear positivist and modernist roots. yet, complexity science, because of its anti-reductionist stance, has burned some holes in the modernist regimentation of time, as discussed above. the latter resonated in the social sciences (reed & harvey, 1992) and led some scholars to call the complexity sciences as a post-newtonian science (adam, 1992; heylighen, cilliers, and gershenson 2006). however, the complexity sciences are anything but a coherent body of knowledge—more a patchwork of ideas than one school of thought—although there are some common themes. non-linearity is mentioned frequently, and there are several strands focused on phase shifts and hysteresis (isaac, 1994; see also elster, 1976). these common aspects in the complexity sciences revolve around the irreversibility of time, as discussed by harvey (2009). irreversibility, here, implies that there is asymmetry between past and future because of the occurrence chance and systemic interactions (prigogine & stengers, 1984; complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 36 university of bamberg press prigogine 1997; harvey 2009). more importantly, the recurring theme is that complexity comes from, and can only be understood through, time. while the complexity sciences may stem from the hard sciences and are primarily concerned with the physical, natural world and its development over time, social time may differ in terms of duration and temporal pace, and in the experience of time (newton, 2003), as mentioned above. to the natural scientist, ultimately concerned with finding fundamental natural laws and necessity, the social experience of time is not very relevant because it is seen as a stratum upon fundamental time that obscures rather than clarifies how things play out over time. as much as complexity scientists from the natural sciences make observations fitting to a post-newtonian worldview, they still usually default to strict time regimes as the structuring device for their findings and simulations. more specifically, non-linearity, phase shifts and hysteresis are found as a rate over uniform intervals, and simulations run by uniform ticks. we observe very little—if any—cross-fertilisation between the complexity sciences and the alternative conceptualisations of time as discussed in this paper. this is unfortunate, as the complexity sciences could benefit from those advanced time conceptualisations as much as the social sciences could take advantages of the numerous time-related discoveries coming from the complexity sciences. a deeper and broader engagement with time can awaken the complexity sciences to how various forms or qualities of time enhance one’s understanding of social complexity. as a step in this direction, table 1 presents an initial review of the multiple and diverse alternative conceptualisations of time. these indicate the reinforced and functional conventions of cutting up time in equal durations: granular, specific units that can be ordered sequentially with (in)variance. these various conceptualisations also reveal repeated practices in the social sciences, which study overwhelmingly outcomes of invariance embodied in routines for instance (cullen & godson, 1975). some scholarship emphasises the multiplicity and diversity of ways to consider time. these highlight how different individual and collective perspectives can be objective (chronos), subjective (kairos), both run in parallel or imagined as layered flows (barabara, 2014). complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 37 university of bamberg press table 1. various alternative time and space conceptualisations that may be used in conjunction with the complexity sciences. conceptual orientation key contributors temporal casing plural and simultaneous modalities or structures mead (1932), gurvitch (1964;1973), luhmann (1982), bergmann (1992), cheng (2017) socially interpreted time-systems frame perspectives on time as modalities or scales. for the former, phenomenological references to the present differentiate horizons (both past and future) and thus stages that can be analysed. for the latter, hierarchical strata add a vertical dimension to temporal analysis. regularities and rhythms zerubavel (1982), sorokin and merton (1937), lefebvre (1985), structuring of activity or temporal location and markers (time-budget, schedules) as well as rates of recurrence, rhythmicity (repetition, sequence, frequency, tempo, synchronisation), or rhythmanalysis. space through concept of timegeography hägerstrand (1967; 1970) spatial interpretations of time through individuals’ and activity paths as represented through the prism (this can be aggregated or transposed to other social units), place (pockets of local order), and projects (activities of individuals and groups that are constrained by rules). multiplicity of times in relation to spaces. a critique of time-space binaries through concept of timespace thrift (2005, 2006), may & thrift (2003) does not subscribe to specific temporal casing but draws on other conceptual orientations and their analytical devices (phenomenology, time-geography, rhythms, cycles, tempo, duration, etc.) to facilitate critiques of time-space dualities. processes, structured in eventsequences and subsequences heise (1993), abell (2004), abbott (1983; 1995), sequences of events and sub-sequences that run in parallel or are a subset of main sequences together form the case. there is no strict delineation of the case in time or space, it could be a series of personal events of an individual but also the sequences of events leading to a global outcome (such as the outbreak of wwi). valuation bourdieu (1963), rezsohazy (1970), pahl (2010; 2014), whillans (2020) individuals’ sense of proximity or orientation to events (distant versus close events affect memory and cognitive processing) and their value associated with temporal durations and horizons. temporal casing expresses valuation through economic productivity or holistic wellbeing. perceived and objective measures of time millonig et al. (2012); delclòsalió et al. (2017); kolarova et al. (2019) self-reported trip duration in surveys and travel diaries, perceived time different from real or clock time, gpstracking, different travel perceptions based on different activities. sample of adjacent points in time at t=1,2, ... n lentz (1990); shumway and stoffer (2017) time series data are a sample of adjacent points in time at t=1,2, ... n and are formed by a collection of random variables. time intervals/cycles (sampling length), sampling rate of autocorrelated time points. overall, regularity of patterns is assumed. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 38 university of bamberg press as shown in table 1, we have multiple conceptualisations of time that span from the subjective perception of time to its fragmentation into milliseconds, as discussed in section 2.1. on the one hand, time, as commonly conceived in the complexity sciences, appears to feature few traces of the alternatives presented in table 1, perhaps except for time series analysis approaches that also incorporate objective forms of time measurement, and the idea that certain empirical phenomenon can play out over different but parallel time scales. on the other hand, conventions about how to conceptualise time and space in the social sciences are the result of a historic process also affected by technological trends and devices, like clocks, calendars, and gps tracing, as discussed in section 2.1.2. overall, we observe the prevalence for systematic and uniform measures of time as linear and regular, and specifically represented as durations and intervals. we contend that these conventions, despite their utility, also convey a questionable sense of certitude and regularity in the practices of research. this is the result of how they distance us from the actual whimsical nature of time, even in the realm of the complexity sciences. this paper is an attempt to reveal the analytical blinders we may have created for ourselves; more importantly, it sets the stage for a necessary debate about how the social sciences account for time. instead of limiting ourselves to certain comfortable and consensually accepted conventions of durations and boundaries, let us consider an alternative way to account for time. temporal casing forms a conduit to achieve that goal. 3.1 temporal casing in temporal casing, we suggest that one can approximate the complexities of time by acknowledging within-case variation as time that is segmented (sampling length) from processes according to specific time points/events (sampling rate). within-case variation concerns the temporal changes and logic that is internal to a particular case. time is qualitative and reflecting not uniformity but diversity (characterised by context, interpretation, layered processes, etc.). this also could be conceived methodologically not only through sampling but in temporal casing. “cases” are a core aspect of every social scientific research. indeed, the difference between case study research and non-case study research is an artificial one; all phenomena in the social sciences are cases (gerring, 2004). cases are as ubiquitous as they are ambiguous in how they are defined and handled (lund 2014). a bare-bones definition of cases conceives of them as an in-depth examination of an instance of some phenomenon of interest (babbie, 2013; gerring, 2004). however, stemming from this is a diversity of perspectives (see ragin & becker, 1992). a case can be anything from a person in psychology and medicine to a decision-making process in public administration to a neighbourhood in urban planning. it can be “a single phenomenon, a narrative unit, an instantiation of a class of events” (dumez, 2015, p.127). cases may be delineated by researchers in time, space, size, number, topic, and many other ways—some of which logically incompatible (gerring, 2004). by definition, they are situated within a context to which they are open, reflecting some of the context’s properties. as such, there is a constant blurring of the boundaries between the case and its context (yin 1994). among others, harvey (2009), carter & sealey (2009), castellani & hafferty (2009), and castellani et al. ( 2016) note the correspondence between the properties of complex systems and the properties assigned to cases as in the literature cited above: developmentally open and with permeable boundaries. cases obtain their structure and processes in interaction with their contexts in very much the same way said structures emerge through the dissipative structure of complex social systems (gerrits & verweij, 2013; gerrits & pagliarin, 2020). the key point here is to accept a critical realist or complex realist (harley, 2009; gerrits & verweij, 2013) ontology that foregrounds the difference between the case as object (intransitive domain, in which reside the cases’ mechanisms, events, and experiences) as such and our knowledge of the case as obtained through the observations (made) available to us (transitive domain, in which reside our experiences of the case). the stratified take on reality as expressed in critical realism complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 39 university of bamberg press explains the experience of many researchers through which they learn that there the many ways in which one can approximate the same object. the boundaries of any complex system are as much a property of the case as it is of the act of describing and defining it by researchers. as much as researchers may look for commonalities between cases, as if they are disembodied and decontextualised units, it is their diversity that forms the biggest empirical puzzle. naturally, similar and stable cases in similar and stable contexts would come a long way towards generating predictions. however, the social scientist has to contend with the fact that case diversity as it develops over time leads to equifinality (goertz & mahoney 2009) and multifinality (george & bennett, 2005). enter the probabilistic and subjective nature of time, as discussed in the previous sections, and it becomes clear why cases are such ambiguous entities. indeed, “… cases appear to be much more complicated than suggested by the rather clear and simple expression ‘case study’” (dumez, 2015, p.44; see also carter & sealey, 2009). “temporal casing” is the process of capturing the case as an object that evolves as precisely as possible through description and/or modelling. the case representing a dissipative structure will likely change in its structure and processes, i.e., the within-case variation discussed above. this calls for dynamic descriptions and models. inevitably, details of the object will be overlooked, ignored, and lost, as no representation can be fully accurate. contrary to social constructivists, however, there is a link between knowledge of the case and the actual object. this calls for a constant reflection that “[…] entails a methodological justification that openly appeals to social scientific criteria of consistency and appropriateness of the researcher’s social categories to the task of explanation.” (carter & sealey, 2009, p. 77). temporal casing as an act on behalf of the researcher can’t be separated from the case as object; interpretation or reading of the data creates additional data that then becomes part of the case study. this point of departure has the following epistemological implications for casing, which also reflect in the methodology of temporal casing: 1. context and history over similarity. cases should be understood as having a history, as following a trajectory through an n-dimensional property space (cf. attractor; see byrne 2004, 2005; gerrits & pagliarin, 2020). while cases may certainly describe similar trajectories, similarities in case properties should not be the primary target of the analysis. convergence and divergence of trajectories is inherent to the way complex systems evolve. 2. qualitative first, quantitative later. as chronos has limited meaning vis-à-vis the real world, it is through case-based qualitative data that one gets a first approximation of how time plays out as within-case variation. interpretation and identification of meaningful time points characterizing a certain set of processes as experienced from within by analysing qualitative data. 3. endogenous time over exogenous time. as each case follows its own contextual logic over time and across certain places, the research should follow the logic internal to the case instead of imposing an external regime such as clock time. importantly, one should acknowledge that various aspects of cases can have their own spatio-temporal duration and frequency. 4. uniqueness and generality. cases are neither completely similar nor completely unique. exactly what is idiosyncratic in cases and what is recurring can only be found through comparison. 5. stages instead of intervals. as internal logic in qualitative terms describing context and history trumps external (quantified) time regimes, the focus should be on the developmental stages of varying length instead of on intervals. 6. boundaries. cases in categories such as organisations, cities, and neighbourhoods have contestable, permeable boundaries as they are approximations of the case as object. this implies that the process of setting boundaries must be a conscientious one (cilliers 2001; jessop et al., 2008). we will now turn to the burgeoning body of knowledge of the governance of megaprojects to illustrate how alternative takes on time alter our understanding of such projects. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 40 university of bamberg press 3.2 a practical illustration of temporal casing in empirical research megaprojects concern construction projects2 that are vast in scope, time, and budget. exactly what can be considered a megaproject or what cannot, is not a given. as a rule of thumb, one could consider anything above a 1 billion euros or dollars a megaproject. importantly, these projects are characterised by considerable complexity in various dimensions: technical (e.g., the use of novel construction techniques), societal (e.g., in the face of diverging societal voices, including protests), political (e.g., facing different coalitions in favour or against the project), physical locality (e.g., covering and linking various areas), scope (e.g., bundling various different subprojects), organisational (e.g., developed and built by a hybrid of public, private, and semi-public authorities), financial (e.g., financed through public-private partnerships), and more. megaprojects tend to be perpetually out of control, resulting in considerable budget overruns, delays, and faltering quality (flyvbjerg et al., 2003). the evaluation of such projects has revealed recurring causes, including overly-ambitious politicians, inaccurate future projections, incentives to maximise budgets, and poor project control (flyvbjerg, skamris holm, and buhl 2005; flyvbjerg, 2014). the findings in these studies derive primarily from statistical analysis, including cross-sectional data series, that subject the cases to uniformisation and regimentation across time and space. we don’t contest the soundness of those findings, but we also observe that individual case studies highlight considerable diversity that is not accounted for in said comparative research. for instance, while the eurotunnel and the øresund connection appear similar in many aspects, a complexity-informed, timesensitive recasting in terms of temporal casing shows that the cases are more dissimilar than similar (gerrits and verweij 2018). this goes for many, if not most, other such projects. as spatial units, they range from being limited by clear boundaries (e.g., elbphilharmonie, germany) to sprawling programmes that cover large and diverse areas (e.g., stuttgart21, germany). some of them can be considered as taking place in a green-fields (e.g., hong kong–zhuhai–macau bridge, china), while others are fundamental conversions of existing built environments (e.g., potzdammerplatz, germany). there is a strong link between space and time. a sprawling redevelopment of a series of connected sites generates a different endogenous and dynamic temporality than a new development at a limited site. other differences between megaprojects also appear when one considers the time dimension. obviously, megaprojects are delivered on different timescales: in some rare cases, megaprojects are completed as scheduled, while others are delivered decades later than originally planned (e.g., berlin brandenburg airport). following our argument, one should be less concerned about how those projects are tied to fixed timescales and more about how their endogenous time plays out. some of them are dense with activities and are experienced as taking place at break-neck speed despite their longevity in calendar time (e.g., noord-zuidlijn, the netherlands), while other projects that required an equal amount of calendar time feature stages during which very few activities take place (e.g., highspeed line berlin–erfurt– bamberg, germany). in a similar fashion, one can observe how the density of activities within cases varies wildly, with some showing very irregular patterns (e.g., hsl zuid high-speed railway line, the netherlands) and others having a more regular distribution. megaprojects hence, appear to offer a particularly appropriate illustration to show how temporal casing can concretely be performed in empirical research. we will focus on two european megaprojects: the elbphilarmonie in hamburg (fiedler and schuster 2016) and the berlin brandenburg airport, berlin (fiedler & wendler, 2016), both located in germany. in what follows, we will make extensive reference to the work by fiedler and schuster (2016) and fiedler and alexander (2016). however, we will not summarise the content of their contributions. instead, we will highlight where their analysis overlaps and differs in the approach of temporal casing compared to what we propose here. our re-elaboration of both 2 arguably, technological projects such as the apollo space program are also megaprojects. for the present purpose, we focus on projects that have a spatial component. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 41 university of bamberg press megaprojects has two main goals: first, we will show how temporal casing is done in practice, by signalling (in italic) to readers the dimensions of temporal casing identified above in section 3.1, and second, we hope to clarify to readers how temporal casing differs from other conceptualisations of time as previously discussed in the paper. boundaries: the elbphilarmonie and the berlin brandenburg airport (ber) are particularly appropriate for the purposes of our illustration because they have clear conceptual, temporal and spatial boundaries. conceptually, the elbphilarmonie and the ber are well-known european megaprojects that have been extensively covered in the literature on megaprojects and in the media. in terms of temporal boundaries, they both emerged after the german reunification in the early 1990s, and the planning and implementation of the ber covered over 25 years (mid-1990s to the opening of the airport in 2021), while the planning and implementation of the elbphilarmonie spanned over approximately 17 years (2000-2017). in terms of spatial boundaries, because of the type of urban function they provide (cultural and tourist activities vs. transport and mobility), both megaprojects have clear spatial boundaries, as the ber is an airport and the elbphilarmonie is a building. however, their conceptual, temporal, and spatial boundaries could also be different: the ber could have larger spatial boundaries if the (city-wide or region-wide) transportation network and (global) mobility flows related to air traffic would be considered. the spatial boundaries of elbphilarmonie could be enlarged by considering the entire area in which it is located (hafencity). the elbphilarmonie, hence would just be an element of a broader transformation occurring in hamburg through the urban renewal of hafencity; this demonstrates a widening of the conceptual and temporal boundaries of this megaproject. this means that, in temporal casing, the entire duration of a process is defined by the researcher (i.e., it is not a given). the definition of the whole duration of a process can be a starting point, but it can also be a later step of the analysis when the cases’ development has been already reconstructed. this will become important for our discussion below on the “stages vs. intervals” as the key characterising feature of temporal casing. uniqueness and generality: both megaprojects are similar and different to a certain degree. conceptually, they are both megaprojects and can be analysed through the same analytical lenses, but their duration is different. they are also similarly located in germany, and the formulation of both projects have emerged after the reunification of germany. however, they are in different states (furthermore, hamburg being a city state) and their spatial extents (both for the projects and of the cities in which they are located) are very different, as well as their demographic profile. as outlined in the first paragraphs of this subsection, the list of similarities and differences could continue; differentiating between relevant and irrelevant similarities and differences is a key challenge in case research. although we cannot address this issue in detail in the present paper (for a reference, see e.g. (berg-schlosser & de meur, 2009; ragin, 2000), we nevertheless want to stress that it is the researcher who has to argue for, and justify the ground for comparability across different cases. the grounds for justification relate back to the dimension of the conceptual, temporal, and spatial boundaries of cases discussed above. noteworthy is that temporal casing is not solely tailored for comparative research, in this illustration we also show how temporal casing is particularly useful to compare the development of different cases characterised by different durations. context and history over similarity: to make the comparison across different cases possible, it is first necessary to examine the specific histories of individual cases (byrne, 2005). this means that, even when the focus is on comparison, the in-depth analysis of individual cases must be rendered by as accurate of a “story” about the case as possible. therefore, researchers have to develop an “intimacy” with their cases (ragin, 2000) or, stated differently, they need to engage with cases’ uniqueness before being able to provide a more synthetic representation of the development of each individual case through temporal casing. in both analyses of the elbohilarmonie and the ber, the authors adopted a historical approach and reconstructed in detail the development of each project—from site identification, to plan formulation, to implementation phases. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 42 university of bamberg press endogenous time over exogenous time: the historical reconstruction of the elbphilarmonie and the ber is a description that allows us to identify the “temporal logic” characterising each case. this means that the case-specific development of each project is shown at the pace it unfolded. endogenous time is typically examined through the identification of periods when the process accelerated or slowed down, or when a certain general event (e.g., financial crisis, fall of the berlin wall) had specific consequences in one case but not in another because of the former’s contextual characteristics. qualitative first, quantitative later: to perform this historical reconstruction on each of the megaprojects, different data sources have been used, from academic and non-academic outlets (e.g., newspaper articles, reports), to chronologically reconstruct the process through which each of the megaprojects emerged and developed. by considering the above aspects together, it becomes clear that fiedler and schuster (2016) and fiedler and wendler (2016) reconstructed the complexity of both megaprojects by examining, over time, the intersection between actor governance and technical and technological aspects of megaprojects (coevolution). for instance, the authors examine the role of the multiple actors involved (private and public actors, independent agencies, their “entrance” and “exit” during the “dramaturgical” development of both megaprojects) and the technical difficulties encountered in the planning and implementation of both megaprojects (e.g. the replacement of the fire alarm system in the case of ber, the problems in producing the individually unique glass windows of the elbphilarmonie with novel production technologies). the authors also identify the main pitfalls of megaproject management (also with the advantage of hindsight) in terms of overoptimism, approximate planning, limited risk management, and vague cost estimation. to this historical and detailed, but necessarily simplified reconstructions of the complexity of actor governance, of technological development, and of (mega)project management, quantitative data can be integrated, for instance, by calculating the number of days, months, or years eventually needed to complete a specific task as compared to the estimated time or the estimated costs as compared to the actual cost of the entire megaproject. however, the historical reconstruction by fiedler and schuster (2016) on the elbphilarmonie and by fiedler and wendler (2016) on ber by using qualitative data is not only descriptive, but also analytical. among the multiple actors involved, the key representatives are identified; among the many technical difficulties and production delays, focus is directed at the selection of main issues; among the complexity of megaproject management, certain key mistakes are subjects of study. furthermore, the authors identify meaningful turning points along the process, for instance when governance arrangements or political scenarios changed, when the megaproject manager was replaced or when one of the main companies went bankrupt. this means that the authors have substantially interpreted the development of their cases. this combination and balancing between empirical data and the authors analytical lenses about the complexity of governance, technology, and management of megaprojects is a characterising feature of qualitative research. researchers perform in-depth analyses to provide thick descriptions of their cases, which are then interpreted and analysed based on the theories that grounded their examination in the first place for the goals of theory-testing or theory-building (broadly speaking). stages instead of intervals: so far, the conceptual and methodological approach adopted by fiedler and schuster (2016) and fiedler and wendler (2016) mostly overlaps with our own approach to temporal casing. what is different, potentially an added value or particularity of temporal casing (both conceptually and methodologically) is first, the connection to complexity-informed perspectives in the social sciences (see abell, 2009; uprichard & byrne 2006), and second, the arrangement of the analysed empirical material into development stages. first, narratives, such as collected qualitative data from interviews or chronologies of a certain process, can be systematised by identifying key actors, turning points, and ordered sequences of events that, together, describe the unfolding of a process as a story. narratives are simplifications and abstractions, but they dialogue with complexity (or, at least, attempt to do it) because they are grounded in complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 43 university of bamberg press the indirect accounts (e.g., interviews to key informants, truly lived episodes of a specific event, newspaper articles, reports, academic articles) that actors provide about a certain process. this means that by using narratives, complexity is approached through the agency of the actors involved in the process (uprichard & byrne, 2006). narratives can represent the complexity of cases by showing the different meanings that actors involved in a process assigned to certain situations or events. actors’ interpretations, too, are part of the explanation linked to the occurrence of an event. narratives, which are composed of both qualitative and quantitative information, can be systematised, for instance, into diagrams (qualitative modelling) and offer a route towards explanatory statements, such as limited generalisations of results, when narratives are compared across cases (uprichard & byrne, 2006). it is probably abell (2009) that, through his highly formalised conceptual framework on narratives, more clearly showed how explanatory statements can be derived from the analysis of action narrative paths. a detailed discussion of abell’s work is beyond the scope of our present article, but here it suffices to highlight, through abell’s framework, how researchers analyse and interpret narratives as an ordered sequence of events that can be formalised into diagrams composed of nodes and arcs. actors’ decisions and events, their duration and effects can be reconstructed through formalised diagrams that may also show “causal” links within a single case. this means that a narrative is not simply a story, but a “causal story” (p. 44) that provides an action path along which actors, their actions, the conditions for their acts, and the effects of those actions are connected and ordered by the researcher to derive causal statements. when narratives are compared, a common causal story might be identified. although fiedler and schuster (2016) and fiedler and wendler (2016) did not make a specific link with this complexity-informed approach to qualitative research, in this illustration we will consider their analyses as complexity-informed causal stories from which key actors, turning points and main effects can be identified (interpreted) to describe the planning and implementation process of the elbphilarmonie and the ber. from this understanding of (mostly) qualitative (but also, quantitative) research, temporal casing comes fully into play when researchers subdivide these stories into stages. this means that key actors and turning points are used to identify where stages in a process “start” or “end” and that they, altogether, form an ordered sequence of events, or trajectory. below, we show how to practically perform temporal casing by using the ber and the elbphilarmonie as examples. given the purposes of our illustration, we reported only the main events and identified the main stages during the planning and implementation of both megaprojects. a much more detailed and complete account could hence be possible. we present both examples of temporal casing in tabular form (see table 2) to show how stages are case-specific and how they can be used to describe at best what happened along the process. however, other visualisations are possible (see the qualitative modelling approach in uprichard and byrne, 2006, or the diagrams for action narrative paths as in abell, 2009). we would also like to remind readers that, as stated above, stages are simplifications, and they are also an analytical lens to segment a complex process in manageable parts that compose the causal story of each case. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 44 university of bamberg press start/end berlin brandenburg airport stages start/end elbphilarmonie stages 1996 consensus among relevant actors about the need for a new airport for berlin 1 1996-1999 process for privatisation failed 1 2000-2003 public ownership of ber through fbb (flughafen berlin schönefeld gmbh), appointment of general manager, but no general contractor and 100% risk of loans guaranteed by the public sector. key management mistakes (optimistic time schedule, fragmented bidding process) 2 2000-2003 consensus among relevant actors about the elbphilarmonie as the new landmark of the city of hamburg 1 2008-2011 construction started, in parallel with the airport design 3 2004-2005 set-up of governance arrangements (“slim” public agency rege— supporting the bidding process and appointment of one big contractor— hochtief ) 2 2012-2013 new general manager appointed; re-building of fire safety alarm system given the new (but expected) regulations 4 2006 2013 contract change to a forfeit model (higher costs for the public actor) and escalation of costs due to a “planning-by-doing” construction. tension among actors and change in staff. 3 … … … … … … 2017 opening 2021 opening table 2. an illustration of temporal casing for the berlin brandenburg airport (ber) and elbphilarmonie, based on the analyses by fiedler and wendler (2016) and fiedler and schuster (2016), respectively. the book chapters have been published in 2016 so the analyses are not up-to-date, and we signalled missing information by using “….”. to conclude this section, we would like to discuss three main characteristics of the temporal casing as illustrated in table 2. first, stages are topic-specific; in table 2 we identified the main stages based on the megaprojects and spatial planning literature: site identification and plan formulation (stage 1), planning phase and governance set-up (stage 2), and implementation process (stage 3 and following). in other domains, stages have to resonate with the specific theories and knowledge developed in their complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 45 university of bamberg press respective research area (e.g., stages in the policy cycle). they should, however, not be followed slavishly but, as we discussed above, meaningfully capture the uniqueness of the cases’ histories. second, the duration of stages is not defined by a fixed temporal unit (like one year, or 5 years), but it varies according to the focus of analysis. moreover, as shown in table 2, not only have stages a different duration within the same case (e.g., in the elbphilarmonie, stage 1 has not the same duration as stage 2), but also across the cases (e.g., stage 2 in the elbphilarmonie has not the same duration as stage 2 in ber). furthermore, for stage 1 in ber, although we subdivided the processes between 1996 and 1999 into two items, we considered it as part of the same stage. therefore, stages can also include multiple events, which might then be qualitatively grouped into one stage. it is always the job of the researcher to analyse the case-specific information and structure it into stages that can effectively (although never “perfectly”) represent the case development. the latter is very important because stages will never satisfy the level of detail desired by researchers using either quantitative and qualitative approaches towards generalisation or case specifics. the different duration of stages and the flexibility needed to identify them explains why we used the word “stages” and not “intervals,” as the latter can suggest fixed time periods. on the other hand, although the word “stage” can imply a step along a pre-defined development path, the term stage is used to indicate a step in a periodisation (or “concatenation”; literally, “chain of events”). temporal casing is not only a description of what happened but, more importantly, it is an analytical effort to systematise a process into an ordered sequence of events. in temporal casing, the building blocks of this ordered sequence of events are (development) stages. each of the stages has individual, case-specific characteristics in terms of content (what happened, where, and who was involved) and duration (how long) which, together, present, in abbott’s terms, a causal story (why). we restate that how exactly this is done depends on the focus of analysis (see above) and on the researcher(s)’s analytical and methodological strategy. there are no best ways to do this, but temporal casing is no mystery in this regard, as it is grounded on conventional qualitative research in the social sciences, so readers can refer to further resources on this domain. third, as it is the researcher who defines the stage-specific durations (see above) and the entire duration of a process (see above on boundaries), temporal casing allows for entire durations that are not only different, but that are also punctuated by a different number of stages with a different duration. hence, temporal casing allows for variety (i.e., it is a diversity-oriented approach, see ragin, 2000): regardless of their entire duration, some processes can be described and analysed by means of two stages, some others by four stages or more (and each of those stages will most likely have a different duration). furthermore, this means that temporal casing can also potentially (and partially) capture the rate of change of a certain process. using a rough measure, which should be improved in the future, the casebased rate of change can be expressed as the ratio between the number of stages qualitatively identified and needed to describe a certain process and its total years of duration. for the stages we identified in table 2, the rate of change would be 0.23 for both the ber (4 stages divided by 17 years) and the elbphilarmonie (3 stages divided by 13 years). this suggests that, despite their different duration and number of stages, and the different durations of their stages, the two megaprojects are characterised by the same rate of change. 4 conclusions in this paper, we started from the observation that scientists continue to wrestle with notions of space and time. the complexity sciences focus on the topic, but with considerable limitations as empirical phenomena are subjected to the regimentation of calendar or clock time. following an overview of alternative ways of understanding time (see table 1), we outlined the properties of time-sensitive casing and revisited the research on megaprojects to highlight what would change had those projects been re-cast in a more time-sensitive manner. we argue that such a re-casting, which we label as temporal casing, complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 46 university of bamberg press constitutes a difference in-kind instead of a difference in degree in research. time and space are primarily endogenous to cases, so case-based research should follow that logic instead of subjecting the case to external logics conceptualising time as regular ticks. however, we don’t want to repeat the default argument that single, in-depth case studies would be better. such studies can be helpful, too, just as a cross-sectional analysis can be. nor is our intention to declare any of these approaches, or others, as obsolete. rather, we observe a faux dichotomy between qualitative and quantitative approaches that misses the main point, namely that cases follow diverging and converging trajectories through the property space. these movements can only be found through structural comparison of the in-depth case materials that highlight the equifinal and multifinal nature of the cases (gerrits and pagliarin 2020). in our empirical illustration, both the elbphilarmonie and the ber feature many of the hallmarks as shown in table 1. they are structured by time-space, contain multiple lineages of events and activities of a varying density, and have a different pacing from the perspective of the people involved. at the core of temporal casing lies the acknowledgement that time units and space units are not a natural given but should be redefined based on within-case variation as described here. in other words, a straightforward comparison between cases across calendar time is therefore not as informative as it seems at first sight but requires interpretation. as we showed in section 3.2, most research practices in temporal casing can appear common sense because they are very much rooted in qualitative empirical research. however, we find these practices are rarely stated explicitly or highlighted by researchers. hence, we identified and showed how the six main dimensions of temporal casing (see section 3.1: context and history over similarity; boundaries; endogenous time over exogenous time; uniqueness and generality; qualitative first, quantitative later; stages instead of intervals) work in practice in the two selected cases. the result of temporal casing as timeand case-based research is rendered, in this paper, in tabular form as per table 2. however, table 2 is not simply a visual support to show how temporal casing works in practice. it is first and foremost an expression of the central effort towards formalisation that characterises temporal casing. the identification of ordered and numbered, development stages for each case not only means that stages can be captured, but also that they can be compared between different cases. temporal casing aims at capturing within-case time variation, but also has the further goal to compare time variation, in the form or ordered stages, across cases. such comparison can occur, for instance, by examining the rate of change of each case and by employing systematic comparative research methods such as qualitative comparative analysis (qca) (ragin, 2000, 2008). temporal casing as a research approach and practice is therefore strongly rooted in configurational comparative methods. in conclusion, we believe that capturing the within-case time variation through temporal casing would beneficially contribute to complexity-informed empirical research in the social sciences, as the power of methods is rendered void if the casing ignores the temporal specificities of social processes. in showing the diversity of ways to conceptualise time, we also highlighted their limits. we further demonstrated how, being rooted in qualitative, case-based research, temporal casing can partially address those limitations. having also revealed the advantages of temporal casing, we hope that researchers will use it in the near future as a basis or groundwork for their own case-based and/or comparative research that aims to be sensitive to “the time of the case.” 5 references abbott, a., 1995. boundaries of social work or social work of boundaries? the social service review lecture. social service review 69, 545–562. https://doi.org/10.1086/604148 abbott, a., 1983. sequences of social events: concepts and methods for the analysis of order in social processes. historical methods: a journal of quantitative and interdisciplinary history 14, 129-147. https://doi.org/10.1080/01615440.1983.10594107 complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 47 university of bamberg press abell, p., 2009. a case for cases comparative narratives in sociological explanation. sociol. methods. res. 38, 38–70. https://doi.org/10.1177/0049124109339372 abell, p., 2004. narrative explanation: an alternative to variable-centered explanation? annual review of sociology 30, 287-310. http://www.annualreviews.org/doi/10.1146/annurev.soc.29.010202.100113 adam, b., 2006. time. theory, culture & society 23, 119–126. https://doi.org/10.1177/0263276406063779 adam, b., 2003. reflexive modernization temporalized. theory, culture & society 20, 59–78. https://doi.org/10.1177/0263276403020002004 adam, b., 1995. timewatch: the social analysis of time. polity press, cambridge. adam, b., 1992. modern times: the technology connection and its implications for social theory. time & society 1, 175–191. https://doi.org/10.1177/0961463x92001002003 aeon, b., aguinis, h., 2017. it’s about time: new perspectives and insights on time management. academy of management perspectives 31, 309–330. https://doi.org/10.5465/amp.2016.0166 altomonte, g., 2016. beyond being on call: time, contingency, and unpredictability among family caregivers for the elderly. sociological forum 31, 642–662. https://doi.org/10.1111/socf.12267 babbie, e.r., 2013. the practice of social research, 13. edition, international edition. ed. wadsworth cengage learning, belmont, ca. barabara, a., 2014. forms of space and time, in: yelavich, s., caccavale, e. (eds.), design as future-making. bloomsbury academic, london and new york, pp. 225–232. beach, d., pedersen, r.b., 2013. process-tracing methods: foundations and guidelines. university of michigan press, ann arbor. bengtsson, b., ruonavaara, h., 2011. comparative process tracing in housing studies. international journal of housing policy 11, 395–414. https://doi.org/10.1080/14616718.2011.626603 bergmann, w., 1992. the problem of time in sociology: an overview of the literature on the state of theory and research on the sociology of time, 1900-82. time and society 1, 81–134. https://doi.org/10.1177/0961463x92001001007 berg-schlosser, d., de meur, g., 2009. comparative research design: case and variable selection, in: rihoux, b., ragin, c.c. (eds.), configurational comparative methods, qualitative comparative analysis (qca) and related techniques. applied social research methods series. sage publications, inc, thousand oaks, ca, pp. 19–33. bourdieu, p. (1963). la société traditionnelle : attitude à l'égard du temps et conduite économique. sociologie du travail, 1, 2444. brenner, n., schmid, c., 2014. the ‘urban age’ in question. international journal of urban and regional research 38, 731–755. https://doi.org/10.1111/1468-2427.12115 byrne, d., 2005. complexity, configurations and cases. theory, culture & society 22, 95–111. https://doi.org/10.1177/0263276405057194 byrne, d.s., 2004. complexity theory and social research. social research update. byrne, d.s., 1998. complexity theory and the social sciences: an introduction. routledge, london. carter, b., sealey, a., 2009. reflexivity, realism and the process of casing, in: the sage handbook of case-based methods. sage publications ltd, london, united kingdom, pp. 69–83. https://doi.org/10.4135/9781446249413.n4 castellani, b., hafferty, f.w., 2009. sociology and complexity science, understanding complex systems. springer berlin heidelberg, berlin, heidelberg. castellani, b., rajaram, r., gunn, j., griffiths, f., 2016. cases, clusters, densities: modeling the nonlinear dynamics of complex health trajectories. complexity 21, 160–180. https://doi.org/10.1002/cplx.21728 cheng, t.-y., 2017. the ecology of social time: an outline of an empirical analytic framework of the sociology of time. time and society 26, 137–164. https://doi.org/10.1177/0961463x15577266 cilliers, p., 2001. boundaries, hierarchies and networks in complex systems. int. j. innov. mgt. 05, 135–147. https://doi.org/10.1142/s1363919601000312 cullen, i., godson, v., 1975. urban networks: the structure of activity patterns. progress in planning 4, 1–96. https://doi.org/10.1016/0305-9006(75)90006-9 delclòs-alió, x., marquet, o., miralles-guasch, c., 2017. keeping track of time: a smartphone-based analysis of travel time perception in a suburban environment. travel behaviour and society 9, 1–9. https://doi.org/10.1016/j.tbs.2017.07.001 dumez, h., 2015. what is a case, and what is a case study? bulletin of sociological methodology/bulletin de méthodologie sociologique 127, 43–57. https://doi.org/10.1177/0759106315582200 durkheim, é., 1915. les formes élémentaires de la vie religieuse [the elementary form of the religious life]: le système totémique en australie: translated by josepph ward swain. george allen & unwin ltd., london. elchardus, m., 1988. the rediscovery of chronos: the new role of time in sociological theory. international sociology 3, 35– 59. eliade, m., 1965. le sacré et le profane. gallimard, paris. ellegård, k. (ed.), 2018. time geography in the global context: an anthology, routledge studies in human geography. routledge, abingdon oxford and new york ny. elster, j., 1976. a note on hysteresis in the social sciences. synthese 33, 371–391. eurostat, 2020. statistical regions in the european union and partner countries — nuts and statistical regions 2021 (no. ksgq-20-092). european union, luxembourg: publications office of the european union. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 48 university of bamberg press fiedler, j., schuster, s., 2016. the elbphilharmonie hamburg, in: kostka, g., fiedler, j. (eds.), large infrastructure projects in germany: between ambition and realities. springer international publishing, cham, pp. 39–85. https://doi.org/10.1007/978-3-319-29233-5_3 fiedler, j., wendler, a., 2016. berlin brandenburg airport, in: kostka, g., fiedler, j. (eds.), large infrastructure projects in germany – between ambition and realities. palgrave macmillan, berlin, pp. 93–103. flyvbjerg, b., 2014. what you should know about megaprojects and why an overview. project management journal 45, 6–19. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2424835 flyvbjerg, b., bruzelius, n., rothengatter, w., 2003. megaprojects and risk: an anatomy of ambition. cambridge university press, cambridge. flyvbjerg, b., skamris holm, m.k., buhl, s.l., 2005. how (in) accurate are demand forecasts in public work projects? the case of transportation. journal of the american planning association 71, 131–146. george, a.l., bennett, a., 2005. case studies and theory development in the social sciences. the mit press. gerring, j., 2004. what is a case study and what is it good for? the american political science review 98, 341–354. gerrits, l., pagliarin, s., 2020. social and causal complexity in qualitative comparative analysis (qca): strategies to account for emergence. international journal of social research methodology 24, 501–514. https://doi.org/10.1080/13645579.2020.1799636 gerrits, l., verweij, s., 2018. the evaluation of complex infrastructure projects. a guide to qualitative comparative analysis. edward elgar publishing, cheltenham uk, northampton usa. gerrits, l., verweij, s., 2013. critical realism as a meta-framework for understanding the relationships between complexity and qualitative comparative analysis. journal of critical realism 12, 166–182. goertz, g., mahoney, j., 2009. scope in case study research, in: the sage handbook of case-based methods. sage publications ltd, 1 oliver’s yard, 55 city road, london ec1y 1sp united kingdom, pp. 307–317. https://doi.org/10.4135/9781446249413.n18 goffman, e., 1961. internados. ensayos sobre la situación social de los enfermos mentales, 2nd ed. amorrortu editores, buenos aires. gurvitch, g., 1964. the spectrum of social time. dordrecht. gurvitch, g., 1973. social structure and the multiplicity of time, in: tiryakian, e. a., (eds), sociological theory, london, pp.17184. hägerstrand, t., 1970. what about people in regional science? papers in regional science 24, 7–24. https://doi.org/10.1111/j.1435-5597.1970.tb01464.x hägerstrand, t., 1967. innovation diffusion as a spatial process. london: university of chicago press. hall, p., hay, d., 1980. growth centres in the european urban system. heinemann, london. harvey, d.l., 2009. complexity and case, in: byrne, d., ragin, c.c. (eds.), the sage handbook of case-based methods. sage publications, inc, london, pp. 15–38. heise, d. r., 1993. narratives without meaning? the journal of mathematical sociology 18, 183 -189. https://doi.org/10.1080/0022250x.1993.9990123 heylighen, f., cilliers, p., gershenson, c., 2006. complexity and philosophy. arxiv:cs/0604072. isaac, a.g., 1994. hysteresis, in: the elgar companion to radical political economy. edward elgar, london. jessop, r., brenner, n., jones, m., 2008. theorizing sociospatial relations. environment and planning. d, society and space 26, 389–401. https://doi.org/10.1068/d9107 kitchin, r., dodge, m., 2007. rethinking maps. progress in human geography 31, 331–344. https://doi.org/10.1177/0309132507077082 kolarova, v., cyganski, r., lenz, b., 2019. chapter seven activities while travelling? travel time perception and travel time use in an era of automated driving, in: ben-elia, e. (ed.), advances in transport policy and planning, the evolving impacts of ict on activities and travel behavior. academic press, pp. 171–206. https://doi.org/10.1016/bs.atpp.2019.07.002 latour, b., 1987. science in action: how to follow scientists and engineers through society. harvard university press, cambridge, massachusetts. le gléau, j.-p., pumain, d., saint-julien, t., 1997. towns of europe: to each country its definition. estat 6, 1–15. https://doi.org/10.3406/estat.1996.6079 lentz, m.j., 1990. time series. issues in sampling. western journal of nursing research 12, 123–128. https://doi.org/10.1177/019394599001200111 levine, r. (ed.), 2006. a geography of time: the temporal misadventures of a social psychologist, or how every culture keeps time just a little bit differently. oneworld, oxford. lefebvre, h., & régulier, c., 1985. le projet rythmanalytique. communications, 41(1), 191-199. doi:10.3406/comm.1985.1616 luhmann, n., 1982. the differentiation of society. columbia university press. lund, c., 2014. of what is this a case? analytical movements in qualitative social science research. human organization 73, 224–234. https://doi.org/10.17730/humo.73.3.e35q482014x033l4 madanipour, a., 2017. cities in time: temporary urbanism and the future of the city. bloomsbury academic, london. may, j., & thrift, n., 2003. timespace: geographies of temporality. london: routledge. mead, g. h., 1932. the philosophy of the present. the open court company, london. miller, h.j., 2005. a measurement theory for time geography. geographical analysis 37, 17–45. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 49 university of bamberg press millonig, a., sleszynski, m., ulm, m., 2012. sitting, waiting, wishing: waiting time perception in public transport, in: 2012 15th international ieee conference on intelligent transportation systems. presented at the 2012 15th international ieee conference on intelligent transportation systems, pp. 1852–1857. https://doi.org/10.1109/itsc.2012.6338777 minca, c., 2009. the island: work, tourism and the biopolitical. tourist studies 9, 88–108. https://doi.org/10.1177/1468797609360599 newton, t., 2003. crossing the great divide: time, nature and the social. sociology 37, 433–457. https://doi.org/10.1177/00380385030373003 nowotny, h., 1992. time and social theory: towards a social theory of time. time & society 1, 421–454. https://doi.org/10.1177/0961463x92001003006 pagliarin, s., gerrits, l., 2020. trajectory-based qualitative comparative analysis: accounting for case-based time dynamics. methodological innovations. https://doi.org/10.1177/2059799120959170 pahl, s., sheppard, s., boomsma, c., & groves, c., 2014. perceptions of time in relation to climate change. wires clim change, 5(3), 375-388. doi:10.1002/wcc.272 parkes, d., thrift, n., 1979. time spacemakers and entrainment. transactions of the institute of british geographers 4, 353–372. prigogine, i., 1997. the end of certainty, 1 edition. ed. free press, new york. prigogine, i., stengers, i., 1984. order out of chaos: man’s new dialogue with nature. bantam books. provonost, g., 1989. current sociology. the journal of the international sociological association 37, 1–129. https://doi.org/10.1177/001139289037003001 ragin, c.c., 2008. redesigning social inquiry. fuzzy sets and beyond. university of chicago press, chicago and london. ragin, c.c., 2000. fuzzy-set social science, 1 edition. ed. university of chicago press, chicago. ragin, c.c., becker, h.s. (eds.), 1992. what is a case? exploring the foundations of social inquiry. cambridge university press, cambridge, usa. reed, m., harvey, d.l., 1992. the new science and the old: complexity and realism in the social sciences. journal for the theory of social behaviour 22, 353–380. https://doi.org/10.1111/j.1468-5914.1992.tb00224.x reszohazy, r., 1970. temps social et développement. bruxelles. roca cladera, j., 2003. la delimitación de la ciudad?: una cuestión imposible? ciudad y territorio. estudios territoriales 135, 17– 36. https://recyt.fecyt.es/index.php/cytet/article/view/75308 rosen, w., 2012. the most powerful idea in the world: a story of steam, industry, and invention. university of chicago press. shen, y., chai, y., kwan, m.-p., 2015. space–time fixity and flexibility of daily activities and the built environment: a case study of different types of communities in beijing suburbs. journal of transport geography 47, 90–99. https://doi.org/10.1016/j.jtrangeo.2015.06.014 shumway, r.h., stoffer, d.s., 2017. characteristics of time series, in: shumway, r.h., stoffer, d.s. (eds.), time series analysis and its applications: with r examples. springer texts in statistics. springer international publishing, cham, pp. 1–44. https://doi.org/10.1007/978-3-319-52452-8_1 simmel, g., 1903. le metropoli e la vita dello spirito, 1995th ed. armando editore, roma. sorokin, p. a., & merton, r. k. (1937). social time: a methodological and functional analysis. american journal of sociology, 42, 615-639. tannier, c., thomas, i., 2013. defining and characterizing urban boundaries: a fractal analysis of theoretical cities and belgian cities. computers, environment and urban systems 41, 234–248. http://dx.doi.org/10.1016/j.compenvurbsys.2013.07.003 thrift, n., 2006. space, place, and time. in: r. e. goodin, & c. tilly; (ed.), the oxford handbook of contextual political analysis, pp. 547-563. oxford: oxford university press. thrift, n., 2005. torsten hägerstrand and social theory. progress in human geography, 29(3), 337-340. doi:https://doi.org/10.1177/030913250502900312 uprichard, e., byrne, d., 2006. representing complex places: a narrative approach. environ plan a 38, 665–676. https://doi.org/10.1068/a37333 van tienoven, t., 2019. a multitude of natural, social and individual time. time & society 28, 971–994. https://doi.org/10.1177/0961463x17752554 wexler, m.n., 2015. time as a commodity waits for no-one: queue culture slows time, technology speeds things up. strategic direction 31, 15–17. https://doi.org/10.1108/sd-07-2015-0109 whillans, a., 2020. time smart: how to reclaim your time and live a happier life. boston: harvard business review press. yin, r.k., 1994. case study research: design and methods. sage, newbury park, ca. zerubavel, e., 1985. hidden rhythms: schedules and calendars in social life, first california paperback ed. ed. univ. of california press, berkeley. zerubavel, e., 1982. the standardization of time: a sociohistorical perspective. american journal of sociology 88, 1–23. complexity, governance & networks – vol. 7, no 1 (2021) special issue: complexity and time in governance, p. 29-49 doi: http://dx.doi.org/10.20377/cgn115 complexity, governance & networks – vol. 4, no 1(2018) special issue: teaching complexity, p. 10-18 doi: http://dx.doi.org/10.20377/cgn-62 10 university of bamberg press teaching complexity in public administration across the globe: an overview nevena ivanovica, lasse gerritsb abstract: as the contributions in this special issue show, there are ample examples of teaching programs at the nexus of complexity and public administration and its sub-fields. however, the examples discussed in this issue do not give us an indication of the extent to which complexity theory or the complexity sciences are taught in curricula worldwide. this contribution presents the results of a thorough internet search to identify those academic programs in political science, public administration and business administration, where complexity sciences are taught in conjunction with all matters public. the search across 193 countries led to selection of 108 programs that corresponded to some or all the defined criteria. we will present characteristics of the selected programs regarding quality of fit, level of studies, field of studies, and type of information about the complexity science approach available on the programs’ websites. keywords: teaching complexity; global survey; internet search introduction christopher pollitt, in his contribution to the edited volume by teisman, van buuren & gerrits (pollitt, 2009), commented that the complexity sciences, while featuring some useful reminders, would remain a fringe interest. almost a decade has passed since that statement. in the meantime, the complexity sciences have been taken up in different disciplines, including ours. they have proven to be most useful when informing ongoing research practices instead of trying to establish it as a discrete, ‘new’ science (byrne & callaghan, 2013; byrne, 2011; gerrits, 2012; morcol, 2002, 2012; teisman, buuren, & gerrits, 2009; teisman & gerrits, 2014). if researchers are more aware of the complexity of social reality, and if they tailor their research accordingly, they may be able to obtain better results. our field has always made the argument that there should be a close relationship between research and education, as can be seen in the contributions elsewhere in this special issue. the question then is to what extent this adoption of the complexity sciences in our research has trickled down to teaching programs across the globe. this contribution presents the results of a thorough internet search that was conducted in order to identify those academic programs in political science, public administration and business administration, where complexity sciences are taught in conjunction with all matters public. we first present the research methodology, then show results of the internet search in terms of regional distribution, and characteristics of selected programs. anevena ivanovic, bamberg graduate school of social sciences, germany, e-mail: nevena.ivanovic@uni-bamberg.de blasse gerrits, otto-friedrich-universität bamberg, germany, e-mail: lasse.gerrits@uni-bamberg.de http://dx.doi.org/10.20377/cgn-62 https://www.uni-bamberg.de/en/bagss/ complexity, governance & networks – vol. 4, no 1(2018) special issue: teaching complexity, p. 10-18 doi: http://dx.doi.org/10.20377/cgn-62 11 university of bamberg press research methodology the internet search was conducted in the period from may 2017 to october 2017 using the google search engine. google, while not perfect, was chosen as it is the most commonly used search engine and appears to be relatively effective for this task. to lower the bias in the search results, automatic filtering options for similar results were disabled, and search was conducted anonymously, starting with an empty browsing history for every search attempt separately. moreover, we considered that popularity bias in the order of the appeared results (due to e.g. pagerank algorithm), had a limited influence the selection of the final list of the programs, as all the results for every search attempt or country were screened regardless of their page rank. although these attempts could not completely diminish bias in the search results, we do believe that the precautions described were enough to provide approximate number of university programs fitting the target type. our search was conducted separately for all 193 countries that are member states of the un. those countries that do not appear to offer academic programs at all, or those countries where a located search on google was not available at the time of our search (e.g. montenegro) where left out. this resulted in 178 remaining countries. the subsequent search query was designed to include: 1. field of studies: public administration, public management, policy science, political science, innovation, business, and management. terms “innovation”, “business” and “management” were included in the query next to other terms closely linked to public administration, as preliminary search attempts provided insight that some programs with the orientation to these topics fit well to the target program. for example, some programs in public management are offered within business schools. we also included complexity science programs that offer specialization in one of the previously mentioned disciplines, e.g. programs that would cover the use of complexity sciences in questions central to our fields. last, but not least, some programs in disaster (relief) management were also included as they emphasized necessity of complexity theories approach in understanding and managing disasters and emergencies; 2. level of study: bachelor and master. we didn’t include post-graduate programs or phd programs. while these also constitute a form of education, post-graduate programs usually offer more flexibility to the students, so we cannot ascertain whether the topic mentioned is also actually taught. this is different for bachelor and master programs, where many courses are an obligatory part of the curriculum; 3. within the programs identified: the inclusion of complexity as a substantive topic. naturally, many universities advertise their programs by telling how complex the world has become, and how important it would be to enroll in their program. we only included those programs where the term “complexity” was used in course titles and / or in the course description. listed literature (e.g. a list of mandatory readings) was checked in case of doubt. this allowed us to differentiate programs that incorporate complexity sciences, from programs that just make a cursory reference to ‘complexity’. this delineation lead to the following search query: "public administration" or "public management" or "public policy" or "* policy" or political or innovation or business or management and masters and bachelors and complexity http://dx.doi.org/10.20377/cgn-62 complexity, governance & networks – vol. 4, no 1(2018) special issue: teaching complexity, p. 10-18 doi: http://dx.doi.org/10.20377/cgn-62 12 university of bamberg press the query was first deployed in the english language. we also translated the query to spanish, french, german, russian, serbian, croatian, macedonian, and bosnian. although using queries in other languages most widely spoken in asia or africa would enrich the results, we were limited to the languages available in the research team. the translated queries were used to search within spain, spanish speaking countries in south america, france and countries where french is in use at the academic level, belgium, germany, austria, switzerland, russia, serbia, croatia, bosnia and hercegovina, and macedonia. an additional search in dutch was not necessary because the english query returned the same results. other languages couldn’t be covered, but we aimed to cover most of the main languages used at universities. upon selection of the preliminary list of programs, every saved link was visited again to check for possible changes in program descriptions or access to websites. a preliminary list of programs was made based on the fit to the criteria regarding topic, level of studies and presence of the term “complex” on the website. this preliminary list included 517 programs. this list was further cleaned after closer examination of the websites’ contents. programs that included term “complexity” in a way not relating to complexity theories approach were excluded. additionally, programs that did not contain aspects of the target disciplines were excluded (e.g. some programs from the field of computer science with complexity approach were included in the preliminary list). after these steps, the final list included 108 programs. the selected programs were then categorized based on the target criteria given above. following previously listed criteria, target program could be described as a bachelor or master level education program where complexity theories are combined with aspects of public administration, public management, public policy, political science, business administration with a focus on public management, or related disciplines. the question about the quality of fit to the target characteristics could often not be answered with a simple ‘yes’ or ‘no’. we often found programs that seemed to fit well, while not necessarily fully-fledged pa-programs (or public management, policy science etc.). for example, programs in disaster management mentioned above (but not all) were included for that reason. the same goes for e.g. programs in computational social science that would offer courses on both complexity theory and on using that in analyzing issues in public policy, public administration etc. in these situations, when programs didn’t fully align with the target description, they were categorized as having a partial fit to the target program. programs categorized as having a full fit were those that complied with all our requirements (i.e. points 1 – 3, where 1 and 3 were the most important ones). all programs had to be checked manually to find enough clues for the assessment. both researchers were involved in this as well as in the subsequent coding. while our search was designed such that we would increase the likelihood of identifying all programs, there were some limitations. firstly, and as mentioned above, programs advertised in other languages than the ones we had at our disposal, could not be included. as such, we may have missed programs in some countries if they didn’t translate their offerings to any of the languages we used in our search queries. secondly, we depended on the information being publicly available. some universities required log-in details (e.g. student accounts) in order to access course descriptions. as such, we couldn’t assess the contents of those programs and the fit with the criteria. thirdly, we could only identify and assess those programs where the information was readable and searchable. while we were able to include information given on websites and in (downloadable) pdf and .doc files, we may have missed information relayed in other formats (e.g. in the shape of a .jpg file, e.g. an online poster). fourthly, we would sometimes be forced to assign a program to both bachelor and master if it appeared that the same or a similar course would be offered at both levels. our coding might have been less precise in those cases, as websites were not always clear about how courses tied in with programs. fifthly, some universities would offer the same or similar programs under different headers, e.g. students from two different programs http://dx.doi.org/10.20377/cgn-62 complexity, governance & networks – vol. 4, no 1(2018) special issue: teaching complexity, p. 10-18 doi: http://dx.doi.org/10.20377/cgn-62 13 university of bamberg press would attend the same course. we couldn’t always make out how this was organized at individual universities so, again, some categorization may have been off. finally, our search was performed on the territory of germany, and search engine adjusted the results for that to a certain extent, which could possibly explain a richer set of results for the search for germany. although the option to allow websites to track the user’s location was disabled, this is often not enough to completely hide this information when browsing. for example, programs at german universities, or generally results in the german language, sometimes appeared even when the search was locating other territories (e.g. asia). in these cases, we ignored those results because they did not result from the search locating precisely germany. results regional distribution of the selected programs out of 193 countries where located search was initially conducted, selected academic programs cover 40 countries (see figure 1). approximately one half of the selected programs (52.78%) are hosted by universities in europe, followed by programs from the territory of asia (14.81%), north america (13.89%), oceania (8.33%), africa (5.56%), south america (2.78%), and central america and caribbean (1.85%). countries with the highest number of selected programs include germany (14 programs), followed by united kingdom and the united states of america (10 programs each), australia, netherlands, and switzerland (six programs each), canada, denmark, italy, new zealand, and russia (three programs each), austria, chile, china, france, ghana, japan, mexico, norway, puerto rico, slovenia, sweden, and thailand (two programs each), belgium, bolivia, ireland, jordan, kyrgyzstan, malaysia, morocco, namibia, nigeria, oman, poland, portugal, somalia, spain, sri lanka, tajikistan, and turkey (one program each). http://dx.doi.org/10.20377/cgn-62 complexity, governance & networks – vol. 4, no 1(2018) special issue: teaching complexity, p. 10-18 doi: http://dx.doi.org/10.20377/cgn-62 14 university of bamberg press figure 1. number of selected programs per country (n=108) overall, 20 programs were found to have an excellent fit to the target program type. most of the best fitting programs were found in europe (netherlands, germany and uk – three programs in each country, denmark, sweden, italy – one program in each country), followed by four programs in north america (usa), three in oceania (australia), and one in asia (malaysia). the number of selected programs within every continent, including categories of the fit to the target program, are presented on the figure 2. http://dx.doi.org/10.20377/cgn-62 complexity, governance & networks – vol. 4, no 1(2018) special issue: teaching complexity, p. 10-18 doi: http://dx.doi.org/10.20377/cgn-62 15 university of bamberg press figure 2. number of selected programs with the good and excellent fit, per continent (n=108) characteristics of the selected programs most of the programs are on the master level (see table 1). close to 60% programs relate to fields of business or public administration, while rest of the programs cover other disciplines (see figure 3). lists of the offered courses were available on most of the websites, allowing to check for those introducing complexity theories. http://dx.doi.org/10.20377/cgn-62 complexity, governance & networks – vol. 4, no 1(2018) special issue: teaching complexity, p. 10-18 doi: http://dx.doi.org/10.20377/cgn-62 16 university of bamberg press table 1characteristics of all selected programs, and those categorized as excellent and good fit to the target program excellent fit good fit overall absolute frequency (n) relative frequency (%) absolute frequency (n) relative frequency (%) absolute frequency (n) relative frequency (%) level of studies master 18 90.00 57 64.77 75 69.44 bachelor 0 0.00 19 21.59 19 17.59 master and bachelor 2 10.00 12 13.64 14 12.97 field of studies political science 9 45.00 20 22.73 29 26.85 business 3 15.00 30 34.08 33 30.56 other disciplines 2 10.00 20 22.73 22 20.37 combination of disciplines 2 10.00 12 13.64 14 12.96 complexity (general) 4 20.00 6 6.82 10 9.26 complexity mentioned in course name 2 10.00 28 31.82 30 27.78 description of the program 10 50.00 41 46.59 51 47.22 both 8 40.00 19 21.59 27 25.00 course list available yes 14 70.00 71 80.68 85 78.70 no 6 30.00 17 19.32 23 21.30 total 20 100.00 88 100.00 108 100.00 http://dx.doi.org/10.20377/cgn-62 complexity, governance & networks – vol. 4, no 1(2018) special issue: teaching complexity, p. 10-18 doi: http://dx.doi.org/10.20377/cgn-62 17 university of bamberg press figure 3. number of selected topics per academic field (n=108) discussion and conclusion we started this contribution with pollitt’s observation that complexity would remain a fringe interest in our field. while we don’t have the data to confirm or reject that claim right now, at the very least we can observe that it hasn’t disappeared from our realm since his comments. in the light of that point, as well as in the light of this special issue, we wanted to take stock of the state of teaching complexity in education programs in our field. we found that majority of selected programs are from europe, north america or asia. the countries with highest number of selected programs are germany, uk and usa, followed by australia, switzerland and netherlands. best fitting programs are found in germany, netherlands, uk, usa, australia and malaysia. most of the programs are on the master level. most of them are in the field of business or political science. regarding the quality of information available on the websites, in most of the cases, information about the complexity approach within the selected programs was available in overall description of the program, or in both description and course name. for almost half of the selected programs, it is not clear whether they could provide training and more in-depth knowledge about complexity even though they do claim to incorporate complexity approach, as there was no indication of specific courses specialized in introducing complexity theories. as mentioned before, there are some obvious methodological limitations to our overview relating to number of languages used in the search queries, biases linked to the search engine, variety of the sources of information about university programs, and quality and availability of information on the websites. nevertheless, the chosen methodology provided an approximation of the number of programs we were after. other possible ways of screening for target programs were considered. more profound search using different sources of information, e.g. extracting list of universities from well-known complexity science institutes, departments, conferences, and 33 29 22 14 10 0 5 10 15 20 25 30 35 business public administration other disciplines combination of disciplines complexity (general) n u m b er o f p ro gr am s http://dx.doi.org/10.20377/cgn-62 complexity, governance & networks – vol. 4, no 1(2018) special issue: teaching complexity, p. 10-18 doi: http://dx.doi.org/10.20377/cgn-62 18 university of bamberg press journals, would account for possible misses of well-fitting programs whose websites didn’t include key words present in used search query. these caveats aside, the overview provided is a first. it tells us that complexity within public administration programs is being taught in quite a few places across the globe. it would be interesting to conduct this search again in about 5-10 years to identify a change in the number of programs, which could possibly tell us something about the trajectory of the perceived relevance of complexity for students and practitioners in our field. acknowledgements the authors would like thank nikolaus jopke for helping out with the german query. references byrne, d., & callaghan, g. (2013). complexity theory and the social sciences: the state of the art (1 edition). new york: routledge. byrne, d. s. (2011). applying social science: the role of social research in politics, policy and practice. the policy press. gerrits, l. (2012). punching clouds: an introduction to the complexity of public decision-making. litchfield, az: emergent publications. morcol, g. (2002). a new mind for policy analysis: toward a postnewtonian and postpositivist epistemology and methodology. westport, ct: praeger publishers. morcol, g. (2012). a complexity theory for public policy. new york: routledge. pollitt, c. (2009). complexity theory and evolutionary public administration: a skeptical afterword. in: teisman, g., buuren, a. van, & gerrits, l. (eds.). (2009). managing complex governance systems: dynamics, self-organization and coevolution in public investments. new york ; london: routledge teisman, g., buuren, a. van, & gerrits, l. (eds.). (2009). managing complex governance systems: dynamics, self-organization and coevolution in public investments. new york ; london: routledge. teisman, g., & gerrits, l. (2014). the emergence of complexity in the art and science of governance. complexity, governance & networks, 1(1), 17–28. https://doi.org/10.7564/14cgn2 http://dx.doi.org/10.20377/cgn-62 network features and processes as determinants of organizational interaction during extreme events complexity, governance & networks-vol 02, issue 01 (2015) 23–44 23 doi: 10.7564/14-cgn10 network features and processes as determinants of organizational interaction during extreme events michael d. sicilianoa* and clayton wukichb adepartment of public administration, university of illinois, chicago email: sicilian@uic.edu bdepartment of political science, sam houston state university email: wukich@shsu.edu despite the widely acknowledged importance of collaboration among participants in governance networks, a limited number of studies have attempted to statistically model the processes by which those networks form. in this article, we explore a range of network features and processes and measure their influence on network formation. we examine the case of hurricane katrina and employ exponential random graph models to identify the drivers of network formation in extreme events. we find that both the attributes of individual organizations and endogenous network processes affect organizational collaboration. understanding these factors is important because the structure of the response network influences information flow, resource exchange, and performance. keywords: network analysis, governance networks, disaster response, exponential random graph models 1. introduction disasters, whether natural or man-made, require multiple organizations to work together. members of disaster response networks mobilize to protect life and property as well as provide citizens with necessary goods and services (comfort, boin, & demchak, 2010; kapucu, 2006). while some collaborative action among participants is planned, many of the activities in these networks develop based on rapid decisions made by participants under conditions of stress and uncertainty (comfort, 1999; drabek & mcentire, 2002). while necessity brings together these organizations (weber & khademian, 2008), coordination and collaboration are not necessarily ensured, and the lack of coordination can lead to suboptimal results (comfort, 1999; kapucu, 2006; tierney, lindell, & perry, 2001). collective action problems have long been a focus of public administration research (frederickson, 1999). considerable attention has been given to the factors that influence * corresponding author. 55976-14-10.indd 23 18/08/15 5:31 pm 24 m. d. siciliano and c. wukich / network features and processes as determinants collaboration in stable operating conditions. research suggests that shared goals, the establishment of trust between organizations, and specific management skills facilitate interaction (bardach, 1999; bryson, crosby, & stone, 2006; feiock, 2007). scholars have also examined how and why agencies interact during extreme events (comfort, 1999, 2007; kapucu, 2006; waugh & streib, 2006). given the condensed time frame for interaction during emergencies, a greater importance is placed on an organization’s capacity to quickly develop and communicate strategies to coordinate action (comfort, 2007; kapucu, 2006). while there is a large body of research on the importance of collaboration among participants in governance networks (agranoff, 2007; provan & milward, 1995), a limited number of studies have attempted to statistically model the processes by which those networks form. according to snijders (2011), the dependency among the ties in a network have hindered the development of suitable statistical network models. consequently, previous work has often relied on qualitative case studies (see agranoff, 2007), or traditional linear methods which assume that the decision maker contemplating collaboration operates in isolation (see feiock, 2004). even when network analysis has been used to study governance networks, the techniques employed tended to offer only descriptive measures of network properties rather than methods that capture generative processes (kapucu, hu, & khosa, 2014). in this study, we use exponential random graph models (ergms) to examine several network generating processes simultaneously while also appropriately handling the structural dependencies present in the data (lusher, koskinen, & robins, 2013). we begin with a review of factors identified in previous research as important determinants of network structure. we then incorporate those factors into a statistical model to examine the interagency governance network that formed in louisiana in response to hurricane katrina in 2005. we pay careful attention to model construction and interpretation in order to provide a clear application of ergms for other researchers who may be interested in applying the methodology. 2. interorganizational network formation it is critical to understand the factors that influence collective action and network formation in public administration research (bardach, 1998; frederickson, 1999). two streams of research in public administration are particularly relevant for their insights on the factors that shape and constrain interagency relationships: (i) institutional collect action (ica) (feiock, 2004, 2013) and (ii) network management studies (agranoff, 2007; agranoff & mcguire, 2003), particularly those concerned with disasters (comfort, 1999; kapucu, 2006). work on ica suggests that organizations engaged in collaborative arrangements accrue transaction costs in gathering information, negotiating agreements, coordinating joint efforts, and monitoring the performance of various partners (feiock, 2004, 2007). the key task, therefore, is to mitigate these costs by establishing shared values and mutual 55976-14-10.indd 24 18/08/15 5:31 pm m. d. siciliano and c. wukich / network features and processes as determinants 25 trust (carr, leroux, & shrestha, 2009; romzek, leroux, & blackmar, 2012). in stable operating environments, interorganizational networks develop over time as organizations assess the credibility and reliability of potential partners (gulati & gargiulo, 1999). unlike practitioners operating under stable conditions, organizations involved in emergency response have very limited time to contemplate decisions regarding interagency collaboration (comfort, 1999; kapucu, 2006). under conditions of uncertainty, individuals and organizations may be forced to rely on organizational attributes, such as sector or level of government, to make assessments of prospective partners. attribute similarity can potentially reduce transaction costs (feiock & scholz, 2010) because it facilitates the establishment of trust and mutual expectations (brass, 1995). the tendency for ties to form among actors with similar attributes or traits is known as homophily (lazarsfeld & merton, 1954). while recent research on governance and policy networks in stable operating environments have produced mixed results with regard to the importance of homophily (henry, lubell, & mccoy, 2011; lee, lee, & feiock, 2012), the speed with which collaborations must form during crises suggests homophilous tendencies will be more pronounced in disaster response systems. hypothesis 1: within the response network, organizations are more likely to form homophilous relationships. another means of reducing transaction costs is to rely on common partners to broker new relationships (kwon, feiock, & bae, 2014; thurmaier & wood, 2002). certain individuals and organizations play key roles in network formation by linking otherwise disparate organizations. studies have identified key actors in response networks that both attract collaborative partners and connect others due to their roles in disaster response plans or their access to information and resources (comfort & haase, 2006; kapucu, 2006; kapucu & demiroz, 2011). when two organizations rely on a common partner to broker interaction, a network structure known as a transitive triad emerges where all three organizations collaborate with each other. transitivity captures the common adage that a friend of a friend is a friend (wasserman & faust, 1994). transitivity has been empirically observed in several studies on policy and governance networks (henry et al., 2011; lee et al., 2012). in a disaster scenario, agencies may rely on existing partners for referrals to connect with prospective partners. these referrals may help to promote trust and diminish the perceived transaction costs associated with quickly forming an ad hoc partnership. hypothesis 2: within the response networks, organizations that share partners are more likely to interact with each other (i.e., form transitive relations). while research using the ica framework sheds light on the context and decision space confronted by public managers, studies on network management offer insight on the specific skills demonstrated by successful network participants (agranoff, 2007). network management scholars have identified the abilities to span jurisdictional boundaries and recruit potential members as valuable skillsets (agranoff, 2007; agranoff & mcguire, 2001; goldsmith & eggers, 2004). in many governance networks, organizations will take 55976-14-10.indd 25 18/08/15 5:31 pm 26 m. d. siciliano and c. wukich / network features and processes as determinants lead responsibility for various administrative and operational tasks, and as a result these organizations assume central roles (provan & kenis, 2008). in disaster response networks, central agencies generally possess significant resources or direct access to resources and information (comfort & haase, 2006; robinson, eller, gall, & gerber, 2013). research has found that the federal emergency management agency, state-level agencies charged with response duties, and local public safety departments are often at the core of disasterresponse networks (comfort & haase, 2006; kapucu, 2006; kapucu & demiroz, 2011). with respect to social network analysis, the degree of an actor is a simple measure of activity or centrality. it is calculated by summing the number of ties that an actor has formed with other members of the network. under a completely random process, the distribution of those ties across all actors, or the degree distribution, follows a normal curve. however, in most social and organizational networks, degree distributions follow a power law distribution (barabási, 2002; barabási & albert, 1999). these networks are often centralized, as a few nodes are highly connected while the majority of nodes have only a few connections. hypothesis 3: within a disaster response network, a handful of key organizations will be disproportionately connected throughout the system (i.e., the network will tend to be centralized). another factor influencing organizational interaction is the concept of propinquity, which captures the physical distance separating two actors. there exists a strong tendency to interact with those in a shared physical space (axelrod & cohen, 1999). some studies suggest that as the distance between two individuals increases, the probability of interaction drops exponentially (allen, 1984; krackhardt, 1994). as with individuals, organizations operating within close geographic proximity may also be more likely to collaborate due to the increased ability to share resources or services (post, 2004) and the reductions in the transportation costs of goods (owen-smith & powell, 2004). hypothesis 4: within the response network, organizations with headquarters located in close geographic proximity will be more likely to collaborate. taken individually, the extant literature on homophily, transitivity, degree distributions, and propinquity is fairly well established. however, much less is known about how these network properties and processes operate simultaneously. for example, how homophily and propinquity function together in generating network connections has not been adequately studied (reagans, 2011). also, transitivity in a network can arise via triadic closure through reliance on shared partners or it can result from homophilious tendencies to seek out partners with similar attributes (robins, pattison, kalish, & lusher, 2007; zaccarin & rivellini, 2010). thus, when transitivity is observed in a network, it is not possible to disentangle the effects of triadic closure from homophily. in order to do so, one needs to use a statistical model, like an exponential random graph model, capable of modeling multiple processes simultaneously (lusher et al., 2013; robins, 2011; robins et al., 2007). this approach has not been widely used to study governance networks. in their review on the state of network research, kapucu et al. (2014) found only 7 articles across thirty-nine public administration journals over a fifteen year time span that used ergms. 55976-14-10.indd 26 18/08/15 5:31 pm m. d. siciliano and c. wukich / network features and processes as determinants 27 3. methods exponential random graph models allow for multiple network generating processes to be tested together (robins, snijders, wang, handcock, & pattison, 2007; snijders, pattison, robins, & handcock, 2006). with these methods, researchers can propose and test micro-level processes that may be at work in generating the observed network (goodreau, 2007). the observed network is simply the network for which the researcher has collected data and is interested in modeling (robins et al., 2007). according to robins et al. (2007), “the network is conceptualized as a self-organizing system of relational ties. substantively, the claim is that there are local social processes that generate dyadic relations, and that these social processes may depend on the surrounding social environment (i.e., on existing relations)” (p. 177). even though the observed network may only be a snapshot of an evolving system, due to the stability and constancy of the micro-level processes driving actor interaction, particular patterns of tie formation will emerge from the observed data (robins, 2011). “these patterns of network ties are indeed the structural signature of the network and provide evidence from which we may infer something of the social processes that build the network” (robins, 2011, p. 484). based on the notation and terminology of robins et al. (2007), exponential random graph models have the following general form: pr exp ( )y y k g ya a a 5 5( )    { }∑1 η where (i) y is a particular realization of the network; (ii) ga(y) are the network statistics corresponding to the network terms or configurations, a, included in the model (i.e., transitivity, homophily, etc.); (iii) ha is the coefficient for a given network statistic; and (iv) k is a normalizing constant to ensure a proper probability distribution. we estimate these models using the ergm package of the statnet suite (handcock, hunter, butts, goodreau, & morris, 2008) in the r programming environment (r core team, 2013). in applying ergms to the network that emerged in response to hurricane katrina, we illuminate potential factors influencing network formation in extreme events. 4. case selection and data in august 2005, hurricane katrina made landfall across the gulf coast, destroying infrastructure and overwhelming communities from florida to louisiana. more than 1,800 people lost their lives in one of the most deadly and, up to that point, the most costly natural disaster in united states history (gall & cutter, 2012). louisiana bore the brunt of the storm as the subsequent flooding of new orleans decimated infrastructure and created a major humanitarian crisis. researchers in public administration and other disciplines paid considerable attention to the case (cigler, 2007; comfort, 2005; comfort & haase, 2006; farazmand, 2007; kapucu, augustin, & garayev, 2009; kapucu & garayev, 2011; kiefer & montjoy, 2006; waugh, 2007; waugh & streib, 2006). explanations for the failure of initiative that characterized the government response (u.s. house of 55976-14-10.indd 27 18/08/15 5:31 pm 28 m. d. siciliano and c. wukich / network features and processes as determinants representatives, 2006) tended to emphasize either individual or system wide failures. the mass media tended to focus on certain individual actors and their failings (e.g., president george w. bush’s perceived disengagement, the inefficacy of fema leadership, governor kathleen blanco’s lack of precision, and mayor ray nagin’s failure to organize an effective evacuation). academic research examined what systemic factors contributed to poor outcomes ( birkland & waterman, 2008; cigler, 2007; comfort, 2006; waugh, 2007). this included work on the effectiveness of mutual aid agreements (kapucu & garayev, 2011; waugh, 2007) as well as the constraints imposed on network formation related to federalism ( birkland & waterman, 2008). comfort, kapucu, and their colleagues, illustrated the features of the response system in terms of the actors involved, their interactions, and the flow of information and other resources (comfort & haase, 2006; comfort, oh, & ertan, 2009; kapucu et al., 2009). findings demonstrated that while certain actors, such as fema, were central in the network, asymmetries existed in which many communities and organizations were not incorporated into the larger response system (comfort & haase, 2006). thus, while previous studies were able to describe the relevant characteristics and important properties of the response network as a whole, they did not allow for inferences regarding the processes that may have influenced the formation of the network. in this article, we complement prior studies by focusing on the interconnections among the organizations in the response system rather than on individual actors or whole network properties. we attempt to advance previous research by employing ergms to statistically model the interactions among the agencies responding to hurricane katrina. our data extends an existing dataset previously used by researchers to understand the structure and composition of the katrina response system (comfort & haase, 2006; comfort et al., 2009). the original data were derived from a content analysis of daily news reports (230 articles) published in the new orleans times picayune and characterizes the response network that formed just prior to, during, and after hurricane katrina (a period of 24 days). for additional information on the data, please see comfort and haase (2006) and comfort et al. (2009). in total, 528 organizations1 were identified along with 673 interactions. the network was then symmetrized to create an undirected network. table 1 provides 1 this number is slightly less than the number originally reported by comfort, oh, and ertan (2009). the difference was due to our merging of several organizations. for example, we combined the louisiana state university manship school for mass communications with the louisiana state university school of journalism as these are the same entity. likewise, we also combined west jefferson general hospital with west jefferson medical center as these names are referring to the same organization with the same physical address. table 1 level of jurisdiction and sector for the organizations operating in the katrina response system city/county state national international total private 23 39 74 6 142 non-profit 30 26 23 3 82 public 136 87 70 11 304 total 189 152 167 20 528 55976-14-10.indd 28 18/08/15 5:31 pm m. d. siciliano and c. wukich / network features and processes as determinants 29 figure 1. geographic locations of organizations operating in the katrina response system. a summary of the level of jurisdiction and sector for the organizations operating in the katrina response. we extended the data by calculating the physical distance between each organization as a dyad level or edge attribute. figure 1 displays a map of the locations of these organizations. 5. variables and measurement we test our hypotheses by first operationalizing the following network features and processes: homophily, transitivity, degree distribution, and propinquity. our data contain information on each organization’s jurisdiction and sector. we assume that due to homophily organizations may be more likely to interact with others who operate within the same sector or on the same jurisdictional level (lee et al., 2012; robinson et al., 2013). to examine the role of homophily in generating ties, we constructed two different variables. the first one was jurisdictional homophily, which indicates whether or not two organizations operate on the same geographic scale (e.g., local, state, national, international). the second 55976-14-10.indd 29 18/08/15 5:31 pm 30 m. d. siciliano and c. wukich / network features and processes as determinants one was, sectoral homophily, which indicates whether or not two organizations operate within the same sector (e.g., public, nonprofit, private). rather than capturing an overall homophily effect for each attribute, we assumed that different homophilious tendencies were likely to exist across the categories of a given attribute. this assumption, known as differential homophily, allows the influence of organizational similarity to be estimated for each jurisdiction type and each sector independently. closed triads and clustering are common features in real networks. early ergms often used a triangle statistic, which was simply a count of the number of closed triads, to account for transitivity. however, the use of a triangle term often led to degenerate network models where the observed network structure could not be recreated from the statistical model (handcock, 2003; snijders, 2002). to help avoid degenerate models, we used a geometrically weighted edgewise shared partner (gwesp) statistic, instead of a simple triangle term to account for transitivity. gwesp is one of several geometrically weighted terms developed by snijders et al. (2006) and extended by hunter and handcock (2006) and hunter (2007) to deal with problems of degeneracy that can arise in ergms. using the notation of hunter (2007), the gwesp statistic is defined as: 5 ∑θ − −θ θ− = − v y e e ep y( : ) {1 (1 ) } ( )i i n i 1 2 where, y is the network, q is the decay parameter, and epi (y) is the number of edges in the network that have i shared partners. many observed networks, including the katrina response, appear to follow a power law degree distribution. in order to investigate the network that emerged in response to hurricane katrina, a geometrically weighted degree (gwd) statistic can be used to help account for degree differentials and is preferred over k-star parameters. gwd is defined by hunter (2007) as: 5 ∑θ − −θ θ− = − u y e e d y( : ) {1 (1 ) } ( )i i n i 1 1 and as with the gwesp statistic, y is the network, q is the decay parameter, and di(y) is the number of nodes in the network with degree i. for a more detailed discussion on the development and use of geometrically weighted dependency terms in ergms see hunter and handcock (2006) and hunter (2007). organizational attributes may also play a role in determining which organizations are most active in the response system. for example, public organizations may be engaged in more collaborative relationships when compared to private organizations. these differences are often referred to as main effects and entered into the model as a set of dummy variables for the organization’s jurisdiction and sector. propinquity is measured by calculating the geographic distance between the organizations in the response system. we assigned longitude and latitude coordinates for each organization. using geospatial analysis, the locations for the organizations responding to katrina 55976-14-10.indd 30 18/08/15 5:31 pm m. d. siciliano and c. wukich / network features and processes as determinants 31 were geocoded using esri’s composite u.s. geocoder. these values were then geoprocessed to produce a table of distances between every organization. the distances were stored in a matrix and enter into the model as an edge covariate. thus, it measures the likelihood of a tie forming between two organizations based on the physical distance between them. 6. analyses and results our analyses involve three different types of network models. we begin with a null or random graph model and then move through dyadic independent and finally to dyadic dependent models. this analytic strategy identifies how coefficients change when new terms are added to the model. for instance, homophilous effects that were significant in a dyadic independent model may no longer be present when controlling for transitivity. while we explore multiple models, it is important to note that only the final, combined model, offers an unbiased estimate of the coefficients. we used the final combined model to assess model adequacy, overall importance of network processes, and goodness of fit statistics. the initial null or random graph model operates as a baseline. the model is built on the simple and unrealistic assumption that ties between organizations in the response system form completely at random. the model includes only an edge term, which functions in the same manner as the intercept in a standard linear model. it is simply the best guess table 2 random and dyadic independent models of the katrina response system dyadic independent models null model main effects homophily propinquity structural effects edges −5.954 (0.053)*** −9.899 (0.776)*** −6.863 (0.108)*** −5.757 (0.068)*** main effects sector nonprofit −0.034 (0.164) sector public 0.929 (0.108)*** jurisdiction city/county 1.252 (0.383)** jurisdiction national 1.549 (0.384)*** jurisdiction state 1.340 (0.385)*** homophily effects both nonprofit 0.891 (0.317)** both private −0.141 (0.287) both public 1.299 (0.118)*** both city/county 0.423 (0.143)** both national 0.976 (0.155)*** both state 0.750 (0.162)*** propinquity distance −0.020 (0.005)*** aic 5010.141 4873.782 4838.192 4986.189 bic 5019.984 4932.841 4907.094 5005.875 log likelihood −2504.070 −2430.891 −2412.096 −2491.094 ***p < 0.001, **p < 0.01, *p < 0.05 55976-14-10.indd 31 18/08/15 5:31 pm 32 m. d. siciliano and c. wukich / network features and processes as determinants at the probability of a tie forming between any two organizations assuming nothing else about the organizations or network structure is known. building on the null model, we add dyadic independent terms one at a time. these terms are the homophily related effects, the main effects of nodal attributes, and edge covariates. these terms are considered dyadic independent because their effect on tie formation is assumed to be exogenous of the network structure. thus, the complicated dependencies inherent in network data are ignored allowing dyadic independent models to be considered within a standard logistic regression framework (goodreau, kitts, & morris, 2009; hunter, handcock, butts, goodreau, & morris, 2008). the results for the null and dyadic independent models are presented in table 2. the parameter estimates for the network statistics included in the models can be interpreted as the log-odds (logit) of individual ties. the general form for ergms displayed previously can be rewritten to express the conditional logit of tie formation:2 ∑η ϑ( )logit p y n actors y g y( 1 | , ) ( )ij ijc a a a 5 5 where yij c denotes all dyads other than yij, and ϑ g y( )a is the change in ga (y) when (yij) is toggled from 0 to 1 (goodreau et al., 2009). as goodreau et al. (2009) note, the logit formulation clarifies the interpretation of the h vector: if forming a tie increases ga by 1, then all else being equal the logit of that tie forming increases by ha. thus, relying on this interpretation, we can easily calculate the odds of a tie forming, exp (∑ahaϑga (y)), as well as the probability of a tie forming, ∑ ∑ η ϑ η ϑ ( ) ( )+ exp g y exp g y ( ) 1 ( ) a aa a aa . in the simple random model, the probability is based solely on the edge term. the probability of tie formation is 0.0026, which we calculated using the following equation: ( ) ( ) − + − exp exp 5.594 1 5.594 . this probability is directly equal to the overall density of the undirected network. again, such a model is rather uninformative and unable to capture the structural features of the observed network. adding dyadic independent terms to the model can help explain some of the structural features. the main effects model indicates the differences in the activity level of different organizational types compared to the base category. the base category for jurisdiction was 2 while we can use a logistic regression framework to aid in the interpretation of the parameter estimates, we cannot use logistic regression to establish the parameter estimates in dyadic dependent models. 55976-14-10.indd 32 18/08/15 5:31 pm m. d. siciliano and c. wukich / network features and processes as determinants 33 set to international and the base category for sector was set to private. the model suggests that public organizations are significantly more active compared to private organizations. for jurisdiction, the model suggests that city/county, national, and state level organizations are each significantly more active compared to international organizations. it is instructive to calculate the probability of a tie forming between two specific organizations. using the results of the main effect model, the probability of a tie between a public national organization and a public state organization can be calculated. the logit value is found by multiplying the statistics ga(y) by their estimated coefficients in the model. here, we need to include the fact that the dyad of interest concerns two public organizations, one of which is a national organization, and the other which is a state organization. using the transformation from log odds to probability gives us: ( ) ( ) − × + × + × + × + − × + × + × + × exp 9.899 1 .929 2 1.549 1 1.340 1 1 exp 9.899 1 .929 2 1.549 1 1.340 1 the resulting probability of the tie, 0.006, can be compared with another dyad where the public national organization is switched to a public international organization (i.e., the base category for jurisdiction). by doing so we obtain a probability of 0.001. holding other attributes of the dyad constant, switching one of the nodes from a national organization to an international organization makes a tie between those organizations less likely. the homophily model examines the role of organizational similarity in tie formation. the model allows us now to investigate the question: are agencies more likely to interact with similar agencies or might they reach out to dissimilar agencies in search of novel information, skills, or other needed resources? in terms of sector based homophily, both nonprofits and public organizations showed tendencies to form ties with organizations in their own sector. for jurisdictional homophily, an increased likelihood in tie formation among organizations of the same jurisdictional scale was found for city/county, state, and national level organizations. homophily for international organizations was not included in the model as there were no international to international ties in the data. the last dyadic independent model, the propinquity model, uses the edge covariate of geographic distance between organizations as a predictor of tie formation. the model suggests that organizations which are separated by greater geographic distance are less likely to work with one another in the response system. each of these dyadic independent terms was combined together in a final model to assess their conditional effect when controlling for other organizational attributes and structural dependencies in the network. the dyadic dependent variables used to capture the structural dependencies are gwesp and gwd. the results of the models for these two terms along with the final combined model that includes all of the dyadic independent and dyadic dependent terms are presented in table 3. the transitivity model indicates that there is a strong tendency for triadic closure to occur in the katrina response system. the positive and significant effect on gwesp indicates that a collaborative tie was more likely to form between two organizations who 55976-14-10.indd 33 18/08/15 5:31 pm 34 m. d. siciliano and c. wukich / network features and processes as determinants shared a common partner. thus, the positive coefficient for gwesp indicates a tendency toward transitivity in the network. in terms of the degree distribution, the negative and significant coefficient indicates that the degree distribution is not centralized but is rather spread more equally across the nodes in the network (koskinen & daraganova, 2013). moving to the final model, we found several significant changes in coefficient size and significance. the most noticeable one is the lack of significance for the main effects. this suggests that other factors in the model were driving the differences in organizational activity. for example, the main effect of city/county is no longer significant, and thus organizations operating at this level are not more active than the base category of international organizations controlling for the other effects. in the final model, city/county has a positive and significant homophily effect, and there is a general tendency to form ties with those nearby. these effects indicate that city/county organizations are unlikely to form ties in the response system unless those ties are with other city/county organizations and/or to other organizations operating within close geographic proximity. the homophilous tendencies of city/county organizations also help explain why coefficient on propinquity has cut in half in the final model. another notable difference between the single process models and the final combined model is the decrease in the size of the coefficient on gwesp. as mentioned earlier, table 3 dyadic dependent and final models of the katrina response system transitivity degree distribution final model structural effects edges −6.244 (0.061)*** −4.487 (0.031)*** −5.622 (0.761)*** gwesp 1.639 (0.081)*** 0.998 (0.089)*** gwd −1.837 (0.049)*** −1.367 (0.111)*** main effects sector – nonprofit −0.371 (0.193) sector public 0.025 (0.374) jurisdiction city/county −0.024 (0.339) jurisdiction national 0.378 (0.329) jurisdiction state −0.004 (0.338) homophily effects both nonprofit 1.854 (0.485)*** both private 0.101 (0.476) both public 0.592 (0.400) both city/county 0.649 (0.241)** both national 0.149 (0.263) both state 0.845 (0.251)*** propinquity distance −0.010 (0.005)* aic 4790.540 4752.930 4530.250 bic 4810.226 4772.616 4677.897 log likelihood −2393.270 −2374.465 −2250.125 ***p < 0.001, **p < 0.01, *p < 0.05 55976-14-10.indd 34 18/08/15 5:31 pm m. d. siciliano and c. wukich / network features and processes as determinants 35 transitivity in networks can arise through homophily as well as through triadic closure. given the drop in the size of the coefficient, it suggests that part of the impact of gwesp in the transitivity model was due to homophily. once homophily was accounted for in the network, the tendency for triangles to form in the network is reduced, though still significant. 7. model fit like other statistical models, ergms can be assessed for overall model fit. one of the primary problems resulting from the use of ergms is model degeneracy. when degeneracy occurs, the model may either fail to converge or if it does converge, the simulated networks based off of the model parameters may produce structural features and patterns that are quite different from the observed data. each of the models presented above successfully converged. within the ergm package in r, a set of diagnostic tools are available to assess the overall fit and adequacy of the model. please see the appendix for a description of these tools and a discussion of model fit with regards to the katrina dataset. 8. discussion our goal was to examine the tie formation in the hurricane katrina response system that emerged in louisiana in 2005 and to test specific hypotheses regarding the network features and processes of homophily, transitivity, degree distributions, and propinquity. in doing so, we provide readers with a clear application of ergms. previous studies have focused on individual actors and/or aggregate network properties, but less emphasis has been placed on the microprocesses and factors that may be at work in generating the system’s structure. the use of ergms allowed us to describe and test several potential processes and to explore their importance for establishing collaborative ties among organizations. the final combined model suggests that certain types of agencies sought to cluster with each other, as opposed to reaching out to more diverse organizations. nonprofits were generally more likely to work with other nonprofits, which supports our first hypothesis. nonprofits perhaps did not have access to state and federal resources in a timely and effective manner. rather than a proxy for trust, nonprofit homophily might be explained based on the cooperative activities in which the organization engaged. nonprofits make up a substantial portion of the united states’ emergency support function (esf) #6, for example, which includes mass care, housing, and human services. the american red cross, the salvation army, and many national faith-based organizations took part in this support function in louisiana and actively coordinated efforts following hurricane katrina. significant homophily effects may therefore indicate potential information asymmetries or conscious groupings of agencies around support function activities. in addition, local agencies, those operating at the city or county level, were generally more likely to work with other local agencies. this finding along with the statistically 55976-14-10.indd 35 18/08/15 5:31 pm 36 m. d. siciliano and c. wukich / network features and processes as determinants significant effect of propinquity provides evidence in support of our fourth hypothesis regarding the role played by geographic proximity. the interactions among local organizations may have been influenced by their physical proximities, but may also have been a product of past relationships, familiarity between personnel, and established levels of trust, all of which are considered to be initial conditions for cooperation (bryson et al., 2006). the reliance of local organizations on other local organizations could also be due to the poor collaborative capacity demonstrated by state and federal agencies. the failure of initiative that characterized state and federal response efforts has been well documented (u.s. house of representatives, 2006). in support of our second hypothesis, the results suggest a strong tendency for responding organizations to form transitive structures. because the model controls for homophily, this tendency toward transitivity exists over and above the clustering effects driven by organizational similarity. the importance of transitivity has been demonstrated in stable operating environments including regional planning networks (henry et al., 2011), adult basic education policy networks (park & rethemeyer, 2014), and in networks related to economic development (lee et al., 2012). our study demonstrates that transitivity is also prevalent in the networks that emerge in response to extreme events. coupled with the homophily effects, the results suggest that the organizations use bonding strategies that is they work closely with multiple partners to achieve desired outcomes (see wukich & robinson, 2013). the coefficient on gwd, which captures the degree distribution of the network, was negative and significant. this result suggests that nodes with a high number of ties were unlikely and thus the network tended to be less centralized. this finding fails to support our third hypothesis. however, certain organizations, based on their sheer number of connections, were more central in the network than others. for these nodes that did have a high degree in the response network, their positions were likely driven by other effects in the model, such as transitivity. as discussed by de la haye, robins, mohr, and wilson (2010), this indicates that organizations which formed numerous ties generally did so in the context of clusters of organizations. 9. limitations and future research while the katrina dataset provided a detailed account of organizational interactions during the first three weeks of the response, some limitations of our study should be noted. rather than viewing daily time slices of the network, the interactions are aggregated to form a picture of the overall response system that emerged immediately after the disaster. while such aggregation does not allow for dynamic network modeling it may help reduce errors in the data. reductions in error are likely as collaboration between two organizations may only be reported by a newspaper on a particular day but the collaboration may have begun earlier and could endure longer than reported. this reality of using archival sources, suggests that aggregation at the appropriate time frame is needed to capture the response network. 55976-14-10.indd 36 18/08/15 5:31 pm m. d. siciliano and c. wukich / network features and processes as determinants 37 while our data controlled for several attributes of the organizations engaged in the response, future research on disaster and governance networks should consider the inclusion of additional covariates. additional variables that could be explored include organizational resources and capabilities (i.e., finances, equipment, personnel). data on these attributes would provide researchers with the capacity to examine, among other things, the role of resource dependency in network formation. future work can build on the network processes and models explored in this paper by adding specific managerial factors into the model that lead to collaboration (agranoff, 2007; agranoff & mcguire, 2001; goldsmith & eggers, 2004) as well as measures of trust or preexisting relationships. statistical models of networks, like ergms, provide researchers with an improved ability to test theory and hypotheses compared to more descriptive network methods or case studies (lubell, scholz, berardo, & robins, 2012; robins, lewis, & wang, 2012). consequently, one fruitful area of future research would be in the comparison of response systems within the same state or country before and after a major policy change or transition in leadership in key organizations. a comparative approach using statistical network models would allow scholars to identify if policy changes altered the importance of or type of factors affecting organizational interaction. the post-katrina emergency management reform act of 2006, for example, enables federal agencies to more proactively engage state and local governments during preparedness, response, and recovery periods. the extent to which a more active federal element might influence network formation and performance is a point of future research. such dynamics could be included in an ergm to explicitly test changes in the propensity for cross jurisdictional ties to form. once we understand some of the underlying processes predicting tie formation, and how those processes have changed in response to policy changes, disaster managers and policymakers can more effectively develop policies to govern the response to extreme events. future research could greatly add to statistical modeling by engaging in simultaneous and detailed qualitative interviews to better recognize the decision processes of organizational leaders as well as organizational capacity to identify where needed resources and expertise exist. 10. conclusions our study builds on previous findings by exploring the network processes that influence organizational collaboration and by illuminating the generative processes at work in building the overall response network. while previous studies identified individual characteristics in isolation and described the network structure, we statistically modeled how homophily, transitivity, propinquity, and organizational attributes simultaneously shape the network. we found that both the attributes of individual actors and endogenous network processes influence the structure of disaster response systems. understanding these processes adds to our ability to evaluate complex systems. these processes may be prominent in emergency situations due to the level of uncertainty and speed by which organizations must make decisions and act. in these situations organizations may naturally rely on 55976-14-10.indd 37 18/08/15 5:31 pm 38 m. d. siciliano and c. wukich / network features and processes as determinants existing partners or similar organizations to form new ties rather than seeking out other organizations that may be more suitable for solving problems. the inefficiencies that can result from this type of connection-making offer support to previous calls to improve communication and coordination during response and to engage in additional networking and preparation beforehand (comfort, 2006; comfort & haase, 2006; kapucu, 2006). therefore, emergency managers must understand and exercise existing plans, but also identify possible partners in the event of catastrophe beyond current planning capabilities, the so-called black swan events (taleb, 2010). this will allow for agencies to rely less on their current network partners and span boundaries to develop relationships or at least identify available resources and contacts prior to an extreme event. we suggest that managers not just count on existing written plans or relationships with their neighbors, but actually exercise those plans, anticipate potential contingencies, and make contact with potential partners in an effort to better access necessary information and resources during an extreme event. 11. acknowledgements the authors wish to thank professor louise k. comfort of the university of pittsburgh for her guidance and generous access to data. she and her team at the center for disaster management (cdm) collected the data used in this article, and they published several papers related to the incident. for more information on cdm please visit http:// www.cdm.pitt.edu/. references agranoff, r. (2007). managing within networks: adding value to public organizations. washington, dc: georgetown university press. agranoff, r., & mcguire, m. (2001). big questions in public network management research. journal of public administration research and theory, 11(3), 295–326. agranoff, r., & mcguire, m. (2003). collaborative public management: new strategies for local governments. washington, dc: georgetown university press. allen, t. j. (1984). managing the flow of technology: technology transfer and the dissemination of technological information within the r&d organization. cambridge, ma: mit press books. axelrod, r. m., & cohen, m. d. (1999). harnessing complexity: organizational implications of a scientific frontier. new york, ny: free press. barabási, a.-l. (2002). linked: the new science of networks. cambridge, ma: perseus. barabási, a.-l., & albert, r. (1999). emergence of scaling in random networks. science, 286(5439), 509– 512. doi:10.1126/science.286.5439.509 bardach, e. (1998). getting agencies to work together: the practice and theory of managerial craftsmanship. washington, dc: brookings institution press. bardach, e. (1999). getting agencies to work together. washington, dc: the brookings institute. birkland, t., & waterman, s. (2008). is federalism the reason for policy failure in hurricane katrina? publius: the journal of federalism, 38(4), 692–714. doi:10.1093/publius/pjn020 brass, d. j. (1995). a social network perspective on human resources management. research in personnel and human resources management, 13(1), 39–79. 55976-14-10.indd 38 18/08/15 5:31 pm m. d. siciliano and c. wukich / network features and processes as determinants 39 bryson, j. m., crosby, b. c., & stone, m. m. (2006). the design and implementation of cross-sector collaborations: propositions from the literature. public administration review, 66(1), 44–55. carr, j. b., leroux, k., & shrestha, m. (2009). institutional ties, transaction costs, and external service production. urban affairs review, 44(3), 403–427. doi:10.1177/1078087408323939 cigler, b. a. (2007). the “big questions” of katrina and the 2005 great flood of new orleans. public administration review, 67, 64–76. doi:10.1111/j.1540-6210.2007.00814.x comfort, l. k. (1999). shared risk: complex systems in seismic response (1st ed.). amsterdam, the netherlands: pergamon. comfort, l. k. (2005). risk, security, and disaster management. annual review of political science, 8(1), 335–356. doi:10.1146/annurev.polisci.8.081404.075608 comfort, l. k. (2006). cities at risk: hurricane katrina and the drowning of new orleans. urban affairs review, 41(4), 501–516. doi:10.1177/1078087405284881 comfort, l. k. (2007). crisis management in hindsight: cognition, communication, coordination, and control. public administration review, 67, 189–197. doi:10.1111/j.1540-6210.2007.00827.x comfort, l. k., boin, a., & demchak, c. c. (2010). designing resilience: preparing for extreme events. pittsburgh, pa: university of pittsburgh press. comfort, l. k., & haase, t. w. (2006). communication, coherence, and collective action: the impact of hurricane katrina on communications infrastructure. public works management & policy, 10(4), 328–343. doi:10.1177/1087724x06289052 comfort, l. k., oh, n., & ertan, g. (2009). the dynamics of disaster recovery: resilience and entropy in hurricane response systems 2005–2008. public organization review, 9(4), 309–323. doi:10.1007/ s11115-009-0098-3 de la haye, k., robins, g., mohr, p., & wilson, c. (2010). obesity-related behaviors in adolescent friendship networks. social networks, 32(3), 161–167. doi:10.1016/j.socnet.2009.09.001 drabek, t. e., & mcentire, d. a. (2002). emergent phenomena and multiorganizational coordination in disasters: lessons from the research literature. international journal of mass emergencies and disasters, 20(2), 197–224. farazmand, a. (2007). learning from the katrina crisis: a global and international perspective with implications for future crisis management. public administration review, 67, 149–159. doi:10.1111/j.1540-6210.2007.00824.x feiock, r. c. (2004). metropolitan governance: conflict, competition, and cooperation. washington, dc: georgetown university press. feiock, r. c. (2007). rational choice and regional governance. journal of urban affairs, 29(1), 47–63. doi:10.1111/j.1467-9906.2007.00322.x feiock, r. c. (2013). the institutional collective action framework. policy studies journal, 41(3), 397–425. doi:10.1111/psj.12023 feiock, r. c., & scholz, j. t. (2010). self-organizing federalism: collaborative mechanisms to mitigate institutional collective action dilemmas. new york, ny: cambridge university press. frederickson, g. h. (1999). the repositioning of american public administration. political science and politics, 32(4), 701–711. gall, m., & cutter, s. l. (2012). 2005 events and outcomes: hurricane katrina and beyond. in c. b. rubin (ed.), emergency management: the american experience 1900–2005 (2nd ed., pp. 191–213). boca raton, fl: crc press. goldsmith, s., & eggers, w. d. (2004). governing by network: the new shape of the public sector. washington, dc: brookings institution press. goodreau, s. m. (2007). advances in exponential random graph (p*) models applied to large social networks. social networks, 29(2), 231–248. goodreau, s. m., kitts, j. a., & morris, m. (2009). birds of a feather, or friend of a friend? using exponential random graph models to investigate adolescent social networks. demography, 46(1), 103–125. gulati, r., & gargiulo, m. (1999). where do interorganizational networks come from? american journal of sociology, 104(5), 1439–1493. doi:10.1086/210179 55976-14-10.indd 39 18/08/15 5:31 pm 40 m. d. siciliano and c. wukich / network features and processes as determinants handcock, m. s. (2003). assessing degeneracy in statistical models of social networks. (working paper no. 39). manuscript in preparation. retrieved from center for statistics and social sciences, university of washington website: http://www.csss.washington.edu/papers/wp39.pdf handcock, m. s., hunter, d. r., butts, c. t., goodreau, s. m., & morris, m. (2008). statnet: software tools for the representation, visualization, analysis and simulation of network data. journal of statistical software, 24(1), 1–11. henry, a. d., lubell, m., & mccoy, m. (2011). belief systems and social capital as drivers of policy network structure: the case of california regional planning. journal of public administration research and theory, 21(3), 419–444. doi:10.1093/jopart/muq042 hunter, d. r. (2007). curved exponential family models for social networks. social networks, 29(2), 216–230. hunter, d. r., goodreau, s. m., & handcock, m. s. (2008). goodness of fit of social network models. american statistical association, 103(481), 248–258. hunter, d. r., & handcock, m. s. (2006). inference in curved exponential family models for networks. journal of computational and graphical statistics, 15(3), 565–583. hunter, d. r., handcock, m. s., butts, c. t., goodreau, s. m., & morris, m. (2008). ergm: a package to fit, simulate and diagnose exponential-family models for networks. journal of statistical software, 24(3), 1–29. kapucu, n. (2006). interagency communication networks during emergencies: boundary spanners in multiagency coordination. the american review of public administration, 36(2), 207–225. doi:10.1177/ 0275074005280605 kapucu, n., augustin, m.-e., & garayev, v. (2009). interstate partnerships in emergency management: emergency management assistance compact in response to catastrophic disasters. public administration review, 69(2), 297–313. doi:10.1111/j.1540-6210.2008.01975.x kapucu, n., & demiroz, f. (2011). measuring performance for collaborative public management using network analysis methods and tools. public performance & management review, 34(4), 549–579. doi:10.2753/pmr1530-9576340406 kapucu, n., & garayev, v. (2011). collaborative decision-making in emergency and disaster management. international journal of public administration, 34(6), 366–375. doi:10.1080/01900692.2011.561477 kapucu, n., hu, q., & khosa, s. (2014). the state of network research in public administration. administration & society. advance online publication. doi:10.1177/0095399714555752 kiefer, j. j., & montjoy, r. s. (2006). incrementalism before the storm: network performance for the evacuation of new orleans. public administration review, 66, 122–130. doi:10.1111/j.1540-6210.2006.00672.x koskinen, j., & daraganova, g. (2013). exponential random graph model fundementals. in d. lusher, j. koskinen, & g. robins (eds.), exponential random graph models for social networks: theory, methods, and applications (pp. 49–76). new york, ny: cambridge university press. koskinen, j., & snijders, t. a. b. (2013). simulation, estimation, and goodness of fit. in d. lusher, j. koskinen, & g. robins (eds.), exponential random graph models for social networks: theory, methods, and applications (pp. 141–166). new york, ny: cambridge university press. krackhardt, d. (1994). constraints on the interactive organization as an ideal type. in c. heckscher & a. donnellon (eds.), the post-bureaucratic organization: new perspectives on organizational change (pp. 211–222). thousand oaks, ca: sage. kwon, s.-w., feiock, r. c., & bae, j. (2014). the roles of regional organizations for interlocal resource exchange: complement or substitute? the american review of public administration, 44(3), 339–357. doi:10.1177/0275074012465488 lazarsfeld, p. f., & merton, r. k. (1954). friendship as a social process: a substantive and methodological analysis. in m. berger (ed.), freedom and control in modern society. new york, ny: van nostrand. lee, y., lee, i. w., & feiock, r. c. (2012). interorganizational collaboration networks in economic development policy: an exponential random graph model analysis. policy studies journal, 40(3), 547–573. doi:10.1111/j.1541-0072.2012.00464.x lubell, m., scholz, j., berardo, r., & robins, g. (2012). testing policy theory with statistical models of networks. policy studies journal, 40(3), 351–374. doi:10.1111/j.1541-0072.2012.00457.x 55976-14-10.indd 40 18/08/15 5:31 pm m. d. siciliano and c. wukich / network features and processes as determinants 41 lusher, d., koskinen, j., & robins, g. (2013). exponential random graph models for social networks: theory, methods, and applications. new york, ny: cambridge university press. owen-smith, j., & powell, w. w. (2004). knowledge networks as channels and conduits: the effects of spillovers in the boston biotechnology community. organization science, 15(1), 5–21. doi:10.1287/orsc.1030.0054 park, h. h., & rethemeyer, r. k. (2014). the politics of connections: assessing the determinants of social structure in policy networks. journal of public administration research and theory. 24(2), 349–379. doi:10.1093/jopart/mus021 post, s. s. (2004). metropolitan area governance and institutional collective action. in r. c. feiock (ed.), metropolitan governance: conflict, competition, and cooperation (pp. 67–94). washington, dc: georgetown university press. provan, k. g., & milward, b. h. (1995). a preliminary theory of interorganizational network effectiveness: a comparative study of four community mental health systems. administrative science quarterly, 40(1), 1–33. r core team. (2013). r: a language and environment for statistical computing. vienna, austria: r foundation for statistical computing. reagans, r. (2011). close encounters: analyzing how social similarity and propinquity contribute to strong network connections. organization science, 22(4), 835–849, doi:10.1287/orsc.1100.0587 robins, g. (2011). exponential random graph models for social networks. in j. scott & p. carrington (eds.), the sage handbook of social network analysis. thousand oaks, ca: sage. robins, g., lewis, j. m., & wang, p. (2012). statistical network analysis for analyzing policy networks. policy studies journal, 40(3), 375–401. doi:10.1111/j.1541-0072.2012.00458.x robins, g., pattison, p., kalish, y., & lusher, d. (2007). an introduction to exponential random graph (p*) models for social networks. social networks, 29(2), 173–191. robins, g., snijders, t., wang, p., handcock, m., & pattison, p. (2007). recent developments in exponential random graph (p*) models for social networks. social networks, 29(2), 192–215. robinson, s. e., eller, w. s., gall, m., & gerber, b. j. (2013). the core and periphery of emergency management networks. public management review, 15(3), 344–362. doi:10.1080/14719037.2013.769849 romzek, b. s., leroux, k., & blackmar, j. m. (2012). a preliminary theory of informal accountability among network organizational actors. public administration review, 72(3), 442–453. doi:10.1111/j.1540-6210.2011.02547.x snijders, t. a. b. (2002). markov chain monte carlo estimation of exponential random graph models. journal of social structure, 3(2), 1–40. snijders, t. a. b. (2011). statistical models for social networks. annual review of sociology, 37(1), 131–153. doi:10.1146/annurev.soc.012809.102709 snijders, t. a. b., pattison, p. e., robins, g. l., & handcock, m. s. (2006). new specifications for exponential random graph models. sociological methodology, 36(1), 99–153. taleb, n. n. (2010). the black swan: the impact of the highly improbable fragility. new york, ny: random house llc. thurmaier, k., & wood, c. (2002). interlocal agreements as overlapping social networks: picket–fence regionalism in metropolitan kansas city. public administration review, 62(5), 585–598. doi:10.1111/1540-6210.00239 tierney, k. j., lindell, m. k., & perry, r. w. (2001). facing the unexpected: disaster preparedness and response in the united states. washington, dc: joseph henry press. u.s. house of representatives. (2006). a failure of initiative: final report of the select bipartisan committee to investigate the preparation for and response to hurricane katrina. washington, dc: us government printing office. wasserman, s., & faust, k. (1994). social network analysis: methods and applications. new york, ny: cambridge university press. waugh, w. l. (2007). emac, katrina, and the governors of louisiana and mississippi. public administration review, 67, 107–113. doi:10.1111/j.1540-6210.2007.00819.x waugh, w. l., & streib, g. (2006). collaboration and leadership for effective emergency management. public administration review, 66, 131–140. doi:10.1111/j.1540-6210.2006.00673.x 55976-14-10.indd 41 18/08/15 5:31 pm 42 m. d. siciliano and c. wukich / network features and processes as determinants weber, e. p., & khademian, a. m. (2008). wicked problems, knowledge challenges, and collaborative capacity builders in network settings. public administration review, 68(2), 334–349. doi:10.1111/j.1540-6210.2007.00866.x wukich, c., & robinson, s. e. (2013). leadership strategies at the meso level of emergency management in networks. international review of public administration, 18(1), 41–59. zaccarin, s., & rivellini, g. (2010). modeling network data: an introduction to exponential random graph models. in f. palumbo, c. n. lauro, & m. j. greenacre (eds.), data analysis and classification (pp. 297–305). berlin, germany: springer. appendix: assessing goodness of fit as noted in the text, all of the models successfully converged. for the final model, we assessed the difference between the sample statistics based on simulated networks (i.e., those generated from the parameter estimates in the final model) to the actual values in the observed network. for each of the terms included in the model, no significant difference was found between the observed value and the sample statistic of that term produced from the simulated networks. once we were satisfied with the final model’s convergence, we examined whether the model could reproduce “out-of-model statistics.” these out-of-model statistics are part of the goodness of fit heuristics described by hunter, goodreau, and handcock (2008). the goodness of fit measures for statistical network models assess the ability of a fitted model to reproduce certain network properties that were not specifically modeled, thus the term “out-of-model” statistics. if the out-of-model or nonfitted statistics are well captured in the simulated distribution of graphs, then there is evidence to indicate that these features may have arisen from the processes included in the model (koskinen & snijders, 2013). the most common out-of-model statistics used to assess goodness of fit are shared partners, degree, and average geodesic distance. the results are shown in figure 2. in each of these plots, the thick black line represents the observed value of a given statistic in the katrina response system. the boxplots are derived from the simulated networks. for the boxplots, 100 networks were simulated based on the parameter estimates in the final model. the relevant network statistic was then calculated for each simulated network, and the distribution of the statistic across the simulations is charted. the first goodness of fit plot reveals the proportion of edges with a given number of shared partners. the plot suggests that our model captured well the shared partner distribution observed in the katrina dataset. while the gwesp term used in the model is assured of capturing the mean number of shared partners, it does not control for the full distribution of shared partners (goodreau et al., 2009). thus, this global property in the network is being adequately captured by the terms included in our final model. the second plot shows the degree distribution. as with gwesp, the gwd term and the main effect terms do not model the full degree distribution. the goodness of fit plot indicates one area where our model does not adequately represent the proportion of nodes with a given degree. specifically, the model underestimates the proportion of nodes with a degree of 1. this indicates that there are more pendants in the observed network than 55976-14-10.indd 42 18/08/15 5:31 pm m. d. siciliano and c. wukich / network features and processes as determinants 43 present in the simulated networks. while the overall shape of the distribution is captured by the model, instances of underor over-estimation represent areas for future research (goodreau et al., 2009) and can indicate the need to explore additional model terms. the final plot reveals the geodesic distances between organizations. these distances represent the number of links that exist between two organizations and is not directly related to any of the terms included in the final model. due to the fragmentation of the network, the plot indicates that the largest proportion of dyads were not reachable (nr). for those that are reachable, the distribution is characterized by a small spike around geodesics of 4, 5, and 6. overall, our final model captures the distribution of distances fairly well, as the observed line falls within each of the boxplots. taken as a whole, we feel our model adequately captures the three out-of-model statistics, and thus provide evidence to indicate that these features could have potentially arisen from the processes included in the model (koskinen & snijders, 2013). figure 2. goodness of fit plots for the final model. 55976-14-10.indd 43 18/08/15 5:31 pm 55976-14-10.indd 44 18/08/15 5:31 pm adaptive governance of coupled social-ecological systems: introduction to the special issue themes complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 1-4 doi: http://dx.doi.org/10.20377/cgn-92 1 university of bamberg press adaptive governance of coupled social-ecological systems: introduction to the special issue themes author: asim zia department of community development and applied economics, university of vermont, u.s.a. e-mail: asim.zia@uvm.edu stockholm resilience alliance (sra) defines adaptive governance as an evolving research framework for analyzing the social, institutional, economical, and ecological foundations of multilevel governance modes that are successful in building resilience for the vast challenges posed by multiscale drivers of change, such as global climate change, rapid technological change, terrorism, socio-economic disruptions, and political coups. the social-ecological systems (ses) framework is an advanced version of elinor ostrom’s (1990, 2005) institutional analysis and development framework. folk and his colleagues (2005) laid out the theoretical foundations for a deeper study of adaptive governance of social-ecological systems. they argue that adaptive governance systems often self-organize as social networks with teams and actor groups that draw on various knowledge systems and experiences for the development of a common understanding and policies, in particular during periods of abrupt change (crisis) in social-ecological systems. folk and his colleagues conclude that “the emergence of bridging organizations seem to lower the costs of collaboration and conflict resolution, and enabling legislation and governmental policies can support self-organization while framing creativity for adaptive co-management efforts” (p. 41). in this context, the sra group laid out two grand challenges for the study of adaptive governance: 1. what are the important multiscale processes in social-ecological systems governance that lead to more or less resilient outcomes on the ground? 2. what are the tradeoffs between management priorities and social-ecological systems for long-term sustainable futures and how do these play out over different scales? some national and international agencies, such as the united nations university institute for the advanced study of sustainability (unu-ias), and a group of scholars are increasingly interested in finding answers to these questions and ask for more applied research on adaptive governance. unu-ias considers gunderson and holling (2002); dietz et al. (2003); folk et al. (2005) and olsson et al. (2006) as key theoretical contributions to establishing adaptive governance as a research field. drawing on five case studies from the american west, brunner and lynch (2006), explored how to expedite a transition toward adaptive governance and break the deadlock in natural resource policymaking. brunner and lynch argue that adaptive governance integrates various types of knowledge and organizations and it relies on open decision-making processes recognizing multiple interests, community-based initiatives, and an integrative science, in addition to traditional science. scholz and stiftel (2005) apply the adaptive governance framework to study water governance issues across multiple spatial and temporal scales. as theoretical and empirical contributions to understanding adaptive governance are growing in volume under multiple research initiatives, this special issue of complexity, governance & networks also aimed to focus on the study of adaptive governance through the parallel theoretical frameworks that have emerged in the fields of public policy, public management and public http://dx.doi.org/ mailto:asim.zia@uvm.edu complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 1-4 doi: http://dx.doi.org/10.20377/cgn-92 2 university of bamberg press administration to study collaborative and network governance approaches in the face of complex public policy problems. the broader shift from government to governance and the hollowing out of the government in this age of contracting out public services to third party vendors further necessitates the need to study the evolving and dynamic nature of governance networks from a complex systems perspective (koliba et al. 2010; zia et al. 2014). a broad call for papers was issued inviting submissions covering the following topics and approaches: • theoretical perspectives on adaptive governance, including but not limited to complexity sciences, complex adaptive systems, governance networks, network governance, collaborative governance, and/or multi-level collaborative governance. • methodological applications to understand and model adaptive governance of social ecological systems, such as governance of food, energy & water systems, governance of natural resources, governance of socio-technical ecological systems (e.g. smart sustainable cities) • social and/or policy learning derived from different governance design experiments • longitudinal and/or panel studies across different geographical and administrative regions the special issue includes five papers that span the broad theoretical and methodological topics included in the call for papers. the first paper by hammond-wagner (2019) advances ostrom’s ses framework by digging deeper into the linkages between institutions and human behaviors. hammond-wagner (2019) argues that the ses framework has struggled to facilitate analysis of environmental challenges beyond common-pool resource (cpr) regimes and the emergence of community-based governance institutions. hammond-wagner (2019) examines attributes of environmental public goods dilemmas that differentiate them from cpr regimes. these include the lack of a behavior-reinforcing link, multi-actor and multi-resource system dynamics, higher levels of uncertainty and complexity, and lack of built-in social capital. in the second paper, webster and pavlovich (2019) present a novel agent based model (abm) that simulates institutional and human behavioral linkages in a coupled natural and human systems framework. webster and pavlovich (2019) argue that decision makers tend to respond to problems rather than prevent them. in political science, this process of responsive governance is associated with complex dynamics such as availability cascades and punctuated equilibrium. however, most authors treat problems as one-time events, like oil spills or political scandals. in the second paper, webster and pavlovich (2019) present an agent based model loosely based on the lake erie watershed to explore how responsive governance evolves along with an on-going but noisy environmental problem: harmful microbial blooms. their conceptual model features a two-level decision process based on jones and baumgartner (2005). meta-agents representing the individual level of analysis “perceive” blooms either directly via observation if they are near a bloom or indirectly through the media. as a meta-agent observes more blooms, their concern increases until it crosses an action threshold, at which point they use simple cost-benefit analysis to select from a range of options. webster and pavlovich (2019) examine two major scenarios, one in which there is a single policy maker managing the entire region (e.g. the national government) and one where there are 5 policy makers, each separately regulating a demographically and geographically distinct region. webster and pavlovich (2019) use the abm generated simulation experiments to explore how variability in risk perception and responsive governance shape the functioning of the entire coupled human and natural system, including biophysical feedbacks. novel theoretical concept of governance treadmill -that captures systemic fluctuation between effective and ineffective governance responses -is introduced and operationalized in the abm. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 1-4 doi: http://dx.doi.org/10.20377/cgn-92 3 university of bamberg press spett (2019), in the third paper, addresses the transboundary management of water bodies that cross political, cultural, and ecological boundaries. the transboundary governance issues entail working with a level of complexity that requires creative, adaptive management strategies to build resilience and increase social capacity in the face of social ecological disturbances. to asses the extent to which such complexity can be managed, spett (2019) explores the application of the socialecological systems framework, proposed by walker and salt (2012), for assessing and managing resilience. elements of this framework are applied on the transboundary lake champlain richelieu river basin, which is a freshwater basin that exists between the united states and canada. mumtaz and ali (2019), in the fourth paper, employ the institutional analysis and development (iad) framework (ostrom 2005) to understand the adaptive polycentric governance challenges emanating from risks posed by climate change in two provinces (sub-national entities) of pakistan. after the 18th constitutional amendment in 2010, the responsibility of implementing climate change policies rests with respective provinces/subnational governments in pakistan. mumtaz and ali (2019) argue that, to manage the complexity of sess, the field of environmental governance emerged as a means of understanding sustainable natural resource use patterns. they further argue that one of the influential approaches to reconciling social and ecological aspects of environmental governance emerged in the form of adaptive governance to deal with uncertainty and complexity of sess. for mumtaz and ali (2019), the adaptive governance literature is an emergent form of environmental governance research that demands to coordinate various dynamic forms of resource management regimes and iad framework is particularly well suited to confront the complexity and uncertainty associated with rapid environmental change. finally, in the fifth paper of this special issue, panikkar et al. (2019) examine the participatory social learning process of localized non-state actors –community and scientific stakeholders – and their efforts at adaptive co-management of the kabul river basin between afghanistan and pakistan, one of the most conflict ridden areas in the world, to enhance trust building, information sharing, collaboration, and capacity building across the basin. panikkar et al. (2019) present action research based approach and deploy novel track-2 and track-3 environmental diplomacy framework to seed ecological cooperation and environmental peacebuilding. for panikkar et al. (2019), the lack of trust among upstream and downstream riparian partners and persistent failures of track 1 diplomacy initiatives has led to an absence of governance mechanisms for mitigating the water security concerns in the kabul river basin. their research shows that science and public diplomacy, democratic participation, and social learning may pave a way to clear local misconceptions, improve transboundary water cooperation, and increase ecological stewardship in the kabul river basin. water cooperation and management is thus a political process which is an emergent phenomena in complex adaptive systems (pahl-wostl, 2009). overall, these five papers present inter-twined themes of adaptive governance in social ecological systems to cope with uncertainty and complexity, risk and insecurity, distrust and competition. while hammond-wagner (2019) and mumtaz and ali (2019) advance ostrom’s polycentric governance theory to tackle complex ses challenges, webster and pavlovich (2019) present a novel theoretical approach of treadmill governance to simulate the emergence of bottlenecks and cooperation in complex non-point source water pollution management contexts. two other papers apply very different theoretical frameworks to analyze transboundary governance issues in two different continents of the world. while spett (2019) applies the well-established adaptive governance based ses framework emanating from resilience alliance in us-canada context, panikkar et al. (2019) break new ground by pitching an action-oriented, and dare i say, a novel reflexive governance approach to bring about ecological cooperation and resolve persistent conflict across afghanistan and pakistan in hindukush mountains. ultimately, i hope that this special issue will ignite new avenues of research and practice to address complex challenges http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 1-4 doi: http://dx.doi.org/10.20377/cgn-92 4 university of bamberg press pertaining to the governance of ses, and enable an interdisciplinary dialogue towards convergence of the emerging theories of polycentric governance, treadmill governance, adaptive governance and reflexive governance of ses. references bizikova et al. (2013) the water-energy-food security nexus: towards a practical planning and decision-support framework for landscape investment and risk management. international institute for sustainable development brunner, ronald d., toddi a. steelman, lindy coe-juell, christina cromley, and christine edwards. adaptive governance: integrating science, policy, and decision making. columbia university press, 2005. brunner, ronald d., and amanda h. lynch. adaptive governance and climate change. boston: american meteorological society, 2010. dietz, thomas., elinor ostrom, paul c. stern (2003). "the struggle to govern the commons." science. 302(5652):19071912. folke, carl., thomas hahn, per olsson, jon norberg (2005). "adaptive governance of social-ecological systems." annual review of environmental resources 30: 411-73. gunderson, l. h., cs holling (ed) (2002). panarchy: understanding transformations in human and natural systems. washington: island press. hammond wagner, courtney (2019). strengthening the institution-behaviour link in the ses framework to faciliate analysis of environmental public goods dilemmas. bamberg: university of bamberg press, complexity, governance & networks – vol. 5, no 1 (2019) special issue, 5-23. jones, b. d., & baumgartner, f. r. (2005). the politics of attention: how government prioritizes problems. chicago: university of chicago press. koliba, c., meek, j., and zia, a (2010) governance networks in public administration and public policy. crc press, boca raton fl. mumtaz, muhammad, & salem h., ali (2019). adaptive governance and sub-national climate change policy: a comparative analysis of khyber pukhtunkhawa and punjab provinces in pakistan. bamberg: university of bamberg press, complexity, governance & networks – vol. 5, no 1 (2019) special issue, 81-100. olsson, pier., lance h gunderson, steve r. carpenter, paul ryan, louis lebel, carl folke, cs holling (2006). "shooting the rapids: navigating transitions to adaptive governance of socio-ecological systems." ecology and society 11(1): 1-18 [online]. ostrom e. 1990: governing the commons. cambridge: cambridge university press. ostrom, e., 2005. understanding institutional diversity (vol. 241). princeton, nj: princeton university press. pahl-wostl, c. (2009). a conceptual framework for analyzing adaptive capacity and multi-level learning processes in resource governance regimes. global environmental change, 19(3), 354–365. pannikar, bindu et al. (2019). transboundary water governance in the kabul river basin: implementing environmental and public diplomacy between pakistan and afghanistan. bamberg: university of bamberg press, complexity, governance & networks – vol. 5, no 1 (2019) special issue, 101-120. spett, emma (2019). building flood resilience in trans-boundary social-ecological systems: adaptive governance in the lake champlain richelieu river basin. bamberg: university of bamberg press, complexity, governance & networks – vol. 5, no 1 (2019) special issue, 65-80. walker, b., & salt, d. (2012). resilience practice: building capacity to absorb disturbance and maintain function. island press. webster, d.g. & pavolvich, tyler (2019). responsive governance and harmful microbial blooms on lake erie: an abm approach. bamberg: university of bamberg press, complexity, governance & networks – vol. 5, no 1 (2019) special issue, 24-64. zia, a., kauffman, s., koliba, c., beckage, b., vattay, g. and bomblies, a., 2014. from the habit of control to institutional enablement: re-envisioning the governance of social-ecological systems from the perspective of complexity sciences. complexity, governance & networks, 1(1), pp.79-88. http://dx.doi.org/ adaptive governance of coupled social-ecological systems: introduction to the special issue themes teaching complexity: the limits of evidence and the ‘prospective’ case study complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 92-102 doi: http://dx.doi.org/10.20377/cgn-48 92 university of bamberg press teaching complexity: the limits of evidence and the ‘prospective’ case study amanda wolf abstract: policy practitioner–students in a master of public policy programme in new zealand describe many problems and processes in their work environments as ‘complex’. yet, they hold firmly to a belief in the merits of ‘evidence’ to guide their advice and decision making in the face of that complexity. this article examines the aims and pedagogy of a two-course sequence designed to help students replace over-reliance on analysing existing evidence with understanding of the ways complexity concepts can aid in estimating possible outcomes of policy interventions. starting with identifying evidence challenges, students learn how to compare a status quo situation and a prospective case of that status quo in which a new policy has been implemented. this method draws on existing scholarship in lesson-drawing for policy applications. students are eased into an appreciation of a variety of complexity frameworks and concepts by looking at a case about which there is, strictly speaking, no evidence. keywords: case study; evidence; lesson-drawing; new zealand; public policy; teaching complexity introduction students in victoria university of wellington’s school of government master’s programme in public policy and public management are primarily working professionals who return to university for study. the programme brings students of both policy and management together for a greater part of their learning, recognising the interdependence and integration of the two fields in practice. students in both degrees share three core courses (out of 12 required for the degree) that cover the overall context, institutions and dynamics of government and governing, national and international societal and economic forces that influence the essential features and interdependence of management and policy, and theories and practices in policy and management. the common core courses also recognise that practitioners need to be able to critically reflect on their specific practices and to have high-level intellectual capability to be both critical and creative in their work. as designed, the programme ensures that classrooms have a great diversity of learners. the courses are interdisciplinary and problem-focused, drawing on selected examples and—importantly—ongoing challenges in students’ workplaces. the explicit intentions are to increase students’ understanding of the relevance of many disciplines to their work, and to sensitise them to the challenges and expectations on them as professionals in this environment. much of the learning in these courses is relevant for advancing students’ understanding of the complexities in policy development and implementation, but ‘complexity’ is not directly addressed as a theme. nevertheless, from the start, students learn in an environment in which they are assisted to look at their practice, in primarily policy or management, openly and reflectively. amanda wolf, school of government, victoria university of wellington, wellington, new zealand, e-mail: amanda.wolf@vuw.ac.nz http://dx.doi.org/10.20377/cgn-48 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 92-102 doi: http://dx.doi.org/10.20377/cgn-48 93 university of bamberg press in addition to the common core courses, students pursue a four-course core (comprising one-third of the degree) specific to management or policy, and complete their degree with shared electives and independent research. students in the master of public policy stream study a twocourse sequence, which i teach with contributions from colleagues. teaching complexity in these two courses is the focus of this article, with a particular emphasis on using the device of a ‘prospective case’. in the next section, the policy cohort and their initial orientation to study are more fully described. following this, my approach to teaching complexity is described, with reference to the course learning objectives, content and pedagogy, including attention to the rationale for using a prospective-case method. a final section presents some reflections on the methods used and their merits in teaching complexity to a cohort of experienced policy professionals. master of public policy students’ initial orientations to ‘evidence’ and ‘complexity’ in their workplaces, policy students are exposed to strong rhetoric and prescriptions concerning the supremacy of ‘evidence’ for developing and implementing policy. consequently, they are intensely distrustful of any information that looks ‘subjective’ or ‘biased’. they applaud claims such as one in a report from the us national research council: “[w]hen the question on the table is what are the ‘real’ conditions or what will ‘probably’ happen if we implement one policy instead of another, science is on balance a more dependable and defensible guide than informed hunches, analogies, or personal experience” (prewitt, schwandt, & straf, 2012, p. 10). closer, to home, they may be employed by a departmental chief executive, who proclaims: “we have to make sure what looks like a good policy idea is backed up by solid evidence and quality analysis”. in workplace discourse, it is equally broadly claimed that policy formation and implementation is ‘complex’. usually this refers to the presence of uncertainty and value clashes, with many inputs and actors at the interface between policy analysis and politics. and while there may be talk of ‘interdependencies’ or ‘unpredictability’—terms that are used in some complexity literature—the typical response is to seek more or better evidence, as the then head of australia’s productivity commission urged: without evidence, policy makers must fall back on intuition, ideology, or conventional wisdom—or, at best, theory alone... . but the resulting policies can go seriously astray, given the complexities and interdependencies in our society and economy, and the unpredictability of people’s reactions to change... . policies that haven’t been informed by good evidence and analysis fall more easily prey to the ‘law of unintended consequences’ which can lead to costly mistakes. (banks, 2009, pp. 4–5) thus, the way many students appreciate their work environment creates a significant block to introducing a deeper understanding of ‘complexity’. accordingly, i adopt a wellestablished pedagogical practice, which holds that for students to learn a new concept or skill, the teacher must first activate some doubt or discomfort with the students’ pre-existing beliefs, then present new material, provide examples, and offer students an opportunity to practice applying the material. over the two courses, i aim to help students enlarge their skills and to understand that (a) there are significant limits to using evidence (in part since it is by definition about the past); (b) they already compensate for these limits using various natural, judgemental (and hence, subjective and biased!) processes that help them to see a little bit into the future; but (c) they don’t have to give up entirely on the rigour and reliability promised by evidence; since (d) skills for working with complexity can complement what they gain from conventional evidence analysis. http://dx.doi.org/10.20377/cgn-48 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 92-102 doi: http://dx.doi.org/10.20377/cgn-48 94 university of bamberg press learning objectives and a stepped introduction to complexity most students undertake study in economics and an introduction to policy analysis before enrolling in the remaining two courses in the policy core. the course prescriptions for the two subsequent courses refer directly to complexity: policy methods and practice: qualitative and quantitative techniques for collecting, analysing, interpreting and applying information and evidence to advance policy objectives particularly under conditions of complexity and uncertainty and in light of given task requirements. policy workshop: integrating theories and practices to produce in-depth studies of selected complex policy issues with a focus on the design and evaluation of alternative policy options. in the policy methods course, the specific learning objectives are framed in a way that accepts that learners (and often their employers) are interested in gaining ‘conventional’ evidence skills, such as determining the specific evidence requirements for a given policy research or analysis task; selecting or adapting methods and practices for various challenges; demonstrating critical awareness of the strengths and limitations entailed in applications; and judging the adequacy of evidence. consideration of complexity, however, begins in the first session. the first exercise is a round-robin to elicit qualities of a ‘good policy professional’, designed to activate some scepticism about the reach of ‘evidence’ in practice. as expected, everyone knows some analytical whiz, and first-round propositions are heavily weighted toward analytic skills, such as the ability to define the problem, to find and assess information or to conduct cost–benefit analyses. soon come ideas about relationships in the environment, such as dealing with political constraints, attending to international norms, and communicating with different demographic groups or other professionals. eventually, a few students venture some qualities consistent with managing complexity and uncertainty. for example, they mention making sense of a messy situation, or intuiting the inclinations and likely responses of a politician and other stakeholders. often, a student will also volunteer characteristics that seem more directly to acknowledge complexity, such as the ability to anticipate or be alert to ‘unintended consequences’ or to the unpredictability of people’s behaviour. however, in offering these suggestions, students may be looking through the prism of an ‘evidence’ or analytic frame, consistent with the banks quotation above. by the end of the round-robin, students are more attuned to some tension in their use of ‘evidence’ and ready to consider my initial definition of complexity: as a condition in which the knowledge required to bring about a desired change is not readily mustered by even the most sophisticated analysis of evidence, but is rather highly distributed in the communities, organizations and individuals where change is to take place and is unknowable at the point when policy decisions are being made. although students arrive wanting to become better evidence-handlers, the exercise forces a small wedge into the door to complexity learning. students realise for themselves that there is more to their practice than developing evidence products, such as analyses and ‘objective’ advice. with the scene thus established, student examine more carefully the assumption of the ‘science’ framing of evidence, drawing on arguments for the variety of information that aids decision makers (head, 2008), the connection between complexity and the tradition of american pragmatism (sanderson, 2009) and the distinction between information and evidence, as described by majone (1989, p, 7): evidence is “information selected from the available stock and introduced at a specific point in the argument in order to persuade a particular audience of the truth or falsity of a statement”. students consider that for many policy challenges, there are unfillable gaps in the evidence base. in their first written assignment, students map out some of the ‘evidence challenges’ in a relatively discrete topic of their choosing, taking into account value differences, and various contextual features. http://dx.doi.org/10.20377/cgn-48 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 92-102 doi: http://dx.doi.org/10.20377/cgn-48 95 university of bamberg press hard on the heels of these efforts to call evidence into question, the class considers definitions of uncertainty, ambiguity and complexity and is introduced to methods and practices to address them. i present three ‘complexity frameworks’ (see table 1), each of which takes a different tack: eppel’s lens (2012), developed from the complexity literature, pawson’s (2012) suggestions for ‘mapping’ a pending policy intervention, and room’s (2011) advice to an ‘agile’ policy professional. i illustrate aspects of these frameworks using a published case study, such as smith and lawrence (2014) on food security and resilience in rockhampton, australia, following flooding that cut off a community’s access to food. such case studies are ubiquitous in the published literature. true to form, such articles have some theoretical point to advance, and it is this that makes them generally seem ‘irrelevant’ or ‘overly academic’ to our cohort of students. yet, almost all such articles contain detailed descriptions of a case, which can be read on their own, and are generally suited to get the reader’s complexity thinking going, with complexity frameworks providing direction and structure: what systems interacted? how did prior changes in the regulatory system change the local response to the drought? what information could have been known about the risks to the supply chain but was available only in retrospect? finally, the students read mary schmidt’s (1993) story of the failure of the teton dam to further sensitise them to the alternative forms of knowledge potentially available to for decision-making, but which are generally dismissed from the ‘evidence’ base. students are allocated one ‘theme’ from table 1 or from schmidt’s types of knowledge, which they consider with respect to the policy topic they are investigating. class presentations ensure exposure to varied applications. the rest of the policy methods course covers evidencehandling skills through units on various methods of collecting, analysing and interpreting evidence and assessing its qualities in specific applications. this treatment reinforces the merits of good analysis, which is not obviated by complexity. learning complexity through prospective case analysis in the policy workshop, students learn to draw and apply transferable lessons from an exemplar (wolf & baehler, 2017). they then undertake a realistic, if somewhat attenuated project that demonstrates their ability to apply these methods, as well as other learning from throughout the policy core. building on the policy methods course (which activates doubt about the full rhetorical impact of the ‘evidence-based policy’ mantra, awakens recognition of some natural aptitude for complexity thinking, and offers exposure to numerous examples of how aspects of complexity may influence policy analysis and decision making), i continue in the workshop to gradually draw students along a learning pathway in a structured manner. the high-level tasks in the policy workshop start with a status quo ‘situation analysis’ of a current policy challenge, which can be ‘cased’ as a policy status quo, or receiving environment (or ‘target case’ in the policy transfer and learning literature) or simply a, with attention to the stakeholders, problemand solution-framings, key contextual influences and tensions among them. next, students make a detailed study of a promising policy intervention that is absent in a, but in place elsewhere (b). by designing an intervention adapted from b for application in the receiving environment (a), students create a prospective case (a′), which is a at time t + 1, in which policy based on insights spawned by b is producing impacts, for better or worse. such prospective-case thinking aids students in taking a broader perspective—to think about the policy ecosystem, comprising actors and resources and, importantly, other policies, and to imagine the introduction of new decisions. in sum, to estimate the impacts in a′, students take into account— at some level and selectively—aspects of the complexity frameworks in table 1. http://dx.doi.org/10.20377/cgn-48 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 92-102 doi: http://dx.doi.org/10.20377/cgn-48 96 university of bamberg press table 1. complexity frameworks eppel complexity lens pawson victore model room agile policymaking toolkit the system whole – whole cannot be understood as the sum of its parts, or reduced to its parts nested, interacting and interdependent systems – characteristics identified at one level of the system are also present in whole multiple interactive systems – reflexive influence patterns, which arise from the ongoing interaction between actors, can result in negative or positive feedback loops adaptation and co-evolution – changes in the environment may stimulate not only system change, but also external change in response change through self-organisation and emergence – change in systems occurs as a result of changes which affect the microdiversity that emerges as a result of feedback loops, adaptation and emergence open systems and socially constructed boundaries – individuals can plan their own actions, not actions of others or their interplay stability is not equilibrium – systems can suddenly and unpredictably undergo large changes of trajectory history influences starting point for change volitions – participants’ motivations and reasoning; interventions don’t work; people’s interpretations and actions do implementation – many hands and long implementation chains; long program processes (many decision-making points) contexts – multiple, open system, interacting; stakeholders, relationships, institutional settings, norms, values; culture, history, and so on time – historicity; sequences of programs; duration outcomes – short, medium and long term; multiple indicators; multiple interpretation of findings rivalry – multiple policies and programs operating: competing, undermining, reinforcing… emergence – elements interact to create new elements of systems; changes based on changes can go in any direction map the landscape – understand the policy arena’s issues and current challenges identify the protagonists – know the players, stakeholders, relationships in policy arena model the struggle – create scenarios to understand how the landscape may evolve watch for tipping points – identify the triggers that could dramatically shift matters tune the landscape – use analytical tools and discussions to move the policy arena into directions that are more productive energize the protagonists – help some of the protagonists build capacity and take other actions to encourage cooperative behaviour toward win-win situations civilize the struggle – help create win-win situations and limit destructive behaviours watch for predators – keep one or more protagonists from unfairly tipping the balance of power and creating a destructive struggle in the landscape sources: abstracted from eppel (2012), pawson, (2013), and room (2011) http://dx.doi.org/10.20377/cgn-48 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 92-102 doi: http://dx.doi.org/10.20377/cgn-48 97 university of bamberg press in the classroom, time is taken to motivate the reasons as i see them for taking a comparative-case approach and to justify the place of these methods in a complexity-ready toolkit. as writers have made clear (for example, thomas 2010, 2011, rehearsing the arguments of macintyre), for one reason or another, generalisation does not work easily in the social sciences, which is a problem for the application of evidence to new decisions. no two particular instances of a policy are ever similar enough, and the differences always cast doubt on the applicability of the generalisation. while replicating studies in many different locations under a diversity of circumstances can help reduce problems of external validity, the high costs and long timeframes associated with repeating sophisticated studies make this prohibitive most of the time. even with replication, problems of external validity tend to persist because social problems, implemented solutions, and contextual settings vary in so many different ways from case to case. in the workshop, following their situation analyses, students find an exemplar elsewhere, assess it to gain an understanding of that other case, then abstract and adapt learning for application in the presenting case. students read rose (2002) and barzelay (2007), which offer methods showing how practitioners and researchers can use existing policies or management processes as cases for learning, based on the congruence between studied cases and prospective decisions. importantly, these methods are intended for application within a causal-analytic frame, and so they do not entail students making a cognitive alteration for which they are philosophically disinclined. ‘lesson-drawing’ and ‘vicarious learning’ methods allow for the transportability and adaptation of “fresh thoughts” (rose, 2005), based on a causal model of the ‘known’ and the ability of policy learners to “use their minds” (barzelay, 2007) to make an effective transfer. nevertheless, some discussion in the classroom gently critiques these two sources, providing students with a critical awareness of the challenges complexity offers in the search for ‘fresh ideas’ and the judgemental elements of ‘using one’s mind’ in the course of deriving and transferring lessons (wolf & baehler, 2017). these include the manner in which relevant fresh thoughts are generated and detected; the enormous number of variables that differentiate cases in unknown ways; and the fact that lessons are from the past but for the future, all of which complicate using one’s mind in any causal-lesson-drawing activity. while students are encouraged to undertake some analysis (for example to estimate costs of implementing the initiative in the new context based on the experience elsewhere), they have little time for actual analysis, and thus they are (intentionally) thrown into the need to make assumptions and judgemental leaps, to make use of analogical thinking, to fall back on guessing. additional learning exemplars are discussed in class, and guests also present some ‘real’ cases involving policy transfer. for example, the wellington city council community services manager has been invited to explain the city’s homelessness strategy. wellington follows a ‘housing first’ strategy, based on some overseas experiences. this compares with several australian cities that have followed a variant of this strategy, known as ‘common ground’ (but for which there is scant ‘evidence’; see parsell, fitzpatrick, & busch-geertsema, 2014). i look for published case studies on policy issues that are socially embedded (rather than concerning a technical matter). the implementation of stockholm’s congestion charge (a good overview is börjesson, eliasson, hugosson, & brundell-freij, 2013), for example, can be used to think about a prospective case in auckland, which is also searching for ways to address congestion and faces the sort of resistance documented in stockholm. another example (steyaert et al., 2007) is used to demonstrate the manner of ‘judgemental leaping’ from the presented case, which is very specific, to a very different prospective case. the authors narrate an example of community-driven ecological restoration of marshlands along the french atlantic coast. maraîchine cows, a local breed, had lost favour during agricultural intensification because of their low milkand beef-production levels. a farmer fond of local history and marsh heritage began to look for the last maraîchine specimens and to work towards http://dx.doi.org/10.20377/cgn-48 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 92-102 doi: http://dx.doi.org/10.20377/cgn-48 98 university of bamberg press rebuilding the population by identifying individual cows, finding farms to house them, organising breeding and so on. in the process, the farmer brought together agricultural schools and ecological museums, but also some regular landowners and farmers. this network of people shared the objective of preserving the marshland’s biological resources. the lesson from this example relates to preserving species and diversity while also providing for community resilience: the case details (french cows on the atlantic coast) matter less than the somewhat more abstract storyline. making a connection requires someone (roleplayed by students) with an actual community-resilience challenge in mind to generate incipient hypotheses (“x here functions like the farmer there”). as further elaborated below, readers act as if they are engaging in a hypothetical dialogue with a scene that reading helps them to visualise and which in turn allows for the emergence of “pursuit-worthy” ideas (mckaughan, 2008) for application in the home environment. reading such cases without the theory parts (like reading a regression study while skipping the numbers) can inspire fresh ideas (uninfluenced by the author’s theoretical objectives). time is taken to consider what makes a ‘case’ and how a case may be complex. i use an ecosystem metaphor to draw attention to interactive systems and feedback loops, emergence, adaptation and co-evolution. i also reflect that cases tend to be relayed in narrative form, which helps to overcome static and single-perspective framings. i emphasise that in policy, complexity is typically not found in statements of ‘problems’ or ‘solutions’, most of which appear entirely straightforward—such as “children consume too much sugar so we need tax sugar-sweetened beverages”. rather, it is the interactions in the world—among varied understandings of the issues, interpretations of the evidence and values—that make successful policy implementation so challenging, and that moreover, these interactions are not just hard to analyse, they are complex and cannot be known or estimated or modelled well at all. instead of trying to overcome these obstacles with the brute force of analytic thinking, students are invited to try to relax that tendency in order to approach cases simply as stories— tales of events and actors and outcomes in unique contexts, since complexity renders such cases unknowable analytically. yet, a policy or management practitioner, with some relevant experience, can observe and gain insights using a combination of abduction and phronesis. briefly, abduction is a form of inference originally proposed by aristotle, but significantly developed by charles peirce. it happens in a case context, when an anomaly is noticed in a stream of observations. noticing something that stands out leads to one or more plausible explanations (douven, 2017). cases can raise a great number of abductive explanations, so there is a need to gauge their usefulness. analogical reasoning can help. in analogical reasoning, a comparison is made between two domains that highlights similarities and supports a hypothesis pertaining to something as yet unobserved in the second domain, but observed in the first (bartha 2013). finally, phronetic reasoning can supply the grounds for applying knowledge to a new problem, especially when the ability to draw analytical inferences through ‘objective’ scientific reasoning is curtailed (thomas, 2011). again, we can look back to aristotle, who introduced phronetic reasoning, a process that starts in the middle of already knowing, with the knower’s biases (prejudices) essentially opening up to what is next to be understood. thomas (2011, p. 24), drawing on others, describes phronesis as involving a “combination of knowledge, judgement and taste, together producing a discernment, through which we see links, discover patterns, make generalisations, create explanatory propositions . . . all the time emerging out of our experience”. as elaborated by gadamer (see thomas, 2011, p. 21), another case creates in a person an ‘image’ that is tentative, but sufficient to be seen and heard in the context of another’s experience . . . but then used in the context of one’s own: it is essentially visual and dialogic. http://dx.doi.org/10.20377/cgn-48 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 92-102 doi: http://dx.doi.org/10.20377/cgn-48 99 university of bamberg press although travelling to the site of the case of interest is desirable, it can be impractical. students, however, drawing on published documents, evaluations and academic case studies can extract a great deal of the basic story, be it about congestion-relief in stockholm or speciespreservation in france. following gadamer’s observation, the student–practitioner can see and hear another story from their own vantage point. that is, the exemplar enables a person to ‘see’ the prospective case by a process of iterating between what can be assessed directly and that which can only be anticipated or seen in the ‘mind’s eye’. as noted by schön (1983), professionals have a capacity to undertake a “conversation” with a past exemplar, to carry a particular result of that conversation forward as a lesson, and then to apply reflections prospectively to the next assignment. if a learner approaches unique cases, suitably prepared by an initial ‘situation analysis’ and with a view to triggering plausible hypotheses through the process of abduction, resulting hypotheses can then be reflected on in light of their implications for action (via phronesis), rather than their fit with any posited pattern, such as a theory or diagnostic category (thomas 2010). there is, of course, no certainty that the ‘right’ lessons will be learned, but this situation affects all learning. learning outcomes and further opportunities many of the methods used in prospective case analysis can continue to be used after a change is implemented, by iterating a revised prospective case accounting for new observations and hypotheses. after the course, it is hoped that students will be inspired to undertake complexityaware behaviours, such as monitoring for anomalies (room, 2011) in newly changed policy ecosystems, and continuing to mentally form and test hypotheses (sanderson, 2009; eppel, turner, & wolf, 2011). with such an ‘experiment-minded’ perspective, grounded philosophically in pragmatism, a practitioner searches for what does not fit the expected pattern, followed by actions (including ‘thought experiments’) that take into account new perspectives arising from plausible explanations about what is occurring. some indicators that students have learning about complexity emerge in discussion and through signs that there is greater sensitivity to complexity when approaching challenging policy issues, and perhaps somewhat less insistence on ‘science’-oriented information when making decisions looking ahead. after the final presentations in the policy workshop, students individually record tips for others on post-it notes. ideas are volunteered that recognise complexity, ambiguity and uncertainty, as well as the unavoidable contextuality of policymaking. sample tips include: • dealing with ambiguity: sometimes a ‘best guess’ is the best you can do—use own insights and judgements to inform as well as any literature • bear in mind: context, context, context • learn from what is out there • different types of knowledge (e.g., tacit, intuitive) are valid • use evidence and judgement: sometimes all you will have is your ‘best guess’ in sum, i hope, but cannot be sure, that students complete these case exercises with their faith in evidence still intact, but also prepared to reject sharp dichotomies, to value more highly varieties of knowledge, and to welcome the emergence and evolution of their own understanding within the constraints presented by their practice environments. initial support for this desired outcome is shown in an assessment, mentioned earlier, at the end of the policy methods course. students are allocated one ‘theme’ from table 1 or from schmidt’s (1993) types of knowledge. http://dx.doi.org/10.20377/cgn-48 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 92-102 doi: http://dx.doi.org/10.20377/cgn-48 100 university of bamberg press with reference to a policy issue that they have been studying individually, they are asked to describe a challenge posed by the allocated theme and to explain and illustrate why it is a challenge. each student prepares a slide and a three-minute presentation to the class, thus ensuring students learn from a variety of applications. policy workshop cases are assessed through a set of exercises, of which those on assessing context and stakeholders are particularly conducive to applying their complexity learning. in addition, a new assignment has been introduced, called ‘learning record’. amongst other lessons they record, i will be able to scrutinise these assignments for evidence of complexity learning. reflecting on the courses more generally, it is clear that complexity can be approached from a variety of philosophical perspectives, including my pragmatism-influenced one. i am aware, however, that when i teach philosophical perspectives, one of the motivations i offer students is that in a public-sector career they are bound to encounter views from different perspectives. it is part of their professional development to learn the hallmarks and distinctions among them, to know their own, and to be open to information presented from another direction. if more time were available, i would choose to go deeper into some of the philosophical underpinnings that guide various approaches, and how these might inform a comparative prospective case exercise. i have wondered about problematising ‘evidence’ and introducing ‘complexity’—however softly—early in the two-course sequence. would it not be wiser to start simply, for example, with textbook methods of analysis? yet, very little in my practitioner–students’ world is simple. largely for this reason, but also inherent in my understanding of complexity and the future-focus of policy work, i do not rely on prepared teaching cases or the ‘case-method’ as used in some business schools. in a small, diverse environment such as they find themselves, students already have refined abilities to work with complexity. they just don’t so label their practices. if i can pull back the screen, and enthuse them with the potential to gain more traction with some new concepts, the learning objectives are already well-launched. moreover, in a small country context, students (as practitioners) are accustomed to looking elsewhere for policy evidence. formalising the prospective case comparison, with a stream of complexity concepts running through it, appears to augment students’ capabilities without either losing them in a heavily theoretical treatment or taking them philosophically too far from their ‘day-job’ in evidence-handling. references banks, g. (2009). challenges of evidence-based policy-making. australian government productivity commission, speech delivered as the anzsog/anu public lecture series, canberra, feb 4. https://www.pc.gov.au/news-media/speeches/cs20090204/20090204evidence-based-policy.pdf. bartha, p. (2013). analogy and analogical reasoning. in e. n. zalta, (ed.). the stanford encyclopedia of philosophy (fall ed.). http://plato.stanford.edu/archives/fall2013/entries/reasoninganalogy/. barzelay, m. (2007). learning from second-hand experience: methodology for extrapolationoriented case research. governance, 20(3), 521–543. börjesson, m., eliasson, j., hugosson, m., & brundell-freij, k. (2012). the stockholm congestion charges—5 years on. effects, acceptability and lessons learnt. transport policy, 20, 1–12. http://dx.doi.org/10.20377/cgn-48 http://plato.stanford.edu/archives/fall2013/entries/reasoning-analogy/ http://plato.stanford.edu/archives/fall2013/entries/reasoning-analogy/ complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 92-102 doi: http://dx.doi.org/10.20377/cgn-48 101 university of bamberg press douven, i. (2017). abduction. in e. n. zalta (ed.). the stanford encyclopedia of philosophy (summer ed.). http://plato.stanford.edu/archives/sum2017/entries/abduction/. eppel, e. (2012). what does it take to make surprises less surprising? the contribution of complexity theory to anticipation in public management. public management review, 14(7), 881–902. doi:10.1080/14719037.2011.650055. eppel, e., turner, d., & wolf, a. (2011). complex policy implementation: the role of experimentation and learning. in b. ryan & d. gill (eds.). future state: directions for public management in new zealand (pp. 182–212). wellington: victoria university press. head, b. w. (2008). three lenses of evidence-based policy. the australian journal of public administration, 67(1), 1–11. doi: 10.1111/j.1467-8500.2007.00564.x. majone, g. (1989). evidence, argument and persuasion in the policy process. new haven and london: yale university press. mckaughan, d. j. (2008). from ugly duckling to swan: c. s. peirce, abduction, and the pursuit of scientific theories. transactions of the charles s. peirce society, 44(3), 446–468. parsell, c., fitzpatrick, s., & busch-geertsema, v. (2014). common ground in australia: an object lesson in evidence hierarchies and policy transfer. housing studies, 29(1), 69–87. pawson, r. (2013). the science of evaluation: a realist manifesto. london: sage. prewitt, k., schwandt, t. a., & straf, m.l. (eds.). (2012). using science as evidence in public policy. washington: national academies press. room, g. (2011). complexity, institutions and public policy: agile decision-making in a turbulent world. cheltenham, uk; northampton, ma: edward elgar. rose, r. (2002) ten steps in learning lessons from abroad, eui working papers 2002/5. http://cadmus.eui.eu/bitstream/handle/1814/1763/rscas_2002_05b.pdf?sequence=1. rose, r. (2005). learning from comparative public policy: a practical guide. london and new york: routledge. sanderson, i. (2009). intelligent policy making for a complex world: pragmatism, evidence and learning. political studies, 57, 699–719. doi:10.1111/j.1467-9248.2009.00791.x. schmidt, m. r. (1993). grout: alternative kinds of knowledge and why they are ignored. public administration review, 53(6), 525–530. schön, d. (1983). the reflective practitioner: how professionals think in action. new york: basic books. smith, k., & lawrence, g. (2014). flooding and food security: a case study of community resilience in rockhampton. rural society, 23(3), 216–228. doi: 10.1080/10371656.2014.11082066. steyaert, p., barzman, m., billaud, j.-p., brives, h., hubert, b., ollivier, g., & roche, b. (2007). the role of knowledge and research in facilitating social learning among stakeholders in natural resources management in the french atlantic coastal wetlands. environmental science and policy, 10(6), 537–550. thomas, g. (2010). doing case study: abduction not induction, phronesis not theory. qualitative inquiry, 16(7), 575–582. http://dx.doi.org/10.20377/cgn-48 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 92-102 doi: http://dx.doi.org/10.20377/cgn-48 102 university of bamberg press thomas, g. (2011). the case: generalisation, theory and phronesis in case study. oxford review of education, 37(1), 21–35. doi: 10.1080/03054985.2010.521622. wolf, a., & baehler, k. (2017). learning transferable lessons from single cases in comparative policy analysis. journal of comparative policy analysis. doi: 10.1080/13876988.2017.1399578. http://dx.doi.org/10.20377/cgn-48 teaching public administrators and leaders to handle complexity complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 80-91 doi: http://dx.doi.org/10.20377/cgn-54 80 university of bamberg press teaching public administrators and leaders to handle complexity harri raisioa, alisa puustinenb, pirkko vartiainenc, juha lindelld, seija ollilae abstract: understanding and working under complexity has become ‘the new normal’ in public administration. hence complexity must also be integrated into teaching and training of public administrators, not only in higher education but also in in-service training and educating wider society. this can be done by combining the ongoing research agenda into courses and teaching methods. this article describes the integration of complexity thinking and teaching in one university, both by giving examples on the courses and methods applied, student feedback received and by anticipating future developments. finally, practical advice for teachers of complexity is given. keywords: complexity; wicked problems; public administration; curriculum; higher education 1. complexity as an integral part of a public management curriculum since 2010, complexity has been an integral part of teaching and research in social and health management at the university of vaasa, finland. the 240 students in the department are mostly adult learners with a background in nursing or social care, who might be studying for either a bachelor’s or a master’s degree. the department also offers an additional master’s degree program in helsinki, in collaboration with the hospital district of helsinki and uusimaa. the majority of students target a managerial position within the social and health care field, which might be that of chief nursing officer, manager of municipal social and health services, or human resource manager. the department also offers a doctoral program, which currently has more than 20 phd students. in addition, the department provides in-service training. complexity came to the attention of the department initially through the research done on the concept of wicked problems. the first such research articles in the wicked-health-care context were published from 2005 (vartiainen 2005; 2008), and the research gradually expanded to include complexity sciences. the first course on complexity was added to the curriculum in 2009, and the first phd dissertation on the topic was published in 2010 (raisio 2010). as research on the theme increased and involved more researchers, the complexity perspective began to be added to other courses as well. today, complexity is a cross-cutting theme in the department’s curriculum. aharri raisio, university of vaasa, finland, e-mail: harri.raisio@uva.fi balisa puustinen, finnish national defense university, finland, e-mail: alisa.puustinen@mil.fi cpirkko vartiainen, university of vaasa, finland, e-mail: pirkko.vartiainen@uva.fi djuha lindell, university of vaasa, finland, e-mail: juha.lindell@uva.fi eseija ollila, university of vaasa, finland, e-mail: seija.ollila@uva.fi http://dx.doi.org/10.20377/cgn-54 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 80-91 doi: http://dx.doi.org/10.20377/cgn-54 81 university of bamberg press in finland, students must pass an entrance examination in order to be accepted to pursue a degree, and this is where the first introduction to complexity occurs, as one of the articles in the pre-reading materials for the entrance examination deals with complexity leadership (lindell, ollila & vartiainen 2014). the department’s bachelor’s degree courses familiarize students with the complexity worldview, and the master’s degree courses extend the understanding of complexity. research interests on complexity have also been expanded at the faculty level, which is evident from the master’s theses and the phd dissertations applying a complexity framework. the department of social and health management also strives to extend the understanding of complexity on the societal level. this takes place for example through podcasts, open university courses, and lectures offered to a range of organizational actors. the approach adopted in the department’s curriculum is that of a “softer” complexity. in richardson’s (2007) classification, this approach could be considered (depending on the content of the course) to align with either the critical pluralist school or the metaphorical school. while the critical pluralists focus on “the limits of our knowledge in the light of complexity, limits that are often trivialized by contemporary scientific thinking,” the representatives of the metaphorical school use complexity as a “powerful metaphorical tool” (richardson & cilliers 2001, p. 6–7). later in this paper, we explore the issues around introducing the “harder” side of complexity (the neo-reductionist school) into the curriculum. for richardson and cilliers (2001, p. 5) this particular side of complexity can be understood as a reductionist complexity science that is “strongly allied to the quest for a theory of everything (toe) in physics.” owing to the department’s grounding in qualitative studies and social sciences, there was a rather natural leaning toward the softer side of complexity. the premise for the strong inclusion of the complexity perspective in the department’s curriculum can be found in boulton, allen and bowman (2015, p. 123) who argue that, “to accept that the world is complex rather than predictable and controllable is to change our approach to everything: our approach to change, to management, to policy development, to evaluation, to leadership – and to living!” the premise applies equally to teaching. the public management curriculum and educational methods have to respond to reality, to the complexity of the world the students of public administration face every day. public management educators have an obligation to enhance their students’ capacity to act in an environment of ever-increasing complexity. this is especially true in the health care sector where “leaders at all levels are faced with some of the most complex and challenging problems confronting leaders” (hutchinson et al. 2015, p. 3021). 2. courses, in-service training and educating wider society in social and health management, the course contents and learning methods are diverse, and below we detail the courses most clearly connected with the complexity perspective, starting with examples where the complexity perspective is most explicit. the description of the courses continues with the level of detail offered reducing gradually until we reach the courses that only touch upon complexity. furthermore, we provide examples of in-service training and attempts to educate the general public. http://dx.doi.org/10.20377/cgn-54 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 80-91 doi: http://dx.doi.org/10.20377/cgn-54 82 university of bamberg press the curriculum edge of chaos is a master’s degree level course that has been on the curriculum since 2009. the course is divided into three modules. the first module helps students extend their understanding of the history and key concepts of the complexity sciences. the 10 principles of complexity by cilliers (1998) and mitleton-kelly (2003) are used as a general basis for establishing an understanding of complexity. the course content seeks to convey concepts that can be difficult to comprehend through real-life examples. the second module includes the perspective of complexity leadership, and in it, complexity leadership theory (see e.g., lichtenstein et al. 2006; uhl-bien, marion & mckelvey 2007) and its applications are explored in detail. the third module is more action-oriented and the method of implementation usually changes every year. past examples of experiments include using dialogue mapping, deliberative forums, and network analysis. dialogue mapping (see conklin 2005; raisio & lindell 2010) is a facilitation tool where a shared display is used and the problem is examined through dialogue. as a tool, it helps to illustrate the multi-dimensional nature and the ambiguity of wicked problems, and the importance of collective intelligence in handling these particular issues. deliberative forums work in a similar fashion, but are more strongly built upon the principles of deliberative democracy (see lindell 2014). during deliberative forums students address an actual real-life wicked issue, acquire information through questioning various experts, take part in facilitated small group discussions, and finally write out their proposals to tackle the issue, which are given to the relevant decision makers. social network analysis [sna] (see e.g., koliba 2014, koliba et al. 2011, 2016) is the most recent development, and attempts to deliver an improved balance between “soft” and “hard” complexity. social network analysis also stresses the importance of relations (ties) between actors and can thus be applied to illustrate the interconnected and intertwined nature of complexity. the students are graded based on a traditional examination. welfare game – the complex nature of social and health services is a bachelor’s degree level course, which was first offered in 2015. the pedagogy of work-related learning forms the basis of the course. the course involves students in simulation gaming (see klievink & janssen 2010). due to the nature of the course, the number of course participants must be limited, for example, to a maximum of 20 students. the course begins with students being informed about the background of simulation gaming and its relationship with the complexity worldview. a practice oriented article describing the experiences of real leaders embracing the paradigm of complexity (raisio & lundström 2015) is obligatory reading material. the article uses three detailed narratives to present to students what complexity means in real life. the actual game is based on real world cases that are processed though different scenarios. students are divided into groups of three or four, and each group has one researcher and one social and health care professional as its instructors. each group has a different opening scenario. the progress of the scenario will vary nonlinearly according to the choices and decisions made by students. the game, however, does not proceed in real time. after each new scenario development, students have on average two weeks to record their responses based on the relevant source material. they are subsequently informed of the instructors’ reactions. the game’s “plot” then takes shape and develops through the interaction between instructors and students. for example, one group struggled with a scenario where they were faced with the consequences of the european refugee crisis. the scenario was set in a small fictional finnish municipality and the students took different roles (e.g., municipal manager and the chair of the municipal council) depending on the nature of each new scenario development, beginning with the preparation for the arrival of asylum seekers. the aim of the course is to prepare the students for increasingly more complex operating environments. students begin to understand how different parts of the whole are http://dx.doi.org/10.20377/cgn-54 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 80-91 doi: http://dx.doi.org/10.20377/cgn-54 83 university of bamberg press interwoven, interact, and hence co-evolve. in addition, the non-linear scenarios encourage the students to develop their capacity for what-if thinking. the course is graded as a pass or fail. the future of social and health systems is a master’s degree course, which has been on the curriculum since 2016. the course combines futures studies, a complexity worldview, and the analysis of social and health care systems. its focus is on futures studies as methods and techniques for administrative sciences, not so much on futures studies as a science. concepts such as trends, megatrends, weak signals, wild cards and black swans, and methods such as scenario building workshops, the futures wheel, and the delphi method are presented and explored. moreover, instead of attempting to teach forecasting (predicting one single best future), the emphasis of the course is on foresight (developing many plausible scenarios for the future). according to curry (2014, p. 13) “it is in scenario building where there is the greatest room for growth in public administration futures research.” additionally, the relationship between foresight and complexity is discussed throughout the course. the course addresses for example the way that complexity challenges scenario thinking in futures studies. as cilliers (2000, p. 28) has stated, to model complex systems, we would have to “model life, the universe and everything,” which is naturally impossible. in futures studies, and in foresight in particular, this can be seen as a widely accepted fact. the promise of creating future scenarios is then not so much in their power to predict the future, but in that such scenarios make us speak out and imagine what the future might hold (see raisio & lundström 2015). the analysis of potential futures for social and health care systems takes place after the opening lectures. students form small groups and select a theme for their group. each group is tasked to write a futures oriented analysis of their selected theme, such as what the hospital district of helsinki and uusimaa might look like in the year 2030; or how technological developments could influence the provision of health care services in the future. for approximately a month, students familiarize themselves with the theme, and analyze the associated trends and megatrends. they look for weak signals and reflect upon different black swan events. they use their analysis to formulate from three to four future scenarios and present them to other students. as a group, the students explore different scenarios and assess their plausibility and desirability. the grade is based on the quality of the written work and the presentation. the citizen’s voice in the welfare society is a bachelor’s degree level course offered for the first time in 2015. the premise of the course is that the ever-increasing prevalence of complexity and of wicked problems calls for a more active citizenry and a stronger role for public participation in public administration. as yankelovich (2015, p. 5) argues “…with all the wicked problems the nation faces, it will be difficult to get back on track without a more thoughtful, more fully engaged public, and without a more public-minded philosophy than now prevails.” the course consists mainly of lectures and an exam. lectures cover three themes: public participation in general, client/patient participation, and deliberative democracy. the focus is particularly on public deliberation, as the growth of deliberative democratic theory has brought to the fore the role of citizens in addressing wicked problems (see raisio & vartiainen 2015). the students are made aware of how taking part in deliberative mini-publics, such as citizens’ juries, deliberative polls, and also participatory budgeting, can help citizens grasp the complexity of public policies. additionally, students produce a written assignment exploring specific participatory and deliberative methods selected from the website participedia.net. in 2017, facilitation training by a qualified facilitator was also included in the content of the course. the aforementioned courses are those most explicitly dealing with complexity, but other courses also touch upon the topic. for example, in one of their first courses, welfare services in a http://dx.doi.org/10.20377/cgn-54 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 80-91 doi: http://dx.doi.org/10.20377/cgn-54 84 university of bamberg press developing environment, students examine the current challenges and opportunities for the development of the welfare society. during the course, complexity and wicked problems are very briefly explained and connected to the context of the welfare society. the bachelor’s level course, from vintage to modern leadership, covers the historical background of leadership development. complexity leadership is briefly examined as a part of modern change leadership. on a master’s level course, human resource management and welfare, one of the topics is conflict management and working life. the course content includes the use of roleplays to foster a realistic understanding of managing highly complex situations. the use of real complex cases as instruments means each student takes a different role in their own group and so is exposed to some extent to reality. after each roleplay exercise, the lecturer debriefs the students on the solutions they applied to handle such delicate and complex situations. in addition, the students are exposed to a complexity perspective on a course entitled project innovations and practical development. on this master’s level course, students, guided by a researcher, act as a research community and implement a real research and development project. one of the more implicit objectives of the course is to help students to understand the complexity of engaging in project work. in-service training in-service training on complexity is provided by the department and varies from the provision of lectures of between two and three hours duration to one-day training modules. the department is, for example, taking part in a middle management leadership coaching program, organized by advocacy organizations for local government employers and for finnish municipalities. in 2017, the program will be run for the third time. the program lasts for one year, after which participants receive a certificate as an indication of their capabilities to hold a management position. in the first year, 137 participants received the certificate. the social and health care department’s role on the program is to introduce the complexity worldview to the participants. during a one-day training module, the first half day includes an introduction to the central tenets of complexity and the main characteristics of wicked problems, alongside an exploration of complexity leadership theory. as a novel content, the so-called emergence of evil (see bella 2006; vartiainen, raisio & lundström 2016) is studied. in raising this darker side of emergence, bella, king and kailin (2003, p. 68), refer to “dark outcomes [which] emerge from interactions among well-intended, hardworking, competent individuals.” the concept resonates strongly with the experiences of the participants of in-service training. the second half of the day sees students participating in a world café (see carson 2011), to encourage them to reflect on and build a shared understanding of complexity and problem wickedness. educating the wider public examples of methods used to educate the general public on complexity include podcasts and an open university course. the relevant podcasts cover the topics of complexity and wicked problems and are hosted by one of the department’s researchers and produced in cooperation with keva, a public-sector pension institution with an interest in workplace wellbeing. the first season began in january 2017 and each season is scheduled to include approximately 10 episodes, each of around 45 minutes duration. the aim has been to invite guests who are senior leaders and have a deep understanding of complexity, and who have used that understanding to develop their http://dx.doi.org/10.20377/cgn-54 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 80-91 doi: http://dx.doi.org/10.20377/cgn-54 85 university of bamberg press leadership style. the open university course, uncertainty: a path to innovation was offered as part of the so-called newspaper university, a tradition stretching back over 20 years that is organized by the university of vaasa, the open university, and the local newspaper, pohjalainen. the newspaper university project ensures popular science articles on a topical theme are published in the local newspaper, and supports the building of an open university course related to those themes. completing the course requires students to expand their reading to incorporate other relevant literature and write an essay on a related topic. the social and health management department took responsibility for the course in 2013. newspaper articles explored the complexity worldview from different perspectives, and were published both in print and online (see sanomalehtiyliopisto 2013), to offer non-subscribers an opportunity to participate. ongoing interest in the course means it has remained among the courses offered by the open university. to date, around one hundred people have taken part in the course. 3. student feedback feedback from the university students and the participants in the in-service training has been positive. one contributory factor might be that the students have mostly been adult learners. one opinion elicited from the feedback received is that having work experience and being faced with complexity in everyday life helps people appreciate what complexity is about. it has often been stated that gaining an understanding about complexity eases the pain of dealing with the rapidly moving, constantly changing environment, where nothing is certain. complexity seems to offer an answer to the question of why everything is as it is. as one the leaders interviewed by raisio and lundström (2015, p. 2) stated, “in a way [complexity] is a liberating image of reality. it just is and we do not need to feel pain about it.” becoming familiar with the complexity worldview enables students and participants to assign names and labels to phenomena and problems they face in everyday life. as theory and reality merge together “a lot of things fall into place.” in particular, the various practical exercises undertaken during the courses are considered useful. for example, the simulation gaming on the welfare game course was initially challenging for the students. typically, in examinations, answers are often considered to be either right or wrong, but in simulation gaming this does not apply. actions taken are in the students’ own hands, which might cause discomfort at first. the course itself proceeds according to the principles of complexity. both simulation gaming and the scenarios created in the future of social and health systems course have helped students to understand the high degree of connectivity and interdependence between the different actors and dimensions of a system and also between the system and its environment, giving rise to complex behaviors and coevolution. similarly, the deliberative forums have been a valuable way of exposing students to the apparent complexity of public service structures and policy. in addition, the various exercises and methods have not only been pedagogical tools, but also practical ones available for use in the future. students have, for example, implemented various deliberative forums at their employers. moreover, they have become more alert to weak signals and observant in imagining different futures. as one of the students recently explained: “weak signals and the search for them led me to react to newspapers somewhat differently than before. or indeed, the theme of the whole course [futures studies and complexity] was awakening, so that in newspapers my attention is drawn to different things than before.” students’ feedback supports the optimism of the teachers that the complexity knowledge gained during the courses also helps students to confront the complexity of the real world; they can be more at ease with complexity and strive to steer their organizations toward the space of http://dx.doi.org/10.20377/cgn-54 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 80-91 doi: http://dx.doi.org/10.20377/cgn-54 86 university of bamberg press possibilities at the edge of chaos. as future managers, they would be more agile and able to adapt as well as more tolerant of ambiguity and failure. however, learners have also reported that the concepts used in complexity can initially be difficult to comprehend, particularly because the language of complexity differs from that they are used to, and hence, absorbing the complexity worldview takes time. course participants have also reported on factors facilitating their learning; and have suggested more discussion sessions during the courses and for the course work to have a stronger connection to practical work. in addition, the participants on in-service training courses in particular have highlighted the challenge of thinking and acting according to the complexity worldview in institutions organized on the basis of a mechanistic view. participants have stressed that existing structures and legislation make it impractical to challenge the mechanistic worldview. that situation highlights a form of imbalance or even a paradox between learning and doing. the various barriers to embracing complexity have recently been explored in detail in the leadership under complexity workshop, organized by the finnish innovation fund sitra (see doz et al. 2017). 4. developments in the future the department of social and health management at the university of vaasa is considering several ways in which the curricula it administers might be developed. in their current form, the curricula highlight the so-called soft side of complexity. depending on the course structure, the complexity perspective is based either on the metaphorical school (mainly at the bachelor’s degree level) or the critical pluralist school of complexity (mainly at the master’s degree level). the hard side of complexity receives considerably less attention. this neo-reductionist school of complexity (see richardson 2007) should feature more strongly in the department’s curricula in the future. for example, agent-based modeling and simulations could be taught and explored in the context of social and health care systems (on agent-based modeling in pa see e.g., koliba 2014; koliba et al. 2011). social network analysis and dynamic social network analysis also seem promising, in that they might bring together the empirical and theoretical sides of complexity in public administration (e.g., puustinen 2017; schipper & gerrits 2015; schipper et al. 2015; schipper & spekkink 2015). combining network theory and complexity theory could be seen as one of the new streams bringing together complexity and pa (see e.g., eppel 2017; koliba et al. 2016; koppenjan & klijn 2014). although promising, network theory, sna, simulations, and modeling also have their weak points. none of them can extensively address the real-life dynamics of a complex social system, where actors constantly interpret and are aware of their own behavior and that of others (e.g., deacon & koutroufinis 2014; puustinen 2017). such dynamics are hard to employ with students since all of them require a significant methodological understanding that neither bachelor’s or master’s level degree courses routinely include. in practice, bringing together the harder and softer forms of complexity could be for example an optional advanced course, which would also be offered to students of other faculties, especially the faculty of technology, and the department of computer science within it. this would combine the metaphorical understanding of complexity and the mathematical, computer based simulations of complex systems. however, there are some potential barriers to such a plan: the scientific languages of public administration and natural or computer sciences do not always accord, and research streams have been, and still are, rather separate (e.g., hidalgo 2015). http://dx.doi.org/10.20377/cgn-54 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 80-91 doi: http://dx.doi.org/10.20377/cgn-54 87 university of bamberg press nevertheless, the multidisciplinary aspect of the teaching of complexity in public administration should be strengthened, and this can be done in ways other than by methodological advances. now the perspective of the curriculum is most clearly focused on the public sector. in the university of vaasa, collaboration could be enhanced not only with the aforementioned faculty of technology, but also with the faculty of business studies, which hosts researchers with a background in complexity science (e.g., luoma 2006). such collaboration could for example open up the area of managing complexity in public-private partnerships as a topic to study. a multidisciplinary approach to teaching would naturally follow the principles of complexity. most clearly, theory-practice dialogue needs to be increased during the courses. this would strengthen the connections between the theoretical side of complexity and the existing management culture. specifically, the power of example should be applied. this could be realized through presentations by visiting lecturers, who would be leaders strongly embracing the paradigm of complexity, such as those who informed the study of raisio and lundström (2015). as an alternative option, webinars and podcasts could be applied. theory, practice, and empirical research on complexity should also be strengthened in a similar manner. this applies both to encourage the more comprehensive inclusion of students in ongoing research initiatives, and enhanced action-oriented research agendas focusing on analyzing complexity and its facets in society at large and organizations in particular. furthermore, in-service training could be extended in the years to come, especially through multidisciplinary collaboration. to date, training has been offered almost entirely to public-sector organizations, and in future, the private and third sectors should also be included. in-service training could also be of longer duration and more extensive, even including the elements of action research in the organization, following for example mitleton-kelly’s emk methodology (mitleton-kelly 2015). adopting a more utopian view, the future objective should be to make complexity thinking a civic skill. this view emerged in one of the small group dialogues in a recent one-day in-service training module. the respondents suggested that social media has created “bubbles and black and white thinking.” originating from there, complexity needs to be understood better and “complexity thinking should be trained.” 5. advice to teachers of complexity in public management contexts based on our experience in teaching complexity, we can offer the following advice. first, it is useful to combine the fields of wicked problems and complexity sciences (see zellner & campbell 2015). the concepts of tame and wicked problems (rittel & webber 1973) help to concretize the differences between the mechanical worldview and the worldview of complexity. for example, often during in-service training the concept of wicked problems is first defined and different resolution strategies explored, and only after that are they connected to complexity thinking. this step-by-step approach makes it easier to take in the concepts, ideals, and practices of complexity. it is especially useful when the wicked problems presented are familiar to the learners, so they resonate strongly with their own experiences. this is also in line with the need for public administrators to recognize their respective theories-in-use applied in their work, and to find the ways in which complexity concepts add value to the existing mental models and extant policies (eppel 2017). as eppel (2017, p. 858) further states, “we need to develop a deliberative sensitivity to complexity friendliness,” both among our students and in public administration in general. http://dx.doi.org/10.20377/cgn-54 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 80-91 doi: http://dx.doi.org/10.20377/cgn-54 88 university of bamberg press second, the role of examples, exercises, and a practical orientation cannot be emphasized enough. the risk is that complexity remains on an abstract, theoretical, and conceptual level, which significantly hinders learning. using different kinds of exercises, such as simulation gaming and deliberative forums, students not only begin to grasp complexity better, but they actually get to operate within complexity. in teaching complexity, the role of the teacher must be emphasized. ideally the teacher acts as a mediator between the abstract theory and the real-life phenomena. classic literature on complexity can be very challenging for students, so the teacher has to be able to illustrate what complexity is about in practice. third, complexity should be an all-encompassing theme in the curriculum, not only a separate course. otherwise an explicit disparity might emerge between what is taught on the complexity course and on other courses. courses should be mutually supportive and their connections should be made clear to all students. for example, in the department’s futures studies course, the interfaces and challenges involved in combining the complexity worldview and forecast and foresight are made explicit. this ensures that the parts comprising the curriculum work in unison to form a coherent whole, rather than as separate bits and pieces that can cause confusion among students. references bella, d., king, j.b. & kailin, d. (2003). the dark side of organizations and a method to reveal it. emergence: complexity and organization 5(3), 66–82. bella, d.a. (2006). emergence and evil. emergence: complexity & organization 8(2), 102–115. boulton, j.g., allen, p.m. & bowman, c. (2015) embracing complexity: strategic perspectives for an age of turbulence. oxford: oxford university press. carson, l. (2011). designing a public conversation using the world café method. social alternatives 30(1), 10–14. cilliers, p. (1998). complexity and post-modernism. understanding complex systems. london: routledge. cilliers, p. (2000). what can we learn from a theory of complexity? emergence 2(1), 23–33. conklin, j. (2005). dialogue mapping: building shared understanding of wicked problems. new york: wiley. curry, d. (2014). the crystal ball of public administration reform: what future prediction tell us about the present. cocops working paper no. 15. retrieved april 18, 2017 from http://www.cocops.eu/wp-content/uploads/2014/06/wp15.pdf deacon, t. & koutroufinis, s. (2014). complexity and dynamical depth. information 2014(5), 404-423. doz, y., hellström, e., kosonen, m., lähdemäki, j. & wilson, k. (2017). reflections on the “leadership under complexity” workshop. sitra working papers. retrieved april 18, 2017 from https://media.sitra.fi/2017/03/15145726/reflections_on_the_leadership_under_complexit y_workshop1.pdf http://dx.doi.org/10.20377/cgn-54 http://www.cocops.eu/wp-content/uploads/2014/06/wp15.pdf complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 80-91 doi: http://dx.doi.org/10.20377/cgn-54 89 university of bamberg press eppel, e. (2017). complexity thinking in public adminstration’s theories-in-use. public management review 19(6), 845-861. hidalgo, c. a. (2015). disconnected, fragmented, or united? a trans-disciplinary review of network science. retrieved april 18, 2017 from https://arxiv.org/ftp/arxiv/papers/1511/1511.03981.pdf. hutchinson, m., daly, j., usher, k. & jackson, d. (2015). leadership when there are no easy answers: applying leader moral courage to wicked problems. journal of clinical nursing, 24(21–22), 3021–3023. klievink, b. & janssen, m. (2010). simulation gaming as a social development instrument: dealing with complex problems. information polity 15, 153–165. koliba, c., meek, j.w. & zia, a. (2011). governance networks in public administration and public policy. public administration and public policy / 158. boca raton: crc press, taylor & francis group. koliba, c. (2014). governance networks performance. a complex adaptive systems approach. in keast, r. & mandell, m. & agranoff, r. (eds.). network theory in the public sector. building new theoretical frameworks (pp. 84-102). routledge critical studies in public management 17. london: routledge. 84-102. koliba, c., gerrits, l., rhodes m. l. & meek j. (2016). complexity theory and systems analysis. in ansell, c. & torfing, j. (eds.). handbook on theories of governance (pp. 364-379). cheltenham: edward elgar publishing. koppenjan, j. & klijn, e-h. (2014). what can governance network theory learn from complexity theory? mirroring two perspectives on complexity. in keast, r. & mandell, m. & agranoff, r. (eds.). network theory in the public sector. building new theoretical frameworks (pp. 157-173). routledge critical studies in public management 17. london: routledge. lichtenstein, b.b., uhl-bien, m., marion, r., seers, a., orton, j.d. & schreiber, c. (2006). complexity leadership theory: an interactive perspective on leading in complex adaptive systems. emergence: complexity & organization 8(4), 2–12. lindell, j., ollila, s. & vartiainen, p. (2014). kompleksisuuden johtaminen [complexity leadership]. in k. lehto (ed.). pirullisista ongelmista hyvään hallintoon [from wicked problems to good governance] (pp. 86–100). vaasa: vaasan yliopiston opetusjulkaisuja 65. lindell, j. (2014). fostering organizational change through deliberations: the deliberative jury in a university setting. tertiary education and management 20(1), 30–43. luoma, m. (2006). a play of four arenas: how complexity can serve management development. management learning 37(1), 101–123. mitleton-kelly, e. (2003). ten principles of complexity & enabling infrastructures. in mitletonkelly, e (ed.). complex systems & evolutionary perspectives of organisations: the application of complexity theory to organisations (pp. 23-50). bingley, uk: emerald group publishing limited. mitleton-kelly, e. (2015). effective policy making: addressing apparently intractable problems. in geyer, r. & cairney, p. (eds.) handbook on complexity and public policy (pp. 111-127). cheltenham, uk: edward elgar publishing. http://dx.doi.org/10.20377/cgn-54 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 80-91 doi: http://dx.doi.org/10.20377/cgn-54 90 university of bamberg press puustinen, a. (2017) voiko verkostoa johtaa? tapaustutkimus sosiaalija terveydenhuollon yhteistoiminta-alueen hallinnan yhteenkietoutuneesta luonteesta. [can networks be managed? a case study on the complexity of governance in a co-operation district of health and social care.] publications of the university of eastern finland. dissertations in social sciences and business studies no. 149. kuopio: university of eastern finland. raisio h. (2010). embracing the wickedness of health care: essays on reforms, wicked problems and public deliberation. acta wasaensia 228. vaasa: vaasan yliopistopaino. raisio h. & lindell j. (2010). kuinka kasvattaa opiskelijoiden työelämävalmiuksia selviytyä yhä haastavammiksi kasvavista hallinnon ongelmista? esimerkkinä dialogikartoitus. [how to increase students capabilities to survive ever challenging public management problems? dialogue mapping as an example]. peda-forum: yliopistopedagoginen aikakausjulkaisu 17(2), 22–26. raisio, h. & lundström, n. (2015). real leaders embracing the paradigm of complexity. emergence: complexity & organization 17(3): 1–5. raisio, h. & vartiainen p. (2015). accelerating the publics learning curve on wicked policy issues: results from deliberative forums on euthanasia. policy sciences 48(3), 339-361. richardson, k. a., & cilliers, p. (2001). special editors’ introduction: what is complexity science? a view from different direction. emergence: complexity & organization, 3(1), 5-23. richardson, k.a. (2007). complex systems thinking and its implications for policy analysis. in morcöl, g. (ed.). handbook of decision making, public administration (pp. 189-222). boca raton, fl: crc press. rittel, h. w. j. & webber, m. (1973). dilemmas in a general theory of planning. policy sciences 4, 155–169. sanomalehtiyliopisto (2013). uutta luova epävarmuus [creative uncertainty]. vaasan avoimen yliopiston julkaisu 37. retrieved april 18, 2017 from http://www.uva.fi/materiaali/pdf/isbn_978-952-476-521-3.pdf schipper, d. & gerrits, l. (2015). the emergence of metropolitan governance: a coevolutionary analysis of the life-and-death cycles of metropolitan governance in the amsterdam metropolitan region. complexity, governance & networks, 1(2), 5778. schipper, d. & gerrists, l. & koppenjaan, j. (2015). a dynamic network analysis of the information flows during the management of a railway disruption. ejtir 15(4), 442-464. schipper, d. & spekkink, w. (2015). balancing the quantitative and qualitative aspects of social network analysis to study complex social systems. complexity, governance & networks, 2(1), 5-22. uhl-bien, m., marion, r. & mckelvey, b. (2007). complexity leadership theory: shifting leadership from the industrial age to the knowledge era. the leadership quarterly 18(4), 298–318. vartiainen, p. (2005). wicked health care issues: an analysis of finnish and swedish health care reforms. advances in health care management: international health care management 5, 163 – 186. vartiainen, p. (2008). health care management in finland: an analysis of the wickedness of selected reforms. review of business 28(2), 41–55. http://dx.doi.org/10.20377/cgn-54 http://www.uva.fi/materiaali/pdf/isbn_978-952-476-521-3.pdf complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 80-91 doi: http://dx.doi.org/10.20377/cgn-54 91 university of bamberg press vartiainen, p., raisio, h. & lundström, n. (2016). kompleksisuuden johtaminen – kohti uutta ymmärrystä [complexity leadership: towards new understanding]. in syväjärvi, a & pietiläinen, v. (eds.) inhimillinen ja tehokas sosiaalija terveysjohtaminen [humane and efficient leadership in social and health care] (pp. 227–267). tampere: tampere university press. yankelovich, d. (2015). wicked problems workable solutions: lessons from a public life. london: rowman & littlefield zellner, m. & campbell, s.d. (2015) planning for deep-rooted problems: what can we learn from aligning complex systems and wicked problems? planning theory & practice 16(4), 457– 478. http://dx.doi.org/10.20377/cgn-54 teaching public administrators and leaders to handle complexity 48 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press evaluating technological progress in public policies: the case of the high-speed railways in the netherlands authors: peter marksa, lasse gerritsb a department of public administration, erasmus university rotterdam, the netherlands, e-mail: marks@fsw.eur.nl b department of political science, university of bamberg, germany, e-mail: lasse.gerrits@uni-bamberg.de main-stream evaluations of failed policies are geared towards finding a limited set of factors that are deemed to have caused the problem. this is particularly so in the case of high-profile public projects such as in technology and infrastructure development. while justified from the point of political accountability, this article presents an alternative view. following insights from evolutionary economics and complex systems about the embedded nature of technological systems and the role of chance next to purposeful planning, we demonstrate that traditional policy evaluations are misguided when geared towards simplistic cause-and-effect relations. to this end, we analyze the reasons for the mixed results in the dutch high-speed railway case. the findings show that, contrary to popular opinions in the political domain, technological progress did take place. however, misalignment between social practices and technological systems masked that progress. keywords: innovation policy, complexity science, socio-technological innovation, policy evaluation 1. introduction high-speed railway projects are important in many ways. they enable people to travel longer distances in shorter time spans, which promotes economic activities. cities that are connected to high-speed networks often benefit in terms of hub functions. high-speed railways are technologically advanced. it requires quality engineering to run a full train safely at 300 km/hr. high-speed railways also carry prestige and are an important means of profiling a country or even people such as ministers. japan, france and germany, and more recently china, have all embarked on such projects for such reasons. when all components fit together, high-speed railways constitute a valuable and durable infrastructure. deciding, planning and constructing high-speed railways can take decades. meanwhile, many things may change; e.g. technologies may have to mature more, passengers may have changed their behavior, and alternative modes of transport may have cropped up. there are many hurdles to take before the first shovel hits the ground and many more before the first train runs. when the components don’t line up the prestigious project can turn into an expensive problem. this is the case for the high-speed railways in the netherlands, which ultimately costed almost twice as much as planned at the start. the dutch government started thinking about high-speed rail in the netherlands at the end of the 1970s. in the early 1990s the first concrete steps were made towards building of a high-speed rail; the so-called hoge-snelheidslijn zuid (high-speed railway link south or hsl-zuid). after more than 20 years of deciding, building and constructing the first high-speed train, called fyra, started revenue service on december 9th 2012. during the mailto:marks@fsw.eur.nl mailto:lasse.gerrits@uni-bamberg.de 49 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press weeks to follow many (safety) incidents occurred after which the rail operators decided to take the fyra-trains off the track on january 17th, never to return anymore. when things go wrong, as such mega-projects often do, common political sense dictates that it is time to evaluate and to pinpoint the ones deemed guilty for the mistakes. the logic of accountability in the public realm holds that, ultimately, responsibility can be traced to one factor or actor (cf. teisman, 2005). traditional ex-post policy evaluation serves two questions: whether the policy has reached its target, or whether the people working in a certain policy field or public organization have delivered what they were supposed to deliver (parsons, 1995). in addition, there is also a tendency to understand that different components of the project as related but essentially discrete elements. in practice, policy evaluation runs into a collection of structural problems, such as establishing causality (noordegraaf & abma, 2005; bressers, 2011), establishing consensus over what constitutes a ‘good outcome’ because of subjective appraisal (bressers & hoogerwerf, 1995), and the fact that any policy is faced with resource restrictions that can ruin the intended outcome (parsons, 1995). policy evaluations are often carried out for control and accountability purposes, which means that learning from unfavorable outcomes takes a backseat (cf. guba and lincoln, 1985 for a critique). when it comes to major public investments such as in infrastructure, policy evaluation is often only carried out when something has gone wrong and the evaluation is inevitably geared towards finding and blaming the ‘guilty’ person, who can then be punished accordingly (cf. kerseboom, 2008, p. 7; kim., 2009). the dutch case is no exception, as exemplified by the following conclusions from the formal parliamentary evaluation report: “the intended service has not been established because other interests prevailed over providing the service. the state was too focused on monetary gains and ns was too focused on retaining its strategic position. not only did that hinder passengers’ interest, it also meant an underutilization of the hsl-zuid.” (parlementaire enquêtecommissie fyra, 2015, p. 4) such statements are understandable from a legal or political point of view – the derailment of those projects provides an excellent opportunity for the opposition to show the world they had been right all along. while such evaluations fit comfortably with the reality of politics, they don’t deliver a genuine insight in the dynamics and mechanisms that govern such programs and projects. indeed, the goal of establishing accountability could even hinder a better understanding of the actual causality behind the unfavorable outcome (cf. verweij, 2015). the traditional policy evaluation perspective that tries to pinpoint the single cause for the problems ignores the configurational nature of socio-technological systems. that is, several socio-technological systems jointly make up the success or failure of programs and projects. arguably all major infrastructural projects, and especially high-speed railway projects, take a long time to be realized from the first thoughts to the actual construction and operation. during the project many technologies will (co-)evolve as well as causing the projects to adjust accordingly to remain successful. in this article, we will contrast the classical view of policy evaluation to an alternative one that builds on insights from evolutionary (economic) theories and the complexity sciences, with particular attention to technology transitions. that is, to understand the evolution of technological configurations we use a perspective in terms of technological transitions. the main question is: what will a complexity-informed evaluation provide in comparison to traditional accountability measures? to this end, we will focus on the development of the dutch high-speed railway project, which – by all accounts – is considered a failure. 50 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press 2. technological transition theory the general idea behind technological transitions (tt) (geels, 2002; rip & kemp 1998; kemp, rip & schot, 2001) is that analyzing technology implies understanding them as configurations, i.e. the alignments of a heterogeneous set of elements (e.g. bijker, 1995; moody, 2009; for an extended argument). configurations are logically connected with the rest of society; e.g. it is embedded in items such as skills, routines, patterns of behaviors. in other words: “[…] societal functions are fulfilled by sociotechnical configurations.” (geels, 2002, p. 1258) technological transition requires a change from one configuration to the other. changing configurations means not only substituting technology but also all other elements in the configuration that are mutually connected and aligned. as long as there is no proper match between the established regulations, infrastructure, user practices, et cetera, new technologies will have a hard time breaking through as they are mismatched with the established socio-institutional framework (freeman & perez, 1988 in geels, 2002, p. 1258). in other words, all components need to be lined up correctly for socio-technological systems to evolve. stability of such configurations is the result of the linkages between the heterogeneous elements, which are the result of activities of social groups producing those elements and linkages (geels, 2002, p. 1259) to understand the coordination of the activities of the different groups the tt-perspective builds on the concept of technological regimes of nelson & winter (1977; 1982); i.e. coordination is the outcome of organizational and cognitive routines. a technological regime is formed if the organizations, and the actors involved, share similar routines, i.e. routine-based behavior. technological regimes create stability because they guide the innovative activity towards incremental improvements along trajectories. rip & kemp add to the technological regime that it is the “rule-set or grammar embedded in a complex of engineering practices, production process technologies, product characteristics, skills and procedures, ways of handling relevant artefacts and persons, ways of defining problems.” (1998, p. 338) in other words, the rules are not only embedded in the engineering practices, but also in the corporate governance structures, manufacturing process and product characteristics. this means many social groups are part of the community and as such the possible trajectory of the technology is influenced by many users beside engineers, i.e. policy makers, societal groups, et cetera (kemp, et al., 2001, pp. 272-273). “regimes are outcomes of earlier changes and they structure subsequent change.” (rip & kemp, 1998, p. 338) as such st-regimes may change, but do so incrementally. this sociotechnical regime provides orientation and coordination to the activities of the involved actors. the sociotechnical regime then functions as a selection and retention mechanism (nelson & winter, 1982) providing stability of the respective sociotechnical configurations. the sociotechnical regimes function within the context of the sociotechnical landscape. these landscapes constitute the wider factors relevant for the respective st-regimes, and at this macro-level the landscapes change extremely slowly. more radical innovations occur at the microlevel of local practices, i.e. the so-called niches (rip & kemp, 1998, p. 338; kemp, et al., 2001, p. 275). there is feedback surging through the nested hierarchical levels. the meso-level of stregimes accounts for stability of existing technological development and the occurrence of trajectories. the macro-level of landscape consists of slow changing external factors, providing gradients for the trajectories. the success of a new technology is governed by what happens within the niche, but also by the developments in the sociotechnical regimes and landscape; e.g. changes at the landscape level may put pressure on the sociotechnical regime creating an opening 51 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press diesel high-speed (1899) diesel 160 km/h (1933) electric 203 km/h (1938) shinkansen: wide gauge, dedicated track, 25 kv ac (1964) hsr narrow gauge (1957) tapered nose (70s) for a new technology niche (geels, 2002, p. 1261). in short, technological transitions occur as the outcome of linkages between developments at multiple levels, but also at multiple technologies. alignment between levels creates a window of opportunity for successful technological transitions; i.e. all components need to be lined up correctly – something that can only be partially managed because certain elements are relegated outside of policy control. this is the main theme in understanding technological change. the transition perspective can be used to evaluate policies: within the larger policy landscape different regimes play a coordinating role for different policies or policy instruments to be successfully implemented. for the regimes to function as coordination all components and networks must be lined up (bressers, 2011). systems can keep each other going but are also highly dependent on each other, i.e. when one misfires, others will misfire too. for a better understanding of why the dutch high-speed railways project failed, it is necessary to assess how and to what extent all systems components aligned. data for this research was collected from multiple archival records: 1191 newspaper articles about the case retrieved from the nexis-database; 40 policy papers issued by the ministry of transport; 28 policy briefs issued by the parliament; and 14 scientific and professional reports published by third parties (more details about the data can be found in gerrits and marks, 2014). in addition, we consulted approximately 50 websites. 3. socio-technological transitions in the case of high-speed railways in the netherlands the following sections will describe the formation of the landscape and the evolution of various niches that, together, lead to the emergence of high-speed railway technology in general, and its deployment in the netherland specifically. this overview is derived from a more extended reconstruction that is available upon request. figure 1 historical build-up of the high-speed rail landscape 52 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press 3.1 setting the standard: landscape formation the development of the dutch high-speed railways is embedded in the landscape of (highspeed) railways that evolved from 1899 into the standard-setting shinkansen service between tokyo and osaka that opened in 1964. this line featured continuously welded rails in a 25% wider gauge designed for 250 km/h. figure 1 shows the evolution towards the standard-setting shinkansen that has had a defining impact on the technological high-speed railways landscape as it is today. the core elements of all such systems are a set of high-speed rolling stock and a dedicated high-speed line to accommodate that rolling stock. the track is to be continuously welded in order to reduce vibrations and misalignment, and to be bolted on semi-continuous concrete bedding for stability. the 25kv ac power is provided via the overhead catenary, tight radii are to be avoided, and level crossings are not allowed. the last standard-setting came after the early years of the shinkansen, when the train’s nose changed from a bullet shape to the tapered shape to reduce air resistance when traveling at high speeds and to deal with the shock waves when entering tunnels. 3.2 short distance high-speed railways: regime definition the netherlands wasn’t a front-runner when it comes to high-speed railways. the development of the french high-speed trains or train a grande vitesse (tgv) in 1972 sparked a discussion among dutch politicians, civil servants and netherlands railways about having highspeed trains, too. thus in 1973 the first proposal by the dutch government was made to connect amsterdam through rotterdam with belgium, the so-called amrobel-plan. the idea was extended with proposals to improve national and international rail traffic in an attempt to offer a viable alternative to long distance road and air traffic. the formal planning procedure was started in 1987 and encompassed three major studies concerning the creation of a so-called corridor running from amsterdam in southerly direction to rotterdam (the netherlands) and brussels (belgium), i.e. the hsl-zuid. the studies encompassed a formal feasibility study, the environmental impact assessment of all possible variants, and the route decision. there were multiple alternatives for track alignment. however, in line with the landscape development mentioned above, a new and dedicated track that would be as short as possible (i.e. from schiphol to rotterdam in a fairly straight line), that would allow a maximum speed of 300 km/hr was the selected option in 1993. what was really new about the hsl-zuid compared to all other highspeed rail lines that time was that it is essentially a short-distance high-speed connection; i.e. the dedicated track is 212 kilometers long of which only 85 kilometers in the netherlands and 35 kilometers in belgium allow for high-speed, and the line has several stops, from north to south: amsterdam central station, amsterdam airport, rotterdam central station, antwerp central station, and brussels midi. due to these factors the average speed on the line is 120 km/h, far below the maximum design speed. the socio-technological innovation of short-distance high-speed rail travel will function as the coordinating regime for the necessary technological innovations in the various niches. within this regime, the most important socio-technological innovations are: 1) public-private partnership; 2) building technology; 3) safety and interoperability; 4) market demands; 5) rolling stock. we will analyze these different socio-technological innovations since after the decision was made in 1993 to go for the shortest route possible allowing speeds of 300 km/h until the collapse of high-speed service on hsl-zuid in 2013. 53 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press privatization uk (70s) deregulation & privatization eu (80s) deregulation nl (late 80s) public-private partnerships/ tendering (90s) privatization (90s) dbfmo catenary etc. (2000) design & construct foundation (2000) concession hsl (2001) 3.3 niche 1: public-private partnerships after the initial decision where the hsl-zuid would be build, the ministry of transport had to decide how and in which manner to finance the building. inspired by the developments of privatization in uk in the 1970s, privatization of public tasks had become increasingly accepted in continental europe as a manner to improve fiscal deficits, but also as a requisite to join the european union (bortolotti & milella, 2006). the wave of privatization in transport caused deregulation and privatization in airlines, ports and airports in several european countries (phang, 2009, p. 23). the privatization idea took off in the netherlands at the end of the 1980s and meant that new forms of financing were now viable; i.e. tendering of tasks and public private partnerships in projects. “it was clear at the time that there were insufficient public funds available to meet the various desires, so private contributions to public projects were considered as a possible solution.” (koppenjan, 2005, p. 136) the dutch government made steps in corporatizing certain state owned organizations and putting them at more distance, but were hesitant about full privatization (the economist, 2002). initially, the construction costs were projected at approximately 3 billion euros. congruent with the discovery of the possibilities of private market mechanisms, the ministry of transport aimed at a public-private partnership with construction and financial companies to share the costs and benefits of the project. the minister expected € 600 million to be financed by private parties. after some attempts to get parties interested in financing, building or operating parts of the hslzuid either completely or in some form of public private collaboration, the ministry decided that (a) tracks, catenary, noise barriers and signaling systems would be put to tender in dbfmo contracts (design, build, finance, maintain & operate), (b) the construction of the foundation and track bed to be put into six separate tenders on the basis of d&c (design & construct) contracts, and that (c) the concessions to operate the hsl-zuid would be tendered out. figure 2 trajectory of niche-cumulation for breakthrough tendering 54 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press the value of the various dbfmo-contracts added up to around € 700 million and another € 700 million was needed for the maintenance fee for 30 years. high construction risks in conjunction with a risk-averse stance from the private companies lead the minister to decide that the state would build the foundations, tunnels, et cetera through traditional design & construct contracts. this would cost around € 2.7 billion, split over the six separate lots. the costs for building the foundation eventually run as high as € 1.9 billion, which is € 250 million higher than estimated. this was due to builders having strong negotiation positions by having formed illegal cartels. the tendering process for the operation concession turned out to be even messier because the ministry had initially allowed the previously state-owned nederlandse spoorwegen (ns) to make an uncontestable bid for the contract. this bid was declared inadmissible because the minister deemed it too low. a public tendering process was started but political pressure from the house of representatives forced the minister to grand the concession to ns – this time in a consortium with klm (royal dutch airlines) and schiphol international airport. this opportunity was then retracted because it trespassed the european regulations for the single railway market. hence the public tender was reopened. four parties made it to the final round. ultimately, ns won the concession by outbidding everyone else with an offer of € 178 annually, € 78 million higher than the next bidder. in the final negotiations, the annual fee was lowered to € 148 million annually because the ministry feared that ns would go down if it would stick to its original offer. the final amount was still considerably higher than what other operators had offered. 3.4 niche 2: building techniques the construction of the foundation encountered some persistent problems, forcing builders to innovate their building techniques: (1) the soil in the netherlands is prone to subsiding, while fast trains require stable and level tracks, and (2) the track was aligned through a more or less pristine natural area called the green heart, for which a large tunnel had to be built in order to preserve its natural state. however, there was no prior experience with building a large tunnel through such suboptimal ground conditions. nearly the full stretch of the hsl-zuid is founded on piles of reinforced concrete, on top of which are concrete slabs to which the rails are attached creating a continuous and level track. the curves have a minimal radius over 4500 meters and bank to enable high speeds. some wide water bodies needed to be crossed, especially the hollands diep, with very long and rigid bridges to keep the gradient under the required 2 percent so trains can go up to 300 km/h while crossing the bridge without dangerous vibrations. high speed trains require considerable energy to run. the standard current (1500v dc) wouldn’t suffice so a separate energy infrastructure for 25kv ac needed to be build. the technologies required for these points, in combination with the build of the green heart tunnel (next sections), meant that the maximum speed could be raised from the conventional (van den brink, 2002, p. 22). in the densely populated country, tunnels had to be built in addition to level tracks and bridges. a 4 kilometer long tunnel between rotterdam and the northern outskirts of the city was built using traditional cut-and-cover technique. the tunnels under the rivers oude maas and dordtsche kil both consisted of immersed tubes of reinforced concrete. however, the tunnel through the green heart was to be built under strict environmental conditions; no damage to the landscape was allowed and most of the building nuisance had to be avoided. as such the traditional methods would not suffice. hence the 7 kilometer long tunnel had to be drilled using a 55 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press tunnel boring machine (tbm). the evolution of the tbm used for the hsl-zuid is depicted in figure 3. the green heart tunnel had to be built in a sand layer, which is very non-cohesive soil that falls apart when digging (crok, 2002, p. 32). due to this geology and to counter the ground water pressure a slurry shield type tbm was to be used. the french-dutch consortium of bouygues/koop developed the tbm aurora specifically for the ground situation of this project. the green heart tunnel was only the fifth tunnel in the world that was built using a slurry type tbm, but the aurora was unique because of its massive diameter of 14.87 meter, creating the biggest single tube tunnel in the world at the time. aurora would dig 2 meters and then tunnel segments of two meter length would be placed, after which the machine would dig for another two meters pushing itself forward on the placed tunnel segments. after some initial delays, the aurora was able to drill quite rapidly and made up for the time lost by drilling more meters per day (20 meters on average) than originally planned. as such the drilling process was concluded before the agreed deadline; which made it one of the most successful innovations of the project. figure 3 trajectory of niche-cumulation for breakthrough large slurry shield tbm wilson cutting discs (1853) complete front of rotating discs (1870s) costly larger front surface diameter (late 1800s / early 1900s) rotting cutting wheels, thrust system, trailing support mechanism small fronts for sewers, etc. (19th century) hard rock tbms soft ground earth pressure balance hi slurry shield machines open face type aurora with 14.87m diameter (2001) 56 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press 3.5 niche 3: safety and interoperability once the foundation and other infrastructures were completed a safety system that would fit the requirements of high-speed rail would have to be installed. such safety systems have been in development since 1901. in the 1950s and the following decades, innovations in radio-based signaling and innovations in traction technology that enabled faster trains, lead countries to develop so-called continuous cab signaling systems. at the end of 1980s, there were 14 different control systems in use across europe. these control systems all had different specifications and as such were incompatible. this meant that trains passing borders from one country to another needed to have multiple systems installed or change locomotives, the latter being the common option. this was impractical because international services were becoming popular. changes of locomotives cost time: multi-system locomotives were overly complex and not as well-developed as they are now. in addition, it is near impossible to read exterior light signals communicating track states at high speeds. in 1989, a working group including transport ministers created a master plan for a trans-european high-speed rail network, in which a european train control system (etcs) was to replace the incompatible national systems, was suggested. early 1990s, this led to the creation of the technical specifications for interoperability (tsi). the baseline for technical specifications were published and tested as part of the european rail traffic management system (ertms) by six railways since 1999. ertms is not a product but defines the tsi such that any company can develop ertms devices whilst maintaining interoperability. the idea is that equipment is mainly built into the train and less at the trackside. at levels two and three the communication between train, track and central post are communicated by gsm reserved for the railways. the first ertms specifications were published in june 2000, and have been amended since then to fit the needs of railway companies and administrators. 57 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press figure 4: trajectory of nichefor breakthrough european train control system while the building of the hsl-zuid is well on its way during may 2004, the minister announced that she opted for ertms (etcs level 2) as the only safety system for hsl-zuid even though that standard was not fully developed at that time. a year later, the minister announced that the completion of the complete hsl-zuid was delayed as the implementation and testing of ermts needed more time. it turned out that implementing ertms in the thalys, a train that would be operated by the joint venture of sncf, db, nmbs & ns, would take at least six months longer. the problem was bigger with the new high-speed train-sets that ns wanted to order. as long as the ertms specifications were still unclear, or constantly being updated (see section 3.7), no trains could be ordered. at the end of 2005 it turned out that etcs was causing problems for high-speed rail all around europe as it was delivered too late everywhere and that the system would often malfunction when implemented. the problems were mainly caused by operators and the european commission constantly making alterations to the ertms-specifications. the development and implementation of etcs has been greatly underestimated. automatic train stops trip arms (1901) radio technology (early 20th century) radio-based signaling systems, e.g. atp, lzb, atb (1950) etcs (1989) digital radio (80-90s) etcs – level 0 (1998) tsi (1993) gsm (90s) ertms (1998) etcs – level 3 (2014) etcs – level 2 (2007) etcs – level 1 (2000) 58 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press 3.6 niche 4: market demands the 1993 decision to build the shortest alignment through the green heart was partially based on the estimated amount of passengers who would be offered a proper alternative to flying or driving. hsl-zuid was seen as a proper competitor to short to medium distance air transport services because of the lower terminal times and the ability to reach directly into the city centres (rodrigue, 2013). however, the dutch government did not anticipate that the european open-skies treaty of 1992 would allow any airline to fly anywhere without a government approval. this regulation caused the entry of many low-cost carriers (lcc) in the european market. this unforeseen development had a major impact on the future of the hsl. at the start of the project the idea was that hsl-zuid would service 10 million passengers that otherwise would fly from schiphol airport. however, this idea became unrealistic due to companies such as easyjet and ryanair entering the dutch market not only via schiphol, but also via regional airports like rotterdamden haag airport. because these airports where in the vicinity of relatively large cities and because they also provided quick boarding times and offered good alternatives for short european trips, i.e. the market that was assumed to be the domain of hsl-zuid. in addition, it also became clear that the expected travelling times between amsterdam and brussels would be longer due to calculation errors on the belgian section of the track and because certain track upgrades between antwerp and brussels had been postponed until 2012. in 2004 mckinsey reported that due to these developments the passenger volume would probably be 17% to 32% lower than expected in the concession agreement from 2001. revenue service between amsterdam and rotterdam started on september 7, 2009, with operator ns using traxx locomotives and coaches, running at a maximum speed of 160 km/h because the dedicated v250 trains were still not ready to run. the travelling time between amsterdam and rotterdam was reduced by 20 minutes due to the almost straight route and the somewhat higher speed. once the thalys and v250-fyra would start servicing the travelling time could even be further reduced. however, ticket prices were significantly higher than regular train services as ns needed to get some returns on its exceptionally high concession fee; the price hike was 60%. as it turned out, passengers were not willing to pay the surcharge on top of the fare price for each single trip between amsterdam and rotterdam. consequently, ns lowered the prices of the tickets but that didn’t attract much more passengers. january 31st 2011 ns again lowered the surcharge because the service was now running at a loss and needed more passengers. also the regular train service between amsterdam and brussels was taken out of commission as the high speed rail service took over that service. all to no avail as passenger were still not eager to get into the more expensive high-speed rail. after the v250-fyra trains were taken out of service and no alternative high-speed trains available, the former slower service between den haag and brussels was restored using traxx locomotives with coaches. a very modest surcharge was asked for the 20 minutes travel time reduction. since then the amount of passengers on hsl-zuid has picked up. 3.7 niche 5: rolling stock the standard set by the shinkansen, which initially only had a high speed of 210 km/h, made it possible for manufacturers to evolve the technology for trains to run at 300 km/h and even higher. however, there are only a handful of manufacturers that are able to mass produce stable 59 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press platforms of high-speed technology for the same reason as there are few aircraft manufacturers: the development cost of new models are immense. the first decades after the launch of the shinkansen very few manufacturers developed and built high-speed trains. recently, more companies have entered the market. by spring 2002 the new concession holder started an international tender for high-speed rolling stock; i.e. 23 high-speed trains. the most important criteria were ‘maximum speed of 220 km/h’, ‘price per seat’ and ‘between 450 and 550 seats per train’. four manufacturers submitted a proposal; french alstom, canadian bombardier, german siemens and italian ansaldobreda. of these four only alstom and siemens had actually developed and built high-speed trainsets themselves. bombardier proposed regular locomotives made suitable for higher speeds. these three manufactures offered modifications of existing stock because designing a new train from scratch for such a small series would not make an interesting business case. ansaldobreda was the only manufacturer that offered brand-new stock. however, ansaldobreda had only built highspeed trains that were designed by other manufacturers, so this would be their first in-house design. in 2003, the minister of transport announced that she would hold ns to their contracted travel time between amsterdam and brussels. according to her calculations, the speed of 220 km/h was insufficient. these stringent requirements as well as the number of seats and the requested exclusive looks, defined a kind of train that didn’t exist at that time and one by one the manufacturers pulled out of the biding process. in the fall of 2003 two candidates remained: alstom with trains running at 220 km/h and ansaldobreda with trains running at 230 km/h. during negotiations close to the deadline ansaldobreda offered an alternative that supposedly could run at 250 km/h, i.e. v250. later that year, the concession holder decided that they would require fewer trains because of the much lower demand. therefore, they ordered 12 instead of the initial 23 trains. alstom pulled out of the tender because the order would be too small to justify the development costs. consequently, ansaldobreda won the order for twelve v250s. the order should be delivered by april 2006. the v250 was designed as an 8-car trainset, of which 3 coaches are used for first class accommodation, giving 127 first class seats out. the total train has a seated capacity of 546. to be able to operate on both dutch and belgian electrified networks the train was designed to operate on 3kv dc, 1.5kv dc and 25kv 50 hz ac overhead power supply. the actual delivery date was postponed several times due to various reasons such as the late decision to implement etcs-level 2. then there was also the fact that ansaldobreda had never designed a high-speed train before. the company had great difficulties in complying with safety requirements and with maintaining build quality and engineering standards. many components were (re)designed during the construction phase, and none of the trainsets were built completely equal. the inaugural service with the v250 took place on sunday december 9th, 2012. during the weeks to follow many trains experienced delays or simply didn’t enter service due to various defects and operating errors. the winter snow of early 2013 damaged the trainsets to such an extent that they were banned from the tracks by the dutch and belgian certification institutes. consequently, the old intercity services were reinstated as a temporary replacement for the v250fyra, while ansaldobreda indicated that it would require several months to fix all problems. the contract with the manufacturer was finally terminated by june 3rd, 2013, as they were not able to fix the problems. the trains have been returned to the manufacturer by september 2014. 60 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press 3.8 niche alignment in the short distance high-speed regime the five niches described above show mixed results. figure 5 shows the discrete contributions of the five niches to the outcome of the short-distance, high speed train travel in the netherlands. the solid lines mean successful innovations, while the dashed lines mean unsuccessful innovations. figure 5 regime buildup through niches public-private partnerships were relatively successful when it comes to the different dbfmo-contracts for building rails, catenary et cetera, and the design & construct contracts for the foundation. the former is very successful when it comes to the development of a new record breaking diameter for the tunnel boring machine aurora. however, the latter did significantly costs more as the builders were able to form cartels, exchange information and influence the bidding process. the tendering for the concession was a success as the state extracted a much higher price than anticipated. however successful for the state, it impacted the ticket price ns would charge and as such affected market demands. all in all the public-private partnership innovation had a mildly positive effect on the design and development of the infrastructure. the building technique is clearly successful as the aurora was able to dig faster than planned through difficult ground conditions and the green heart tunnel was finished before the market demands: 1 – modal split due to lcc (90s) 2 – travel time vs price (2011 rolling stock: 1 – design demands 2 – limited # manufacturers (2004) 3 – ertms problems (2007) safety & interoperability: 1 – ertms (1998) 2 – etcs (2007) short distance high-speed: hsl-zuid (1993-2013) building technique: 1 aurora (2001) 2 – pillared dedicated 25 kv ppp: 1 – dbfmo (2000) 2 – d&c (2000) 3 – concession (2001) 61 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press agreed deadline. the other techniques, e.g. a completely pillared track or long bridges, also contributed to the successful build of the track. the development of a european wide safety system took significantly longer due to many factors ranging from unclear definitions by the european commission to unstable software releases and equipment failure. that is, the ertms / etcs innovation delayed the building of both the track and the trains and significantly influenced the costs of building and operating hslzuid. the idea of using high-speed trains in order to affect consumer behavior and force a change in the modal split failed completely as travelers did not find the high-speed service compelling enough to pay a considerable surcharge in comparison to the price of car travel or regular train services. in addition, the unforeseen emergence of low cost carriers meant that travelers could now board faster, nearer to their homes at relatively low prices. as ns and the dutch government learnt: high-speed railways are not automatically considered superior to other modes of travel. lastly, the strict design demands as dictated by the concession, lead to a botched tendering process that drove ns into the arms of ansaldobreda and its untested, hurried proposal for the v250 trainsets. the problems with ansaldobreda were intimately connected to the problems with ertms problems and with a lax quality assessment on behalf of ns and by the certification institution. the trains arrived 6 years after the deadline, to be taken out of commission a couple of weeks later. ideally, niches all add up to a successful socio-technological innovation. in this particular case, it didn’t happen. the case very clearly demonstrates that the technological and social dimension are mutually dependent. it takes alignment of all niches – e.g. agreement over ermts and the technology to build trains – to move a step forwards. even though certain sociotechnological innovations were successful they were not properly aligned with policy innovations such as public-private partnerships in such projects. ultimately, ns had to be saved from bankruptcy three times by the dutch government in the years to follow. the current situation sees ns running make-shift services at a loss with a solution still being far away. 4. conclusion we started this article with the observation that classical policy evaluations are geared towards finding that one variable that supposedly controls for all the problems encountered. in the political arena, this often boils down to pinpointing and blaming a person or organization, implementing an extra set of rules, and to move on. while we don’t want to fully deny the idea that someone can just have committed a grave error, we do believe that major public projects that includes innovative technology, such as hsl-zuid, cannot be judged on the basis of simple premises about causes and effects. the case demonstrates that success of the newly developed short-distance high-speed railways depends on the alignment of underlying coevolving technological as well as social niches. misalignment of these niches means that the innovation at the regime level takes place in starts and fits. it may also mean that the innovation in one niche propels or hinders the innovation in another niche, making the innovation at the regime level fail. this is not to say that innovation doesn’t take place at all – after all, progress has been made since the first ideas to launch a high-speed railway service in the 1970s – but that progress is haphazard and did not take place according to the policy goals and planning. arguably, the sociotechnological innovation framework provides insights into why certain public policy innovations 62 complexity, governance & networks special issue: complexity innovation and policy (2017) 48-62 peter marks, lasse gerrits: evaluating technological progress in public policies: the case of the high-speed railways in the netherlands doi: http://dx.doi.org/10.20377/cgn-42 university of bamberg press were successful or not, without having to nail the failure to human error. that is, complexityinformed theories about socio-technological innovation as the cumulation of evolving niches can lead to a considerably advanced appreciation of the complexity of public policies. references bortolotti, b. & milella, v. (2006). privatization in western europe: stylized facts, outcomes, and open issues, fondazione eni enrico mattei. bressers, j. t. a., & hoogerwerf, a. (1995). het beleid en het beleidsveld. bressers, n. (2011). co-creating innovation: a systemic learning evaluation of knowledge and innovation programmes. van den brink, e. (2002). een pijl door het groene hart, de ingenieur (december): 22-26. ce delft (2003). to shift or not to shift, that’s the question: the environmental performance of the principal modes of freight and passenger transportation in the policy-making context. delft: ce solutions of the environment, economy and technology. clewlow, r.r., sussman, j.m. & balakrishnan, h. (2014). the impact of high-speed rail and low-cost carriers on european air passenger traffic, transport policy, 33: 136-143. doi: 10.1016/j.tranpol.2014.01.015. crok, m. (2002). dwars door de smurrie. de ingenieur (december): 27-35. decker, k. de (2013). high speed trains are killing the european railway network. low-tech magazine: doubts on progress and technology. http://www.lowtechmagazine.com/2013/12/high-speed-trains-are-killing-the-europeanrailway-network.html dekker, v. (1998) tunnel hsl ook door nood schiphol overbodig, trouw, 20 maart economist, the (november 2002). coming home to roost: privitisation in europe. url: http://www.economist.com/node/1200637 geels, f.w. (2002). technological transitions as evolutionary reconfiguration processes: a multi-level perspective and a case study. research policy, 31: 1257-1274. doii:10.1016/s0048-7333(02)00062-8. gerrits, l., & marks, p. (2014). vastgeklonken aan de fyra: een pad-afhankelijkheidsanalyse van de onvermijdelijke keuze voor de falende flitstrein. bestuurskunde, 23(1), 55-64. kemp, r., rip, a. & schot, j.w. (2001). constructing transition paths through the management of niches. in: garud, r. karnoe, p. (eds.), path-dependence and creation. lawrence eribaum, mahawah, nj: 269-299. kerseboom, j. (2008). leren van evalueren: over het niet-gebruik van beleidsinformatie. bestuurskunde, 17(2), 41-46. kim. (2009). na het knippen van het lint: het ex-post evalueren van infrastructuur. den haag: mnisterie van verkeer en waterstaat. koning, p. de. (2012). snelle en luxe trein ‘wordt me door de strot geduwd’. nrc handelsblad, 28 november. koppenjan. j. (2005). the formation of public-private partnerships: lessons from nine transport infrastructure projects in the netherlands. public administration, 83(1): 135-157. doi: 10.1111/j.0033-3298.2005.00441.x. lincoln, y. s., & guba, e. g. (1985). naturalistic inquiry (vol. 75). sage. mraz, s.j. (2011). the latest in high-speed train technology, machine design. http://machinedesign.com/news/latesthigh-speed-train-technology nelson, r. r., & winter, s.g. (1982). an evolutionary theory of economic change, 929-964. noordegraaf, m. & abma, t. (2003), ‘management by measurement? public management practices amidst ambiguity’, public administration, 81(4), pp. 853-871 parsons, w. (1995). public policy. cheltenham, northampton. phang (2009). transformation of the urban rail sector through ppp, journeys: 23-35. rip, a. & kemp, r. (1998). technological change. in: rayner, s. malone, e.l. (eds.). human choice and climate change. vol. 2., batelle press, columbus, oh: 327-399. rodrigue, j-p. (2013). the geography of transport systems (third edition). new york: routledge. teisman, g. r. (2005). publiek management op de grens van chaos en orde: over leidinggeven en organiseren in complexiteit. academic service. verweij, s. (2015). once the shovel hits the ground: evaluating the management of complex implementation processes of public-private partnership infrastructure projects with qualitative comparative analysis. http://www.lowtechmagazine.com/2013/12/high-speed-trains-are-killing-the-european-railway-network.html http://www.lowtechmagazine.com/2013/12/high-speed-trains-are-killing-the-european-railway-network.html http://www.economist.com/node/1200637 http://machinedesign.com/news/latest-high-speed-train-technology http://machinedesign.com/news/latest-high-speed-train-technology evaluating technological progress in public policies: the case of the high-speed railways in the netherlands public administration and complexity or how to teach things we can’t predict? complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 1-9 doi: http://dx.doi.org/10.20377/cgn-61 1 university of bamberg press public administration and complexity – or how to teach things we can’t predict? mary lee rhodesa, elizabeth eppelb keywords: complexity; pedagogy; public administration; education 1. introduction since the beginning of the 21st c, a group of public administration scholars have been researching complexity theory and its application in the public administration and policy contexts (koliba et al 2016, rhodes et al 2011, head 2010, morcol 2013). in spite of calls encouraging the application of complexity theory to practice (oecd 2009) and opposing claims that it would remain a ‘fringe interest’ (pollitt 2009), there has been little attention paid to how this research has translated into practice or into the teaching of public administration that would facilitate this translation. initial forays into literature regarding the teaching of public administration makes it abundantly clear that academics (and practitioners who published in academic journals) perceived a fundamental and growing gap between theory and practice in public administration going back decades, which was made particularly evident at the turn of this century in a series of articles in public administration review (bolton & stolcis 2003; newland 2000; stivers 2000). hartmann & raadschelders (2011) explain: "the general, and rather stereotypical, explanation is that of a growing gap between practitioners and academicians, with the former being interested in usable knowledge while the latter are concerned with advancing science (p.1)". closing this gap is an important area of interest for both scholars and practitioners. with this in mind, a group of complexity and public administration scholars convened a panel in irspm 2015 to discuss how complexity theory could help to close this gap and how they were or were not incorporating complex systems and complexity into their teaching. this panel turned into a 'call for inputs' from attendees and other interested parties in 2016 for information on how complexity ideas were being taught in the classroom. that call evolved into this special issue on teaching complexity to public administration students and practitioners – to put an initial ‘stake in the ground’ regarding how theory is beginning to be translated into practicerelevant learning. amary lee rhodes, university of dublin, ireland, e-mail: rhodesml@tcd.ie belizabeth eppel, university of victoria in wellington, new zealand, e-mail: elizabeth.eppel@vuw.ac.nz http://dx.doi.org/10.20377/cgn-61 mailto:rhodesml@tcd.ie complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 1-9 doi: http://dx.doi.org/10.20377/cgn-61 2 university of bamberg press we begin this editorial with the story of how one of the editor’s public policy experience and research led her to teaching complexity to practitioners and how respondents to our call for inputs came to a similar conclusion, albeit via different paths. the main body of this editorial is a reflection on the key points from the articles. they have inspired us to examine why & how we teach complexity and how the budding ‘community of practice’ of scholars and practitioners might evolve over time and impact on the practice of public administration. why teach complexity to public administration students? as noted above, the first attempt by the authors to solicit input from other complexity scholars on how they were or were not using complexity theory in the classroom was during a panel at irspm 2015 in birmingham, uk. the panel topic was "complexity theory in public administration: state of theory and practice" and the range of papers presented there suggested that although complexity theory had certainly made inroads into public management scholarship indeed it was being incorporated into a wide range of theoretical and practical investigations there was little convergence on which aspects of complexity theory were generating the most interest or had been tried and rejected as not relevant. furthermore, a facilitated discussion on the use of complexity theory in general in the classroom did not generate much enthusiasm and the panel chairs resolved to consider whether or not there was a critical mass of public administration scholars attempting to use complexity in the classroom given the still emerging state of relevant theory. the editors of this issue then examined their own experiences to discover key decision points in their journey to the teaching of complexity to public management students. one of us had decades of practical experience of policy development and public management and her frustration is described below: “i was acutely aware that theories of public policy and public management were limited for understanding the world in which i worked. generally they focused on only part of the world such as the economic efficiencies of the design of a funding system, or the ideal institutional arrangements, ignoring the interactions these changes would have on existing patterns and behaviours of other actors. the explanation these various extant theories provided was a static one that took little account of interactions among multiple actors, layers of institutions or reflexive changes over time. worse still, when it came to how to work with the multiple scales of this world from the individual to the organisation, the institutions and the system level, the theories usually only focused at one level. they left me as a practitioner to work out how change at one level interacted as a whole to produce the patterns we experienced.” leading on an education policy development project at the time, she encountered the cynefin framework (e.g. kurtz & snowden, 2003; snowden, 2001; snowden, 2002, 2005; snowden, 2010) which provided a sense-making device for examining information from a range of perspectives such as the different system views identified above and the various layers of organisations within each of those systems. a cynefin-based workshop for around 70 stakeholders was held and it became evident to participants that current patterns and pathologies in the education system had their origins in decisions made by various political, sector and policy actors in previous periods. taking these ‘path-dependencies’ into account made it both more likely that policy would have its intended effect and less likely that public administrators could http://dx.doi.org/10.20377/cgn-61 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 1-9 doi: http://dx.doi.org/10.20377/cgn-61 3 university of bamberg press assume that it would. in other words, monitoring and responding to feedback over time was just as important as the initial policy intervention choice. however, the recommendations arising from the cynefin-based workshops were largely ignored. the tertiary education policy interventions that ensued resulted in a number of surprises and not the result the minister intended (eppel, 2012). it also gave rise to phd research on how complexity theory might help us to better understand public policy processes and their implementation. by the time she was finishing her research on the tertiary education policy processes, the public policy and public management use of complexity theory as an explicit theoretical lens seemed to be everywhere (geyer & rihani, 2010; innes & booher, 2010; rhodes, murphy, muir, & murray, 2011; teisman, van buuren, & gerrits, 2009). since then, and assisted by forums such as compact (gerrits & marks, 2012; meek & marshall, 2014), special editions (e.g. teisman & klijn, 2008) and conference panels (e.g. irspm 2013 and 2015), the complexity scholarship has grown and more use of it has been made in theory. the combination of a critical mass of theoretical insights, personal experience and perceived need led her to incorporate the teaching of complexity to dgov (doctor in government) students at victoria university in wellington. specifically, she structured her approach to orienting complexity theory towards practice under the broad themes of policy processes and governance in particular the treatment of multiple actors, nonlinear interactions and dynamism over time. having re-examined our own journeys to adopt complexity for the study and practice of public administration, we issued a “call for inputs” in 2016 that asked scholars in our network seven questions about their teaching of complexity to public administration students. this call was designed to gather comparable data across the group, but also allowed for considerable elaboration in the responses as desired. of the 26 people contacted, ten responded to the call which was rather more than we had expected given the relatively low level of engagement with the topic just one year earlier. responses varied from short (sometimes one word) answers to full course syllabi and commentary on the effectiveness of the module and teaching approach. in the appendix to this editorial we have provided a summary of the call and the responses received. our respondents had similar motivations for incorporating complexity into their curricula as we did (see summary of motivation below), although some would have come from quite different starting points. problems in the real world were not getting solved and complexity appeared to offer a more aligned and comprehensive theoretical lens with which to perceive and develop solutions. combined with expanding research and communities of knowledge, this provided the scholarly and social impetus for a change in what and how we taught. overview of teaching complexity to public administration students the range of responses from this initial data gathering exercise was fed back to contributors and they were asked to contribute more substantive papers on their experiences with teaching complexity to public policy and management students. in addition, we received a proposal to undertake a search on the internet for higher education programmes incorporating complexity into curricula aimed at public administration students, the results of which follow this editorial. the process of writing reviewing and revising the resulting papers further enriched our understanding of the pedagogy of complexity teaching when we saw the variety of ways scholars http://dx.doi.org/10.20377/cgn-61 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 1-9 doi: http://dx.doi.org/10.20377/cgn-61 4 university of bamberg press had integrated their personal learning about complexity and tried to embrace it in their teaching. we summarise some of the key learning points below. level / location of students taught: complexity appears to be taught principally at the masters level as indicated in the survey results (ivanovic & gerrits) and in the contributions in this issue. however, we have aimed to include as broad a spectrum as possible in this issue and we have examples of teaching that include workshops embedded in professional / vocational education and training (raisio et al and van der cingel) through to executive education short courses (hazy); from engaging with the public through podcasts and ‘newspaper university’ programmes (raisio et al) to developing communities of practice involving scholars and practitioners to develop a toolkit for intervening in complex public systems (haynes et al). unfortunately, we were not able to include an article on the approach to developing and delivering a full-fledged complexity-led programme, but we are aware that these exist and hope in the future to publish more detailed information and analyses of these. ivanovic & gerrits report that countries with the most programmes (that could be found via an internet search) were in germany (14), the uk (10) and the usa (10), but that the spread of these programmes around the world was surprisingly broad. 53% of the programmes identified were in europe, however. see the ‘heatmap’ in figure 1 below. figure 1: “heatmap” of location of complexity programmes globally in 2017 motivation for incorporating complexity in the curriculum: the introduction of complexity into the curriculum is largely problem-driven, with reference to the ambiguity, unpredictability, ‘open-endedness’ of public administration policy and service problems. http://dx.doi.org/10.20377/cgn-61 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 1-9 doi: http://dx.doi.org/10.20377/cgn-61 5 university of bamberg press teachers and students perceive that the world in which public policy makers and administrators operate is not stable or controllable, and that traditional theories, models and management tools assume (to a greater or lesser extent) that it is. the inevitable failures that arise from this mismatch drives dissatisfaction with theory and the search for better models to fit the experience of practitioners and the empirical observations of scholars. there is also an undercurrent of psychological distress mitigation in the motivation for the teachers’ adoption of complexity perspectives and the students’ embracing of these by the end of the learning experience. raisio et al call this “easing the pain of dealing with a rapidly moving, constantly changing environment”. pedagogy: case studies referenced in the first round of ‘inputs’ gave way to more formal / informal problem based learning (pbl) approaches in the articles contained herein. gerrits & wirtz describe this in detail as the fundamental architecture of their pedagogical approach, and haynes et al reference pbl explicitly as informing their approach to developing a community of practice, interrogating case studies and developing the toolkit for systemic interventions. wolf’s use of ‘prospective case studies’ is an innovative combination of a traditional case study approach and problem-solving, underpinned by the application of complexity lenses. in several of the contributions to this issue there are specific recommendations of which complexity frameworks, models and theories work best within the learning objectives and pedagogies described. hazy incorporates a rich set of theoretical contributions into his executive education programme on leadership, while wolf summarises three useful lenses for students to apply to the public policy domain they encounter and seek to influence. in all of the papers, authors make reference to the need to shake up students’ belief systems (i.e., challenging their assumptions about public management systems and strategies) and to get students to consider a range of different perspectives on problems that are not easily analysed or categorised. this need to ‘activate doubt’ (wolf) in students is done in various ways, sometimes simply by encouraging participants to share their stories (hazy, haynes) – which tends to be effective with more experienced students – and other times by actively challenging students assumptions about how things work and how to intervene to achieve outcomes (wolf). raisio et al suggest that introducing students to the concept of ‘wicked problems’ (rittel & webber 1973) is a good way to “concretize the difference between the mechanical worldview and the worldview of complexity”. finally, there is a distinct pattern in the description of the role of the teacher in a complexity-informed classroom. raisio et al describe this role as being a “mediator between abstract theory and real-life phenomena”. this description resonates within each of the contributions to this issue as teachers help students understand better the world they live and work without prescription – just a different perception. this role is fundamentally linked with the role of facilitating a student-led learning process – which is clearly described in gerrits & wirtz, and is present in each of the contributions in this issue. assessment: in the initial call for inputs we were struck by the plethora of assessment approaches mentioned and this was borne out by the articles in this issue. while examinations do make an appearance, the range of assessment approaches is something to warm the heart of any 21st c pedagogue. action-learning tools such as journals, diaries, and reflective essays play a prominent role, as do issue, scenarioand ‘futures’ analyses of cases and current issues in public management. role-play, games and simulations are used, but it is not clear how these are or are not incorporated into formative or substantive assessment. the overall sense we get from these http://dx.doi.org/10.20377/cgn-61 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 1-9 doi: http://dx.doi.org/10.20377/cgn-61 6 university of bamberg press articles and the response to the original call is that assessment is not at the core of the planning, development and implementation of complexity-oriented education – and this may not be a bad thing. in fact, it may only be possible to assess students’ comprehension and/or ‘mastery’ of the concepts of complexity as they apply to public administration when they are faced with real issues in real-time. we are aware of a complexity-led masters programme in the netherlands that incorporates and internship-based masters thesis, as well as a case-based consulting type project for a ‘commissioning’ public organization along the lines of what is normally seen in experiential mba programmes. this would appear to be the most appropriate approach to assessing students’ facility overall with the concepts and practices of complexity-informed public policy development and management. learning outcomes / feedback: perhaps due to the broad range of pedagogies and assessments, there is little in the way of patterns in learning outcomes. this is an area that could benefit from further research and reporting. student feedback is also difficult to get a clear view of based on these articles although most report that students ‘value’ what they have learned and feel that the material is relevant to the practice of public management. in the case of haynes et al, the community of practice developed an action & intervention ‘toolkit’ that is an excellent example of how scholar-practitioner interaction can result in theory-based practice. many of the authors here report that students initially feel uncomfortable with either (or both) the theories and the pedagogy, but they tend to come around by the end of the module to the view that “teaching methods are well suited to the subject matter” (gerrits & wirtz: p. 43). echoing the observation made above, it is perhaps impossible to know the immediate learning impact of these courses on students or indeed on the organisations that employ them. conclusion this special edition began with the question, ‘how do we teach things we can’t predict?’ and along the way also answered the question, ‘why teach complexity?’. the answer to ‘why’ is firmly rooted in the challenges of increasingly wicked problems in the public sphere and the apparent alignment of the principles, processes and perspectives of complexity theory with the problem domains. the jury is out as to whether or not public administration professionals who are exposed to complexity theory in an academic environment are more likely to succeed and/or thrive when faced with such problems, but at least, for some, it “eases the pain of dealing with the rapidly moving, constantly changing environment, where nothing is for certain” (raisio et al: p. 85). we were both struck in reviewing the responses to the previous call and these contributions how many people had started alone in their search for explanations and had gradually moved towards like-minded colleagues to build communities of practice in both researching and teaching complexity. it appears that there is a need for a critical mass of committed scholars and teachers in close proximity to get a foothold in existing curricula. students and practitioners appear to like and appreciate the learning, however, we have little information as to the success of these courses within the wider curricula in which they operate. the high-level scan of programmes around the world suggests complexity elements are fairly marginal, but it would be instructive to understand if this is a growing or static proportion (ivanovic & gerrits). our view at this point in time, is that in order to further develop the teaching of complexity to practitioners, the process described by haynes et al seems to be the most viable, with the most likely audience being healthcare managers or urban planners / city managers. http://dx.doi.org/10.20377/cgn-61 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 1-9 doi: http://dx.doi.org/10.20377/cgn-61 7 university of bamberg press complexity theory is making headway – but still struggles with credibility in the face of more linear, quantitative, hierarchy based theories–in the academy and even more so in practitioner’s world. nevertheless, the evolution of theory into the classroom is underway and may be gathering steam. we need to undertake more focused research to understand the full cycle of relevant teaching and learning and to examine how (if) this teaching is translating into practice more explicitly. finally, we need to assess whether those students who go into public administration or work in organisations that are engaged in delivering public services are using complexity-informed approaches successfully in a world where uncertainty and path-dependency are the norm. this is a massive research challenge and will require ongoing collaboration among scholars, practitioners and governments to address effectively. references bolton, m. j., & stolcis, g. b. (2003). ties that do not bind: musings on the specious relevance of academic research. public administration review, 63(5), 626-630. boulton, j. g, allen, p. m. , & bowman, c. (2015). embracing complexity: strategic perspectives for an age of turbulence. oxford, uk: oxford university press. eppel, e., gill, d., lips, a. m. b., & ryan, b. (2008). better connected services for kiwis: a discussion document for managers and front line staff on better joining up the horizontal and the vertical. wellington: institute of policy studies, school of government, victoria university of wellington. eppel, e. (2009). the contribution of complexity theory to understanding and explaining policy processes. (doctor of philosophy), victoria university of wellington, wellington. http://hdl.handle.net/10063/1202. eppel, e., turner, d., & wolf, a. (2011). complex policy implementation: the role of experimentation and learning. in b. ryan & d. gill (eds.), future state: directions for public management in new zealand. (pp. 182-212). wellington: victoria university press. eppel, elizabeth, & lips, miriam. (2016). unpacking the black box of successful ict-enabled service transformation: how to join up the vertical, the horizontal and the technical. public money and management, 36(1), 39-46. doi: 10.1080/09540962.2016.1103417 gerrits, l., & marks, p. (eds.). (2012). compact 1: public administration in complexity. litchfield park, az: emergent. gerrits, lasse, & marks, peter. (2015). how the complexity sciences can inform public administration: an assessment. public administration, 93(2), 539-546. doi: 10.1111/padm.12168 geyer, r., & rihani, s. (2010). complexity and public policy: a new approach to 21st century politics, policy and society. abingdon, uk: routledge. hartmann, j., & raadschelders, j. (2011). practitioners and academicians: about gap, authority and bridges. pa times, (august/september), 25. head, brian william (2010) "public management research", public management review, 12:5, 571-585 http://dx.doi.org/10.20377/cgn-61 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 1-9 doi: http://dx.doi.org/10.20377/cgn-61 8 university of bamberg press innes, j. e., & booher, d. e. (2010). planning with complexity. abingdon, uk: routledge. koliba, c., gerrits, l., rhodes, m. l. & meek, j. w. (2016). complexity theory, networks and systems analysis. in handbook of theories of governance, cheltenham, uk: edward elgar kurtz, c. f., & snowden, d. j. (2003). the new dynamics of strategy: sense-making in a complex and complicated world. ibm systems journal, 42(3), 462-483. meek, j. w., & marshall, k. (eds.). (2014). administrative stractegies for complex governance systems: challengies of making public administration and complexity theory work. litchfield park, az: emergent publications. newland, c. a. (2000). the public administration review and ongoing struggles for connectedness. public administration review, 60(1), 20-38. oecd. (2009). applications of complexity science for public policy: new tools for finding unanticipated consequences and unrealized opportunities. global science forum. retrieved: 19/10/2016: https://www.oecd.org/science/sci-tech/43891980.pdf pollitt, c. (2009). complexity theory and evolutionary public administration: a skeptical afterword. in: teisman, g., buuren, a. van, & gerrits, l. (eds.). (2009). managing complex governance systems: dynamics, self-organization and coevolution in public investments. new york ; london: routledge rhodes, m. l., murphy, j., muir, j., & murray, j. a. (2011). public management and complexity theory: richer decision-making in public services. new york: routledge. rittel, h. w. j. & webber, m. (1973). dilemmas in a general theory of planning. policy sciences 4, 155–169. snowden, d. j. (2001). narrative patterns. the perils and possibilities of using story in organisations. knowledge management, 4(10). snowden, d. j. (2002). complex acts of knowing: paradox and descriptive self-awareness. journal of knowledge management, 6(2), 100-111. snowden, d. j. (2005). multi-ontology sense making. a new simplicity in decision making. management today yearbook 2005. snowden, d. j. (producer). (2010, 3 july 2014). the cynefin framework. retrieved from http://www.youtube.com/watch?v=n7oz366x0-8 stivers, c. (2000). “an impossible job”: report on the building bridges tour. public administration review, 60(3), 194-199. teisman, g., & klijn, e.-h. (2008). complexity theory and public management: an introduction. public management review, 10(3), 287-297. http://dx.doi.org/10.20377/cgn-61 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 1-9 doi: http://dx.doi.org/10.20377/cgn-61 9 university of bamberg press appendix: ‘call for inputs’ and summary of responses edited text of call for inputs: as part of our special issues in public management review (2017) and complexity governance & networks (2018) on complexity in public administration, we would like to gather data from our network about if and how researchers in complexity teach their students about complexity. in order to participate in this project, we would ask that you respond to the following questions if you have in the past, are currently, or are planning in the near future to teach complexity and/or complex systems to a group of students: • what led to your decision to teach complexity and/or complex systems? • at what 'level' did you teach this topic? • what part does the complexity component play in the curriculum? • what complexity and/or complex system concepts do you cover? • what pedagogical techniques do you use to teach complexity, e.g., case studies, simulations, role play, 'chalk & talk', etc.? • please describe the students' role in the learning process and how you assess their learning outcomes. • what sort of feedback have you received from students and/or colleagues about the material taught, either during the course or perhaps subsequently? why teach complexity: a mix of problem-driven reasons and being exposed to complexity in msc/phd studies "i want my students to be better prepared to select, adapt and justify policy [choices]." teaching level: primarily msc/mpa, but may also be taught in final year ug, phd and professional doctorate levels. which concepts / role in curriculum: huge range in curriculum role from the basis for a full masters programme in "governance and management of complex systems" to a brief introduction in a lecture (or two) to highlight how a complexity perspective might frame analysis. complexity content is equally diffuse difficult to discern a pattern across respondents. pedagogy / assessment: case studies are prominent in the pedogogy described. students are encouraged to engage with the material early on and find their own way through it. readings are different across every respondent/module. students demonstrate their understanding through class discussion and essays often related to the case studies, but not exclusively. there were one or two respondents who used simulations / role-play / field work. less use of 'chalk & talk' and exam-based teaching/assessment than would traditionally be seen in a taught masters (imho). feedback / other: students were reported as being uncomfortable at first with the content / teaching style(s), but over time (if they stayed in the module) they came to like/appreciate the material and its relevance in the 'real world'. comments include: "they either love it or hate it"; "students are eager to learn about ways of thinking that can move us beyond traditional strategies that are not working"; "the framing of issues is so helpful why has no one ever introduced us to these ideas before?" http://dx.doi.org/10.20377/cgn-61 developing public service knowledge and learning about complex systems: using a community of practice to integrate theory and practice complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 46 university of bamberg press developing public service knowledge and learning about complex systems: using a community of practice to integrate theory and practice mary darkinga, philip haynesb, julia stroudc abstract: a community of practice knowledge exchange (cpke) comprising of practitioners and academics with an interest in complexity theory was formed. the central activity of the cpke was an agreement that each would develop a case study that integrated complexity theory and challenges in practice. in the first phase, members agreed definitions of core concepts and used them to describe the systems they were working in. frequent contact was via a virtual learning environment. this activity was supplemented with face-to-face contact, in the form of half-day workshops. after several months, the cpke agreed two core pieces of reading to provide a theoretical basis. these were popular because of their applied focus. the cpke evolved to a second phase. diverse case studies of practice-based challenges were used to share experiences of complex systems. a key event was the presentation of case studies as complex systems using a shared conceptual language generated by earlier learning. in the final phase, the cpke considered interventions into the case studies and created and applied a management toolkit to develop approaches to possible management interventions. through the development of the toolkit, the cpke became interested in the role of values in a complexity-based practice and how coherent and shared values could aid more informed interventions. the use of complexity theory changed the ontology of management practice by facilitating an understanding and acceptance of uncertainty and that optimal approaches often required a relational and cooperative approach built on shared values, rather than an instrumental and singular management orthodoxy. keywords: communities of practice; knowledge exchange; blended learning; complexity theory introduction complexity theory had been taught by several staff in the school of applied social science at the university of brighton since the late 1990s, across a diverse range of courses, including public policy and management, social policy, and social work. the main early focus in this learning and teaching was on the relevance of complexity theory to forms of social practice, including public management, social policy, and social work. content was first developed for a postgraduate certificate in public service management, from 2003, and – in the same decade the ma public service partnership. these qualifications were replaced by an mpa in 2007. amary darking, university of brighton, großbritannien, email: m.l.darking@brighton.ac.uk bphilip haynes, university of brighton, großbritannien, e-mail: p.haynes@brighton.ac.uk cjulia stroud, university of brighton, großbritannien, email: j.stroud@brighton.ac.uk http://dx.doi.org/10.20377/cgn-45 mailto:m.l.darking@brighton.ac.uk mailto:p.haynes@brighton.ac.uk mailto:j.stroud@brighton.ac.uk complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 47 university of bamberg press in 2007, several staff engaged in this activity began to meet specifically to share their current experiences of using complexity theory in learning, teaching, and research. at this stage curriculum interventions were largely focused on occasional specialist lectures and seminars on generic courses, these teaching sessions offering an alternative and critical perspective and highlighting some of the important new theoretical texts becoming well known at this time (byrne, 1998; cilliers, 1998). several staff in the school had started to present papers and to publish in this area (haynes, 2003, 2008; stroud and warren adams, 2013). resulting discussions in the local academic community included mapping the emerging theoretical framework. a connection was also made with a colleague at the brighton medical school (bms) (see price at al. 2015). a small community of practice (wenger, 1998), at first entirely based in the local academic community, was established to share experiences of the use of complexity theory and how it was informing curriculum and research developments. a community of practice has been defined by wenger (1998) as a group of likeminded people who share an enthusiasm and commitment for elements of their practice and want mutual support to improve the performance of their practice. communities of practice often experience a strong sense of shared professional values. the group were also influenced by problem-based learning (pbl) as this had been taught and practiced by the academic staff with an interest in complexity theory. problem based learning is a learning participant centred approach to teaching and learning (savin-baden, 2000). participants learn by sharing open-ended problems, often through real world case studies, and seek to address issues by using a variety of information content alongside their own and other’s experience. in 2009, haynes led a bid for development funding, on behalf of this small academic group. this was in response to a call from the uk esrc to bid for funding for knowledge exchange. the uk research council was seeking to fund work that could articulate the scope of areas of new academic theoretical work that was beginning to have public impact, including in areas of social and professional practice. the call was an ideal opportunity to take forward the aspirations to do more with integrating complexity theory with innovations in practice. in late 2010, an award was made to fund a project at the university of brighton, entitled: systems and complex systems approaches in public policy and practice: a knowledge exchange between academics and practitioners (res-192-22-0083). this project provided an important pilot for action learning between academics and public managers and professions. the ideas learnt were pivotal in developing aspects of the curriculum on the newly evolving mpa. aims of the project the funded project aimed to create a more substantial community of practice whereby academics would meet equally with social practitioners. this was to be a forum for collaboration in policy analysis and public management development. building on the social learning theory and practice of wenger (1998), it aimed to take place in an equally shared professional space, occupied by both academics and practitioners. by enabling practitioner input to the critical reflection about complexity theory and its application, academics sought to harness theoretical insights, identify policy, and practice solutions, and devise best practices for learning, teaching, and research. the outcomes were to improve academic and practitioner understanding about complexity theory, to develop working tools for public policy makers, managers, and professionals, to provide specific developments for the higher education public policy and management curriculum, and to enable wider national and international dissemination of findings when the project finished. http://dx.doi.org/10.20377/cgn-45 https://app.researchfish.com/awards/viewdetails/0?gorderby=organisation&filter=esrc-es/h04454x/1 https://app.researchfish.com/awards/viewdetails/0?gorderby=organisation&filter=esrc-es/h04454x/1 https://app.researchfish.com/awards/viewdetails/0?gorderby=organisation&filter=esrc-es/h04454x/1 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 48 university of bamberg press a post hoc evaluation of the funded project identified that learning in the community of practice knowledge exchange (cpke) had taken place in three phases: formation and induction, using theory to describe policy and practice, and using theory to intervene in policy and practice. beyond this was the wider and long-term impact of the project, in terms of its influence on university teaching, research, and consultancy. it is important to note that these phases were not planned into the project, ex ante, but emerged through the participatory approach to facilitating the community of practice. phase 1 formation and induction the academic and community partner team responsible for winning the grant funding were the founding members of the cpke. further participants were recruited through an external website managed by the university's community university partnership programme (cupp) and via direct requests for practitioner participants from public bodies that had supported the project concept. this included reference to the support available such as covering expenses and providing reading and information support via university library and information services. those expressing an interest were contacted by the project lead to make sure they had a full understanding of the project’s aims and what would be the responsibilities of a member who joined. given the degree of commitment required to join, this meant that some interested parties did not feel they had adequate time to give to become members. twenty-four participants were successfully recruited. they represented public service providers, voluntary and community organisations and university staff/research students. a good balance of academics and practitioners was achieved. once new members with no formal link to the university had been enrolled as community academic fellows, to give them access to the necessary information systems, the community was closed to outsiders, and all those enrolled invited to an induction event. several members identified themselves as both academics and practitioners with dual professional roles. the duality of these members varied from those who were working in two, part-time jobs, one as an academic and one as a practitioner, to those whose major role was one or the other; for example, a practitioner who was also an occasional visiting lecturer at the university. the large number of these overlapping roles meant that that any simplistic division of academic and practitioner was far from the experience of most members of the cpke and not usually presented as a duality in the day to day experience. for example, those with major practitioner roles often made significant and leading contributions to the conceptual definitions, but perhaps with a more example-based approach, rather than starting with an abstract definition of a concept. the cpke agreed their emergent and evolving learning and development process through a process of active participation. at the face-to-face induction, all participants were encouraged to explain their interest in complexity theory and to indicate what they hoped to gain from being involved in such a developmental and learning process. all were expected to join a closed virtual learning environment managed by cupp. the induction included training into use of the virtual leaning environment and opportunities to try it out with specific exercises. each person who joined the community of practice agreed at the outset to develop a case study where they would attempt to integrate theory and practice. it was also agreed that the final case studies were to be presented at a closed conference that included an evaluation of the project. http://dx.doi.org/10.20377/cgn-45 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 49 university of bamberg press during the first phase of the project, members shared ideas for reading and gradually agreed on some core reading. the reading activity was used to facilitate an online process whereby members shared core concepts from complexity theory, and sought to identify examples from their research and practice and then to agree core definitions. at the end of this phase, two pieces of reading that had been used in the process were agreed as ‘most influential’ and conceptual definitions used in phases 2 and 3, largely came from these two sources (meadows, 2009; snowden and boone, 2007). this first phase of the project had rapidly established a closed community of practice in a three-month period and this enabled it to provide a knowledge exchange (ke) for sharing ideas about systems and complex systems between academics and practitioners. this was the beginning of reflection on how theoretical ideas were best integrated into practice. the cpke had some initially difficulty in agreeing abstract definitions of key complexity concepts, but when the cpke moved to consider descriptions around real living systems, the definitions became focused on working examples. seeing the practitioners as equal learners enabled the academics to focus on conceptual definitions that were not too abstract but rather grounded in real world examples. phase 2. using theory to describe policy and practice in the second phase, the ckpe sought to describe public and voluntary organisational systems they were involved in, using the concepts that had been explored and defined in the earlier community activity. each member was asked to present a case study from policy or practice and then to describe it as a complex system using the agreed theoretical framework. a face-to-face learning day was used for each person to share their case study and to gain peer assistance in constructing it as a complex system. the case studies selected by the participants were diverse. they included macro policy issues like the economy and public finances, and national energy policy, but also local examples of specific provider services in both the statutory and nongovernment sectors. examples were how regional and local government could best support changing local dynamics in tourism, how social workers best respond to risk in partially closed family communities, and how an urban community centre could adapt to meet growing and changing social needs caused by government and economic failures. all participants used diagrams to explain their system case studies. although no fixed diagrammatic methods were prescribed by the project, it was apparent that the diagrams developed and presented used similar symbolic representations to each other. for example, lines with arrows to show the flow of processes and activities, and boxes to indicate fixed point events or groups of actors (like parts of an organisation or team). the scale of the diagrams varied considerably depending on whether the participant was interested in macro, meso or micro level system behaviour. some diagrams were monochrome while others used the addition of colour. these diagrams bore similarities to those used by historical soft systems approaches to management decades earlier, but no direct connections had been made by the cpke with this field of publications (e.g. checkland and scholes, 1990). phase 3: using theory to intervene in policy and practice in the third phase, having shared the case study descriptions, the participants agreed to work on implementation plans for their systems. in effect, they would map possible interventions onto http://dx.doi.org/10.20377/cgn-45 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 50 university of bamberg press their system descriptions and consider with the cpke the likely implications for attempting these different interventions. one early group exercise was used to consider a hierarchy of management interventions. the most influential text for this development was meadows (2009). from this, the cpke identified seven core aspects of management intervention: resources and their use, identifying change, using information, setting rules, understanding self-organisation, directing the organisation by giving it purpose and radical change based on strong core shared goals and values. the community decided that the values and realism of a complex system approach to management should result in a radical model that reversed the likely focus and ordered use of these interventions, as indicated by the direction of the arrows and the order of the management actions in table 1. therefore, influenced by meadows’ (2009), observation about the transformational impact of changing values, and strategic goals clearly aligned with values, there was an appreciation by the cpke that culture and values are deeply ingrained elements of public organisational systems and that there are often the hardest aspects to change. the presentation of the toolkit (table 1) invites users to ‘turn the kit on its head’ and to start with cultural interventions from the ‘bottom up’. building team and organisational cultures and making them resilient through adaptability was argued to be at the core of a management practice that uses the insights of complexity theory. this can only be done by an interactive process whereby managers and leaders are constantly communicating and sharing messages with all staff. several members have since presented the toolkit at conferences and similar, and it has influenced other publications (price, et al, 2015). an agreed and popular method for mapping interventions onto the existing descriptive system diagram was to use post-it®sticky paper notes. each note represented a different idea for an intervention. the ideas for interventions were much influenced by meadows (2009) ideas about how best to intervene in systems. popular examples were: the use of stocks and flows of resources (including human resources); the use and presentation of information into specific parts of the system (to attempt to increase certain types of relational feedback); and modifying system rules. also popular was encouraging reinforcement of what was seen as helpful and functional self-organisation in one part of the system, by providing reinforcing information about it into another part of the system. there was also some more fundamental discussion about how to alter the overall trajectory of the system by changing its values and objectives, but given the macro and holistic aspect of this, it was difficult to summarise such possible transformations on the diagrams. on the final meeting day for the cpke, when each of the case studies were presented, a rich variety of interventions were in evidence, with much discussion and debate among the cpke members about the likely possible outcomes of such interventions, if they were to go ahead. the cpke agreed to publish the final toolkit, to make it available for future education, training and consultancy, and to promote the impact of the ke via an externally facing website (https://www.brighton.ac.uk/ssparc/research-projects/the-brighton-systems-knowledge-exchangeproject-.aspx). the complex systems toolkit was designed to assist public policy and management students in developing an integrated approach to public management that is systemic in its approach. table 1 summarises the toolkit (the full package is available at https://www.brighton.ac.uk/ssparc/research-projects/the-brighton-systems-knowledge-exchangeproject-.aspx). http://dx.doi.org/10.20377/cgn-45 https://www.brighton.ac.uk/ssparc/research-projects/the-brighton-systems-knowledge-exchange-project-.aspx https://www.brighton.ac.uk/ssparc/research-projects/the-brighton-systems-knowledge-exchange-project-.aspx https://www.brighton.ac.uk/ssparc/research-projects/the-brighton-systems-knowledge-exchange-project-.aspx https://www.brighton.ac.uk/ssparc/research-projects/the-brighton-systems-knowledge-exchange-project-.aspx complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 51 university of bamberg press the project evaluation at the conclusion of the project, there was an evaluation conducted by all those involved. members expressed a desire to continue their learning and those not based in the university were concerned about how to make the online learning environment simpler to use and more intrusive (to remind members to participate). some felt that it needed to be connected to a personal e-mail address as a more regular aide memoire. some felt that if the same kind of community of practice was attempted in the future the emphasis needed to shift more from online learning to additional face-to-face learning sessions. practitioners were keen to have occasional invites to share their knowledge at the university, for example as visiting lecturers and to some extent, this proved to be possible. all felt that the project needed to be simply recorded and presented as an external website and this was achieved. the involvement of practitioners in the action of learning about theory was an essential element to the success of the project, it moved beyond theoretical debates about abstract concepts to real world policy and management issues and decisions. a key part of the success was seen to be that the phases of the cpke were participatory and emergent, in that every three months all members reflected on the direction of the project, its progression and how it could ensure it achieved its final goals. a historical analysis of the working of the cpke identified the three key stages in the process, but these were not planned. they emerged through the participation. this emergent structure and process was closely linked to the overall demands of the project and the expectations that were set at the induction phase and the role of the principal investigator and project administrator in reminding the participants of the importance of the focus of the original goal to integrate theory and practice and to develop case studies. http://dx.doi.org/10.20377/cgn-45 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 52 university of bamberg press table 1 the action and intervention toolkit radical ‘public value’ approach a. resources and their use inputs and their use and flows in the system identify any blocks in system flows build stocks and reserves as buffers develop resources (including staff development) analyse workflow and work process [see seddon’s (2008) work on demand failure] b. identify types of change feedback that reinforces itself, often exponentially feedback that has implicit or explicit controls on it unmanaged change versus managed change c. use of information information about the system, its analysis, and ‘active ‘ or ‘passive’ communication back into the system d. setting the rules explicit rules and their effects implicit rules and their effects subversion of ruleseffect of deciding to change the rules traditional management approach http://dx.doi.org/10.20377/cgn-45 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 53 university of bamberg press table 1 (continued) the action and intervention toolkit radical ‘public value’ approach e. empowering self organisation understanding the power of self organisation, its functional and dysfunctional effects how can we empower the best aspects of selforganisation? f. directions and purpose the ability to generate shared goals, aims and objectives as powerful tools of change the link with values and cultures g. radical change building core and primary values that enable a critical sense of the priority issues that need action the ability to generate shared goals, aims and objectives as powerful tools of change the link with values and cultures building core and primary values that enable a critical sense of the priority issues that need action traditional management approach http://dx.doi.org/10.20377/cgn-45 54 university of bamberg press complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 discussion and conclusion formation and induction of the cpke while the funded cpke included a strong online element to enhance regular and flowing communication, it also included some clear stages for the community to regroup around face-toface contact. these events proved especially important in critically reflecting on the community’s own learning process and agreeing how it should progress. the evaluation provided strong feedback evidence that this was a collective shared view and that if the community had been constructed entirely online, the full dynamics and momentum of the community and its learning progress would not have been possible. this provides evidence of the importance of blended learning and the limitations of distance learning that is entirely online. the latter often experiences problems with motivation, commitment, and progression, as is well documented (bach et al, 2007). using theory to describe policy and practice an important aim of the funded knowledge exchange was that it required an approach to theory that was grounded in practice-based case studies and therefore assisted in moving the academics from a preoccupation with a debate about the definition of abstract and meta theory. previous scholarly work in the application of complexity theory to the social sciences had focused on issues relating to the importing of theoretical ideas from the physical sciences into the social sciences and the validity of the resulting theoretical framework and concepts (byrne, 1998; klijn, 2008; teisman & klijn, 2008; teisman, van buuren & gerrits, 2009; byrne & callaghan, 2013). there had been (and continues to be) some concern about the reliability and consistency of complexity concepts in the social sciences with different definitions sometimes being applied in different disciplines and circumstances. at first, there was a lack of empirical and practice-based work to illustrate the usefulness of the theory. in part, this was because of the meta theoretical nature of complexity theory and its holistic and integrated approach to social sciences. the requirement built into the funding bid that the closed community of practice consist of an equal focus on theory and practice and academic scholarship and public practice ensured that the scholarly work had to articulate its theoretical narrative into a real-world setting. the project ran concurrently with a developing international use of complexity theory in public management that increasingly focused on an organisational systems narrative and this was an important influence on the direction that the project and its later influence on curriculum took. the turn to complex systems (byrne, 1998; cilliers, 1998) in the applied social science approach to complexity theory provided more emphasis on practice (klijn, 2008; rhodes, et al, 2011), and loosely connected the complexity approach with a range of contemporary approaches to applied systems theory in management (seddon, 2008; rhodes et al, 2011; allen, et al, 2011). using theory to intervene previous public policy scholarly work (hill, 2005) had argued the importance of a theoretical framework in public policy and administration that could clearly separate description and http://dx.doi.org/10.20377/cgn-45 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 55 university of bamberg press prescription. it is one thing to be able to define a complex policy and organisational environment, in order to categorise it and make sense of it. it is another matter to be able to intervene in this complexity with some degree of confidence and to claim that one at least in part knows what one is doing. of course, the need for humility in intervening in complexity is one of the primary insights to have emerged in the turn to complexity and complex systems, and this was certainly a feature of reflections in the community of practice documented here (klijn, 2008; etzioni, 2014; byrne, 2014). these insights about the use of theory in practice proved a critical part of the discussions in the project cpke in phase 1 when it was seeking to get beyond debates about the boundaries of complexity theory and the nature and definition of its primary concepts. clarity around the differences between adequate description and reflection as contrast to the challenges of intervening in working public system were pivotal in moving into the second and third emergent phases of the learning community process. the conclusion of the learning here was not negative or overly cautious, nor indeed was it that ‘nothing works’. rather, it was that intervention is as much of an art as a science and that how things are done, is of as much importance, as what is done. relational and communication issues were often expressed as central in complex systems although these were difficult to capture in conventional approaches to organisational and systems based diagrams and models. the ultimate test of theorising complexity in public policy and management is the usefulness of the concepts to the art of practice. the academic members were pleased by the ability of the community in the final phase of the cpke to articulate together the integrative management model in the toolkit (table 1). this provides evidence that a complex systems approach to public management can be juxtaposed in relatively straightforward ways with some existing management approaches, especially those that are integrative and abstract by nature of their meta approach (i.e.: the management of change, and strategic management). perhaps, what was more surprising and radical in the development of the toolkit, was the desire to place public values at the centre of public management and to see values as a major element at the core of any public policy system. coherent values were seen as essential to such a system if it was to be functional and have operational chances of ‘success’, in terms of improving the quality of public life. this element is more novel and distinct when an integrative complex systems approach to management is compared to other major integrative approaches like managing change and managing strategy. while values may well play a role in these latter practices and how they are modelled (for example, values are at the centre of visionary approaches to strategy) they are by no means always at the centre of previous integrative approaches. the desire for values to be the kernel of a complex systems model of practice was perhaps the most interesting outcome. in this community of practice experience, this was influenced by the popularity of the meadows (2009) approach with practitioner members, and the fact that meadows’ original systems synthesis of environmental issues was values based. for example, one of the partners who joined the cpke was a service manager at an adult learning organisation based in a socio-economically disadvantaged area of their city. they came to the knowledge exchange with a case study concerning how to manage the multiple requests for data they received from funders (a situation familiar to many non-government organisations). their main concern was how to maintain the dignity of clients who were required to answer multiple, often personal questions in order to satisfy funder data requirements. the data requested was of little use to the organisation; it served their funders monitoring requirements http://dx.doi.org/10.20377/cgn-45 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 56 university of bamberg press but did not constitute helpful feedback for them. using meadows (2009) concepts, they identified requests for data as an excessive flow of information. in order to intervene in this situation they identified the need to assert their value position to funders which they described as 'evidencing what we do in a dignified way'. affording their clients dignity became a value position around which they sought to reorganise information management needs in their organisation. the practice discourse of primary interest to the cpke approach could be summarised as: an acceptance of the limitations of a mechanistic approach to cause and effect management approaches, and an appreciation of peer approaches to reflective decision making. this was a collaborative approach to dealing with complex issues. it maps with what snowden and boone (2007) have called the ‘complexity domain’. while the dynamic values influencing public managers at the cognitive and micro levels are in themselves complex and changing (box, 2015), they also interact with a more instrumental meso and macro holism of applying public value to defined policy outcomes (moore, 1995; 2013). these two approaches, while originally defined separately in the literature, are also increasingly discussed and debated together and argued to be connected (haynes, 2017). the nature of these connections and how they evolve in policy and practice over time needs more empirical research, but such a framework of entangled cognitions and cultures is itself a manifestation of a complex social system. conclusion in the subsequent years that the learning method here and the toolkit have been applied to the mpa, values have remained a central and important element to student participants. this influence of values has in the last decade been commensurate with a turn to public values in public administration scholarship, and identifying and specifying the diverse public values of practitioners in the public services box, 2015; bryson et al, 2016; haynes, 2017). during the cpke and in subsequent teaching practice, the use of complexity theory in public management moved somewhat away from a science-based method towards a philosophy of practice. here the approach is less about diagnosing what multiple methods can best detect the complexity of process and service production, but has become much more about actors acknowledging together the wide-ranging challenges and instability of the world in which they operate and that relational and supportive practices are needed that address the scale of these complex challenges. our experience in working with practitioners to solve complex problems in the public domain is that an appreciation of complexity theory and its concepts drives a change in perspective. together we came to accept that an element of uncertainty is inevitable, but to realise that it is the product of our dynamic, interactive, and innovative society. this results in a rather different ontological perspective to the previous historical approaches of rational and predictive management science. a key challenge after the conclusion of the funded and closed community of practice was how to ensure the learning was appropriately translated to other settings. discussions about the most appropriate place to include it, in a contemporary mpa curriculum, focused around the interactive action learning benefits of the process experienced by the funded community of practice. http://dx.doi.org/10.20377/cgn-45 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 57 university of bamberg press the three phased methods identified by post hoc evaluation of the cpke were subsequently implemented in the university of brighton mpa where it became a core element of the module strategy and planning in the public and voluntary sectors. it has run numerous times in this context and has evaluated well by students on each occasion. in particular, students liked the integrative nature of the exercises and saw it as a good evaluation tool to use before considering new strategies and change processes. the method was also influential on the development of the second edition of managing complexity in the public services (haynes, 2015) and in particular the chapters on performance and strategy. the toolkit has been further articulated and explained in price et al (2015). acknowledgements this project was funded by the esrc grant res-192-22-0083, systems and complex systems approaches in public policy and practice: a knowledge exchange between academics and practitioners we would like to acknowledge the contribution to the learning in this article as contributed by members of the brighton complex systems cpke and students on the brighton mpa. in particular, ceri davies, jim price, carla ricaurte quijano, chris warren-adamson, and david wolff. our thanks also to the reviewers of the earlier versions of this article for their helpful critique and ideas and assisting the development of the article. references allen, p., maguire, s., mckelvey, b. eds. (2011) the sage handbook of complexity and management. london: sage. bach, s., haynes, p. & lewis smith, j. (2007) online learning and teaching in higher education. maidenhead: open university press. box, r. (2015) public service values. new york: routledge. bryson, j., alessandro, s., bennington, j. & sorensen, e. (2016) towards a multi-actor theory of public value co-creation. public management review, online (http://dx.doi.org/10.1080/14719037.2016.1192164), pp. 1-15. byrne, d. (1998) complexity theory and the social sciences – an introduction. london: routledge. byrne, d. (2014) thoughts on a pedagogy of complexty.complicity: an international journal of complexity and education 11:2, 40‐50 byrne, d. & callaghan, g. (2013)complexity theory and the social sciences: the state of the art. abingdon: oxon: routledge. checkland, p. & scholes, j. (1990) soft systems methodology in action. chichester: wiley. cilliers, p. (1998) complexity and postmodernism: understanding complex systems. london: routledge. etzioni, a. (2014) humble decisionmaking theory. public management review 16:5, 611-620. http://dx.doi.org/10.20377/cgn-45 https://app.researchfish.com/awards/viewdetails/0?gorderby=organisation&filter=esrc-es/h04454x/1 https://app.researchfish.com/awards/viewdetails/0?gorderby=organisation&filter=esrc-es/h04454x/1 https://app.researchfish.com/awards/viewdetails/0?gorderby=organisation&filter=esrc-es/h04454x/1 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 46-58 doi: http://dx.doi.org/10.20377/cgn-45 58 university of bamberg press haynes, p. (2017) understanding the influence of values in complex systems based approaches to public policy and management. public management review. 1-17, published online 17-09-17 http://dx.doi.org/10.1080/14719037.2017.1364411 haynes, p. (2003) managing complexity in the public services. maidenhead: open university press. haynes, p. (2015) managing complexity in the public services. 2nd edition. oxon: routledge. hill, m. (2005) the public policy process. 4th edition. harlow: pearson. klijn, e.-h. (2008) complexity theory and public administration: what's new? public management review, 10:3, 299-318. meadows, d. h. (2009) thinking in systems: a primer. london: earthscan. moore, m. (1995) creating public value: strategic management in government. cambridge, ma: harvard university press. moore, m. (2013) recognising public value. harvard: harvard university press. price, j., haynes, p., darking, m., stroud, j., warren-adamson, c. & ricaurte, c. (2015) the policy maker’s toolkit. in r. geyer and p. cairney, (eds.) handbook on complexity and public policy. cheltenham: edward elgar. rhodes, m. e., murphy, j., muir, j. and murray, j. a. (2011) public management and complexity theory: richer decision-making in public services. oxon: routledge. savin-baden, m. (2000) problem based learning in higher education: untold stories. buckingham: society for research into higher education and open university. seddon, j. (2008) systems thinking in the public sector. axminister: triarchy press. snowden, d and boone, m. (2007) a leader’s framework for decision making. harvard business review. november, 2007 https://hbr.org/2007/11/a-leaders-framework-for-decisionmaking. stroud, j and warren adams, c. (2013) using complexity theory in kinship practice. child and family social work. 20:4, 407-414. teisman, g. & klijn, e.-h. (2008) complexity theory and public management: an introduction. public management review, 10:3, 287-298. teisman, g., van buuren, a. & gerrits, l. (2009) managing complex government systems: dynamics, self organization and coevolution in public investments.. london: routledge. wenger, e. (1998) communities of practice: learning, meaning and identify. cambridge: cambridge university press. http://dx.doi.org/10.20377/cgn-45 http://dx.doi.org/10.1080/14719037.2017.1364411 a. resources and their use 22 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy author: jason potts school of economics, finance and marketing, rmit university, melbourne, australia, e-mail: jason.potts@mit.edu.au one interpretation of complexity science is that it distinguishes two types of science—an equilibrium science of forces, as begun by newton, and a complexity science of rules, as exemplified by wolfram (2002). if you accept that argument, then there are also two types of economics—an equilibrium economics in which forces move resources around the economy, and a complexity economics in which generic rules structure knowledge in an economy (dopfer & potts, 2008). however, this also implies two types of economic policy—a policy framework based on reallocating resources and a policy framework based on redesigning rules (colander & kupers, 2014). modern economic policy generally engages in both, but we argue that this reflects the idea that modern economic policy has not caught up with complexity science. we illustrate how this difference plays out in the particular domain of innovation policy. keywords: complexity science, evolutionary economics, rules, innovation policy 1. pluralism in science and economics, and therefore in policy a fundamental premise of the modern rational approach to economic policy is that it should be based on what, in the vernacular, is called ‘proper science’ or ‘good economics’. the implicit case, robust against even the largest shock (mirowski 2015), is that there is only one of these proper and true forms—variously called neoclassical economics, mainstream economics, textbook economics, or neoliberalism (by its critics)—and that the various other forms of economics, those of heterodox stripe for instance, such as post-keynesian or marxist economics, are weaker forms that are not based on ‘proper science’ and certainly do not constitute good economics. in this ‘monotheistic’ worldview there is only one fundamental truth, but we must ever be vigilant because many flawed imposters surround the one true way. in this worldview, good economic policy, as that which is rationally guided and scientifically defensible, is based on the ‘one true economic model’ and avoids the siren-song seductions and free-lunch promises of the false models. good policy is based on good economics, which is based on good science. bad policy is based on bad economics, which is based on bad science. that, in stylized essence, is the modern catechism of economic policy. the question that this paper addresses concerns how evolutionary and complexity economics fits into this whole intellectual and technocratic exercise, and from there considers whether it is a proper and legitimate basis for policy. in essence, is it part of good economics and good science, and thus a valid and proper foundation for policy? or is it not? is it closer to marxism—once thought by many to be proper and good, but now consigned to the bin of bad economics and bad science? or is it closer to behavioural economics—once thought somewhat dubiously ‘psychological’, e.g. herbert simon’s conceptions of bounded rationality (simon, 1955) and its emphasis on non-standard lexicographic preferences (earl, 1986), but now refurbished with http://dx.doi.org/ mailto:jason.potts@mit.edu.au 23 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press proper scientific methods and credentials (experiments, game-theoretic-foundations, nobel prizes) and these days a respectable element of contemporary economic policy (e.g. ‘nudge’ policy, thaler & sunstein, 2008). when framed in this manner, the question would seem to be: is complexity economics canon? is it part of the one true way? because if it is indeed so, then and only then it is safe for policy use. yet i will argue that the monotheistic model—i.e. in which there in only one true scientific model, which begets the one true economic model, which begets the one true policy model, and about which all other models are false—is actually itself false (maki, 2001). rather, i want to argue for a pluralistici or dualistic (polytheistic, in the metaphor) approach in which there are two distinct worldviews, both of which are true, and both of which range over the domains of good and proper science, economics and policy. this isn’t a compromised or accommodating position, but one that corresponds to two distinct ontological claims about the world: namely whether we are dealing with a closed system or an open system (prigogine & stengers, 1984; dopfer & potts, 2004). there is a true world of science, economics and policy for closed systems, and a different but equally true world of science, economics and policy for open systems. what we have come to think of as the ‘proper science’ and ‘good economics’, in the sense alluded to above, is an approach that is true in a closed world. this is tightly modeled on 19th century physics, which was also a science of a closed world (mirowski, 1989; potts, 2000). much of our modern approach to economic policy is implicitly based on this manner of thinking. however, a different approach holds in an open system—i.e. an evolutionary system in which novelty emerges from within and a dissipative system that operates far-from thermodynamic equilibrium to maintain various levels of internal complexity (foster, 2005; herrmann-pillath, 2013). this requires a new kind of science, and a new approach to economics. it will also require a new approach to policy. the basic argument i make here is that the evolutionary and complexity approach to economics is true in the dualistic (polytheistic) sense of referring to an open system conception of the world in general and of the economy in particular (a different true system holds over a closed system worldview). the purpose of this paper, however, is to elaborate what this means for economic models and methods in general and to translate these into economic policy in particular. the case for a pluralistic/polytheistic worldview has distinctly practical implications regarding the effective design and conduct of economic policy. it does so by highlighting the very different phenomena, in the respective closed and open system cases, that policy seeks to effect. the closed system world is essentially construed as a world of economic ‘forces’, and which eventually arrive at some equilibrium configuration. policy is about shifting that equilibrium; i.e. a reallocation of resources. but the open system world is essentially construed as a world of economic ‘rules’, and from which emerge particular economic structures and processes. policy is about the design of rules to achieve particular structural or process outcomes. an evolutionarycomplexity approach does therefore imply a different economic policy perspective, but this is not a superficial distinction, even at the level of pluralistic economics. rather, the origins of this more fundamentally trace from a different approach to science. this paper will outline this case, focusing on the different ways this manifests in two views of innovation economics—and the definition of the innovation problem as an economic problem—and the implications for innovation policy. http://dx.doi.org/ 24 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press 2. two kinds of science from the modern perspective of complex systems, it is becoming increasingly clear that an important, or at least plausible, distinction can be made between two kinds of science (bertalanffy, 1968; simon, 1996; wolfram, 2002; mitchell, 2009): the equilibrium sciences, and the complexity sciences (a.k.a. the science of complex systems). the equilibrium sciences are built on the study of forces, and are exemplified by the use of ‘newtonian’ analytic methods based on algebra, topology and calculus. the complexity sciences are built on the study of rules, and are exemplified by analytic methods based on discrete, evolving, algorithmic systems. now this distinction is rarely clean in practice, for it is certainly the case that discrete non-linear systems can be transformed into continuous linear approximations ‘for the purpose of analysis’, and there are domains of analysis with interwoven characteristics of both (e.g. markov chain monte carlo, evolutionary game theory). and moreover, the claim for two kinds of science is far from unanimously acknowledged; indeed, it is only those who self-identify on the complexity side that seek to make this distinction, i.e. those affiliated with the likes of the santa fe institute in new mexico, or similar. furthermore, this division does not neatly occur across subject domains (e.g. physics, chemistry, economics, anthropology, et cetera), or even across broader historical and methodological aggregates (e.g. natural sciences, social sciences, sciences of the artificial, et cetera), both of which are the sort of divisions that give rise to interdisciplinary resolutions. rather, this difference occurs as a division about appropriate methods for the study of two broad classes of phenomena that correspond to closed systems and open systems (bertalanffy, 1968; prigogine & stengers, 1984). at this stage it does not really matter what these systems are made of, or what they are specifically open or closed to for instance energy, matter-energy, information, ideas, and knowledge. what does matter is their analytic representation, which can be understood in terms of solution concepts. by and large closed systems can be analytically represented with equilibrium solution concepts. note that these can be static or dynamic, in which time is one dimension of analysis. the type of sciences that are based on equilibrium solution concepts are sciences defined by forces; i.e. a vector of forces, or force fields. forces are conveniently represented mathematically in real vector space (rn) using algebra, topology and calculus. most 19th and 20th century sciences are of this type, from modern physics to neoclassical economics. methodologically, these sciences are reductionist and deductivist. open systems, however, have complexity solution concepts. i want to paint with a broad brush here and define complexity as a state-space at the edge-of-chaos (kauffman, 1993; mitchell, 2009) that refers to the study of self-organization processes and spontaneous and emergent order between the mutually interacting parts of a (complex) system (mitchell, 2009). this conception of a new kind of science has been percolating for a while now (hofstadter, 1979; nicolis & prigogine, 1989; wolfram, 2002; hendrickson & mckelvey, 2002), and has evolved along many different lines and approaches (e.g. dissipative systems, cellular automata, agent-based modeling, network analysis, genetic algorithms, and so on) and across effectively all scientific disciplines. this is increasingly indexed as 21st century science. however, what holds these together is not subject matter but a commitment to information-theoretic computable approaches that are best characterized as studying not forces (in vector-space), but rules (in computational space). methodologically, these sciences are emergent and inductive. now good arguments can and indeed have been made against this claim: some directly, some obliquely. the universal ‘baconian’ argument is that there is just the scientific method – that of conjecture and refutation (popper, 1963) – and whatever analytic tools are brought to bear http://dx.doi.org/ 25 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press on the questions of science are simply that: different tools used in science. thus there is science and non-science, rather than different types of science. a different way of viewing the unified thesis is to argue that even the major distinctions between natural and social science will yield to a deeper model of the unity of knowledge, or consilience (wilson, 1998), which is particularly associated with a unified evolutionary approach. the argument advanced here is not inconsistent with these universalist or unified claims, but rather refers to different ways of knowing different types of systems, and reflect different trade-offs over different methods. both natural and social sciences are amenable to closed and open system representations. closed system representations have many advantages, not the least of which are highly developed underlying mathematical theory and tractable analytic solutions. but open system representations are naturally geared to study unfolding processes and to explain change and emergence, but come at the price of tractable analytic solutions. but if we accept these trade-offs actually imply that from a practical perspective at the level of choosing methods and questions there are two types of science, then it can be reasonably noted that these choices align with a concept of the sciences of forces (broadly the 18th to 20th century sciences), and with an new and emerging sciences of rules (mid-20th century onwards). a science of forces sets up an analytical model of all major or relevant causal variables acting on the explanandum (‘things to be explained’), and derives the equilibrium prediction of the state of the system once all these forces have worked their way through the system. the enormous benefit of this mode of science is that it is rich in well-understood tools and methods in which the objective is to discover how observed reality is explained by the interaction of forces. a science of rules works differently, seeking to understand how a small number of interacting rules can produce an emergent order that corresponds to an observed reality. while newer, the price you pay for adopting this approach is that the tools while in some sense far more powerful are also harder to work with, and less well-understood. the objective in this approach is also different, namely to discover the initial conditions and rules that generated the observed reality, and the principles by which the system transforms itself through time. these different types of science have different solution concepts: equilibrium versus complexity. in consequence, these two types of science map to two different types of economics. classical and neoclassical classical economics is based on equilibrium sciences and is analytically formulated in terms of (fields of) forces, and complexity and evolutionary economics is based on complexity sciences and is analytically formulated in terms of (systems of) rules. two types of science imply two types of economics. that is the line of argument in section 2 above and in section 3 below. but the reason we make these arguments comes in section 4, when two types of economics implies two types of policy, which we illustrate with innovation policy. for it is here that we extract the core argument of this paper, reviewed in section 5, in which we run backwards through this same logic, that complexity policy therefore has its own foundation in not just economics, but in science. there exists, in other words, a distinct foundation for complexity-based economic policy. 3. two types of economics? two types of economic policy? economics is a curious type of science. the classical economists—especially ricardo, mill and marx—sought to develop a model of an institutionally rich and historically evolving but ultimately closed system, for example, theorizing about the notion of a long run stationary state (dopfer & potts, 2008, ch.1). the main economic policy implications that came from that era related to the logic and efficacy of broad policy settings, such as the debates about free trade. http://dx.doi.org/ 26 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press during the twentieth century, a new model of economics emerged that took a more direct approach to maximizing aggregate social welfare. this built on the positivist approaches to economics as a science argued by robbins (1932) and friedman (1953). this newly professionalized model was led by the likes of arthur pigou, paul samuelson and jan tinbergen, who advocated an activist engineering approach of the sort that was in the ascendency at the time in respect of large scale economic planning, industry policy and keynesian macroeconomics. this modern approach to economic policy was based around a theoretical conception of an economy as a system of market forces (the arrow-debreu general equilibrium paradigm) in which a policy approach (e.g. that based on aggregate demand management, industrial planning or market failure) was developed that sought to instrumentally control the economic system through the various policy levers and mechanisms that were available to governments through manipulation of market prices, incomes or spending. but a different approach to economics can be read in the more heterodox schools of economic thought, particularly those associated with evolutionary economics, behavioural economics, institutional economics, agent-based computational economics and complexity economics (arthur, et al., 1997; velupillai, 2000; holt, et al., 2011). these various schools of thought would at first seem to lack overarching commonalities, although potts (2000) argues they are all characterized by analytic representation in non-integral space (cf. integral space in the fieldtheoretic equilibrium models (mirowski, 1989)). but what they all share is that they are based on a theoretical conception of the economy as made of generic rules (dopfer. et al., 2004; dopfer & potts, 2008). these heterodox schools of thought consistently argue that the economy is made of knowledge in the form of institutions, in the widest sense of that term (north, 1990; loasby, 1999; williamson, 2000), and that economic agents and firms alike are made of habits and routines as units of knowledge (nelson & winter, 1982). economies are complex systems because knowledge is a complex system, and the process by which knowledge changes and grows is an evolutionary process (ziman, 2000; potts, 2000). economies evolve because knowledge evolves. now there is of course a great deal of overlap between these two approaches and it is an almost cartoonish simplification to say that on one side lies an equilibrium model of forces allocating scarce economic resources, while on the other lies a complex systems model of evolving rules. most obviously, institutional economics can be understood in both senses. but it is also interesting that institutional economics has significantly fractured down this line (hodgson, 1998, 2007). this basic difference between two approaches to economics reveals itself in the different conceptions of what an economy is and what it is made of, and of the corresponding economic problem. in the mainstream equilibrium models an economy is made of (scarce) resources (i.e. commodities), which are allocated over an economic space that can be described by a vector of prices and quantities; i.e. the commodity space. technologies and preferences parameterize that space and each point in the space represents a different configuration of allocations. some of these are efficient. within this positive analytic context, the purpose of policy is to artificially intervene in prices and quantities in order to achieve efficient or otherwise desirable allocations of resources. this is the physics approach to economics translated into the engineering approach to economic policy, which reached its apogee in the work of paul samuelson, an engineer turned economist, and broadly in the program of work of the cowles commission (mirowsk,i 1989, 2001). however, complexity and evolutionary models start from a different fundamental conception of what an economy is made of: namely that an economy is made of knowledge, not of resources. this knowledge has various forms, including: technical knowledge (technologies), social coordinating knowledge (institutions), organizational competence-carrying knowledge (firms), behavioural and cognitive knowledge (habits, routines), and so on. patterns of economic http://dx.doi.org/ 27 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press activity, and economic growth and dynamics, are a consequence of the structure of that knowledge and of changes in that knowledge. in this view, the knowledge base of an economy is a complex evolving system. the economic problem is centered about the effective discovery, coordination, use and replication of knowledge, as joseph schumpeter first clearly explained. it is this knowledge that enables the production of commodities and the exploitation of resources, and overall is what explains the wealth of nations (beinhocker, 2006; mccloskey, 2016). the essential point is that an evolutionary and complexity-based approach to economics expresses an underlying approach to science that seeks to explain economic reality in terms of complex interacting rules, not as a field of forces. the essential difference is that rules are structured as complex systems and evolve; forces are not and do not. an equivalent way to state this methodological proposition is that an evolutionary-complexity approach to economic analysis is a study of the growth of knowledge in which the constituent matter of an economic system is understood to be different forms of knowledge, ranging over technical knowledge, organizational knowledge, institutional knowledge, behavioural knowledge, and so on. it is this generic knowledge (which i have referred to here under the category of generic ‘rules’, following dopfer and potts (2008)) that makes possible economic operations such as production, consumption, coordination and exchange. this structure of knowledge forms a nested complex open system that evolves under the evolutionary mechanisms of variation, selection and differential replication (loasby, 1999; ziman, 2000). rules are complex systems that evolve, forces do not. an economics of rules is thus fundamentally different from an economic of forces. we may summarize our argument so far as follows: the complexity era has afforded us a new kind of science (wolfram, 2002) that, when combined with evolutionary theory, has given us a new kind of economics (potts, 2000; mckelvey, 2004; foster, 2005, 2006; beinhocker, 2006; colander, et al., 2011; kirman, 2011; arthur, 2013; elsner, et al., 2015; kauffman, et al., 2015). but do we yet have a new kind of policy? i want to argue that this is indeed the next step. but our starting point is that there is, as yet, very little recognition of that imperative. nevertheless, a new modern paradigm for public economic policy in the evolutionary and complexity science age is indeed emerging (room, 2011; dolphin & nash, 2012; durlauf, 2012; colander & kupers, 2014; elsner, 2015). there are a number of nascent forms of a new type of policy approach that proceeds from a rules-based science and economics. this can be observed in aspects of the new institutional economics (ostrom, 2005) and the economics of mechanism design and market design (roth, 2015). but a particularly interesting aspect of this can be observed in the study of innovation economics. in this next section 4 we show how two distinct views on the nature of the innovation problem can be discerned that broadly line up with the two views of economics and two views of policy discussed in the above section. 4. the nature of policy: two views of the innovation problem the two views of economics—a scientific model of forces, and the corresponding equilibrium model; and a scientific model of rules, and the corresponding complexity model—can be observed in two distinct views of the innovation problem, variously as a market failure problem, i.e. forces, equilibrium allocation problem, and as a collective action problem, i.e. rules, complex coordination problem (allen & potts, 2016). these two distinct interpretations of the innovation problem translate into two distinct models of innovation policy: one corresponding to government intervention to reallocate resources, and thus solving a market failure problem; the other corresponding to governance in the discovery of effective rules (organizations and institutions) to coordinate collective action. http://dx.doi.org/ 28 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press in the equilibrium view, the innovation problem is diagnosed as a misallocation of resources that is then corrected through innovation policy instruments to reallocate resources – indirectly via incentives, i.e. artificial rent creation, or directly via government coercion, i.e. tax and spend. but from the complexity perspective, the innovation problem is diagnosed as a collective action problem associated with coordinating distributed knowledge under conditions of high uncertainty. this sort of problem is resolved not by moving resources around, but by discovering effective rules that will enable coordination to occur and for new order to emerge (potts, 2014). from the equilibrium perspective, what a policy solution looks like is a reallocation of resources. that is a surface rearrangement of an economy, leaving its deep structure unchanged. but from the complexity perspective, what a policy solution looks like is the discovery and adoption of new rules. that is a deeper change in the economic order, and these new rules will then cause subsequent changes in the surface level of the economy to unfold. the innovation problem the innovation problem was first clearly expressed by nelson (1959) and arrow (1962b) as a problem in the production of new information under competitive market institutions, diagnosed as a species of market failure in which “a free enterprise economy underinvests in invention and research (compared with an ideal) because it is risky, because the product can be appropriated only to a limited extent, and because of increasing returns” (arrow, 1962a: 624). the innovation problem has several distinct policy solutions (davidson & potts, 2016). it can take a coasian form – attach property rights and bargain to equilibrium – which is the intellectual property approach; a pigovian form – identify the externality and correct with subsidy – which is the r&d tax credit approach; a marxian-keynesian form through industry planning and targeted government spending to redirect existing organizations toward the innovation target; and arrownelson-freeman form to create non-market research organizations, such as universities or public science institutes, and to assemble these into an effective innovation system. innovation policy institutions range from decentralized and market focused – such as intellectual property rights, or regulatory support – that operate through the legislative and judicial branches, to mixed models operating through the tax system – such as r&d tax credits or targeted industry subsidies – to centralized and planning focused – such as public science institutes – that operate through government agencies (flanagan, et al., 2011). while some are market-facing, these are all government mechanisms. intellectual property is a state-sanctioned monopoly, and r&d tax credits work through the tax system. public procurement models, for example in defense contracting or medical research, are often publically funded but delivered by private organizations, whereas public science is usually both public funded and publically produced. innovation is an economic problem because inputs into innovation have alternative uses. but the innovation problem is further confounded because the output is information produced under uncertainty (arrow, 1962a). first, the output has fundamentally uncertain value ex ante. this makes it difficult to optimize on levels of input spending. it also incentivizes the rational agent to strategically wait for some other agent to reveal that information. second, there are often significant fixed costs (indivisibilities) associated with research that are irrecoverable in a competitive market with marginal cost pricing. again, this incentivizes the rational agent to wait. third, ideas are appropriable, or non-excludable. again, the first mover is at a competitive disadvantage. in the absence of significant first-mover advantages, such as learning-by-doing (arrow, 1962b), the rational agent will underinvest in invention, as the production of new information, from the social welfare perspective, will experience market failure (bleda & del rio, http://dx.doi.org/ 29 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press 2013). this is the standard diagnosis of the innovation problem (martin & scott, 2000). the innovation problem is a problem of the competitive production of information under uncertainty, giving rise to pigovian externalities.ii it is equivalently a public goods problem (lindsay & dugan, 2013), or a problem of market structure (aghion, et al., 2005). this connects the innovation problem to the economics of efficient and effective institutions (north, 1990; acemoglu & robinson, 2012), where we can clearly see the innovation problem and its solution expressed in terms of the discovery and design of effective rules. consider the institutions that facilitate innovation and economic growth and development. for instance, the schumpeterian entrepreneur is a source of innovation in an economy, but that same behavior can be destructive of economic growth if channeled through political rather than market means (baumol, 1990). the effectiveness of the rule of law predicts productive entrepreneurship, and therefore innovation. innovation is also inhibited by extant regulation, much of which is actively sought by incumbent firms in order to protect their own interests against ‘creative destruction’ by new entrants. regulation by-and-large inhibits innovation (stigler, 1971; thierer, 2014). furthermore, the innovation problem also shows up as a design problem in the overall crafting of innovation institutions, both as a problem of designing and implementing an efficient institutional mix of innovation policies (borras & edquist, 2013), and in the assembly of an effective national innovation system (freeman, 1995; dodgson, et al., 2011; potts, 2016). the economics of innovation can be framed as a problem in microeconomic incentives that are consequence of the peculiar properties of the production of information in competitive markets, namely the existence of fixed costs in research and development, the appropriability of the resultant product, and its uncertain value. in a competitive market, this leads to private underinvestment. but the economics of innovation can also be framed as a problem in public goods provision because new ideas are non-rivalrous and non-excludable. this suggests public provision. however, the economics of innovation can also be framed as a problem in political economy, given that solutions to these problems involve the creation of new institutions with their own incentive problems, and the evolved institutions of a market economy will have their own effect on the allocation and intensity of entrepreneurial action. this covers neoclassical economics, evolutionary economics, institutional economics, and political economy. but a further angle on the innovation problem is to view it as a problem of collective action. the innovation commons commons-type solutions to social dilemmas make use of effective governance rules to regulate access to the shared resource and to punish those who violate the rules (ostrom, 1990). if we view the innovation problem from the perspective of ‘who pays for the research and development?’ then the collective action problem is to pool collective resources and to exclude free-riders. but the innovation problem is not necessarily a pooling of money to create a fund or indeed any pooling of fungible resources, say to create a laboratory. the innovation problem can also be formulated as a problem of pooling distributed information about the prospects, costs, risks, and opportunities of the new idea or technology. this information is distributed in the sense that different agents will have different parts of it, and these parts may not have much value or meaning in and of themselves. it is only when combined that these elements of information become of value in estimating the nature of the idea and the outlines of the opportunity it presents. this pooled information is the central valuable resource in an innovation commons because this is the information that entrepreneurs need in order to invest in the new idea, which is the origin of the process of innovation. http://dx.doi.org/ 30 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press an innovation commons (allen & potts, 2015) is a further class of institutional solution to the innovation problem. but it is a solution that comes not from the state (government), but rather from civil society (governance). it is closer to a spontaneous order than a planned institutional system (potts, 2014). for this reason, the innovation commons does not usually form part of the various instruments of innovation policy yet is a near pure example of a bottom-up institution to resolve the innovation problem as a collective action problem.iii an innovation commons is both similar and different to other types of commons, encompassing natural resource commons (ostrom, 1990) and the ‘new commons’ (hess, 2008; bertacchini, et al., 2012). the similarities accrue in the sense that an innovation commons is manifestly an institutionalized community with rules of governance. as frischmann (2013) observes, a “commons is an institutionalized community practice, a form of community management or governance.” in this definition, the “basic characteristic that distinguishes commons from non-commons is institutionalized sharing of resources among members of a community.” (madison, et al., 2010b, 841) the rules by which an innovation commons works are similar to the rules by which a natural resource commons works, as captured by ostrom’s wellknown eight design principles for managing the commons (poteete, et al., 2010). the value of the innovation commons comes from the value of the pool of information that is gathered there, and the basic problem is the same, namely seeking to resolve a social dilemma in contribution subject to free-riding. in an innovation commons, an individual can access new information about the uses and opportunities inherent in a new technology, or the meaning of the information that they already have in a way that would be costly or impossible in the absence of that institution. this combines with the natural human propensity to cooperate (bowles & gintis, 2011) to arrive at a new institutional watershed in the cooperative possibilities of the coordination of economic production. at the surface level of analysis, in a field-of-forces model of the economy, the innovation problem presents as a problem of market failure in the production of new knowledge under competitive market conditions. from a policy perspective, the problem here is clearly diagnosed as a misallocation of resources that can be corrected with some intervention. but at a deeper level of analysis, in a rules-based model of the economy, the innovation problem presents in a different form, namely as a problem of coordination between agents in which distributed agents need to arrive at a system of rules that will enable effective cooperation in order to coordinate resources, including information, toward mutually satisfactory ends. the solution here involves the discovery and implementation of effective rules. and what the theory of the innovation commons highlights is that such rules will often emerge from the bottom-up, arising from the actions of a motivated and incentivized community. in other words, the sort of economy policy that emerges from a complexity perspective does not necessarily flow from government as top-down public policy, but will often emerge from community-level as civil-society instigated bottom-up discovery of effective rules. 5. complexity policy follows from complexity economics, which is about rules as complexity science emerged through the latter part of the twentieth century it soon became clear that it offered the prospect of a new kind of science (wolfram, 2002), implying a new kind of economics (colander, et al., 2011; arthur, 2013). the latest phase of this unfolding development is recognition that it also seems to imply a new kind of economic policy (durlauf, 2012; colander & kupers, 2014). a key claim made in this paper is that this policy bifurcation is deep, tracing to the distinction between a science of forces and a science of rules. the two http://dx.doi.org/ 31 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press different types of policy follow from two types of economics, which follows from two types of science, namely a science of forces – exemplified in general equilibrium economics, and a science of rules – exemplified in complexity economics. the approach to economic policy based on a science of market forces is one in which an equilibrium set of prices and allocation of resources can be displaced by the policy-maker, as a kind of chief-engineer, in order to achieve a preferred resource allocation. but the approach to economic policy based upon a conception of the economy as made of knowledge, and based on a science of rules (e.g. dopfer & potts, 2008), is substantially different because it occurs at a deeper level in the economy. the equilibrium approach to economic policy, as based on a field-of-forces model of the economy, seeks to intervene in the surface layers of an economy by redistributing resources towards a more efficient allocation. but the complexity approach to economic policy, as based on a generic rule model of the economy, seeks to change the deep structure of the economic order (see also pelikan & wegner, 2003). douglass north (1990) identified this with institutional change; james buchanan (1990), building on brennan and buchanan (1985) and hayek (1960), called this constitutional economics; and oliver williamson (2000) identified it with four emergent levels of long run dynamics. dopfer and potts (2008) call it the generic level of analysis and argue that economic policy based on evolutionary-complexity approach is inherently generic policy, which is to say rules-based policy. now while it is manifestly true that all economic policy as practiced by governments involves both modes – deep and surface, constitutional (change in rules and institutions) and operational (change in resources, distributions, prices) – the point here is that complexity policy emerges from complexity economics, which emerges from complexity science, which is fundamentally a science of rules. complexity policy is ipso facto about rules. an important but widely underappreciated implication of this rules-based approach is that policy does not exclusively or even necessarily come from, or work through, the standard democratic/technocratic model of the modern policy instruments of government. colander and kupers (2014) use the standard complexity theory trope to portray this as a top-down versus bottom-up distinction, but the radical implication here is that effective policy may emerge directly from civil society, arriving as an associational and community driven dynamic. we are, for example, starting to observe a lot of this emerging from the so-called ‘sharing economy’, about which top-down policy is variously playing catch-up, or seeking to formalize many norms and practices that have evolved spontaneously to address collective action problems in the affected communities. if policy means effective rules, then whatever makes effective rules is a policymaker. innovation policy illustrates this. i outlined above two distinct formulations of the innovation problem. from the field of forces viewpoint, the innovation problem is diagnosed as an allocation problem where too little is invested in new knowledge, and is resolved with policy to reallocate resources toward innovation in order to maximize some social welfare function. however, from the complexity perspective the innovation problem is diagnosed as a collective action problem for dealing with the entrepreneurial uncertainty surrounding a new technology, and resolved with effective rules to enable agents to cooperate to pool resources and information to overcome that uncertainty. government innovation policy transfers resources or creates rents, which is top-down by degrees. but a pure bottom-up innovation policy is the emergence of the ‘innovation commons’ when a community of enthusiasts for a new technology figure out rules to enable them to pool and share information and resources to explore and develop the new technology into exploitable opportunities, including what may eventually become firms, markets and industry organizations. these emergent coordinating rules form a solution to the innovation problem: they are, in effect, policy. http://dx.doi.org/ 32 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press there are two kinds of policy in economics because there are two kinds of economics, because there are two kinds of science. that has been the argument of this paper. the complexity perspective in economics implies that policy is the discovery of effective rules for collective action. this contrasts with the mid-20th century engineering inspired model in which policy is about reallocating resources across the surface of the economy. these rules are what the new institutional economists call ‘the rules of the game’. but these are not necessarily at the macro level of the economy, and nor are they well-understood at the micro-level of analysis. dopfer and potts (2008) call these ‘meso rules’—in that they are common to a population of micro-agents, and are the building blocks of the complex systems that compose the macro-economy. they are the proper target of economic policy from the complexity perspective. in the field-of-forces approach to economics, there is macro-economics and macro-policy, and micro-economics and micro-policy. but in the complexity approach to economics, which is based on a science of rules, there is just meso-policy, but which in turn has micro and macro consequences. complexity economics therefore does not just suggest some new tweaks and riffs on the extant approach to economic policy. instead it is far more revolutionary: a whole new meso-based rule-centered approach to economic policy because it is a whole new economics based on a whole new science. references acemoglu, d., robinson, j. (2012) why nations fail. crown books: new york. aghion, p., bloom, n., blundell, r., griffith, r., & howitt, r. (2005). competition and innovation: an inverted-u relationship. quarterly journal of economics, 120(2): 701–28. allen, d., & potts, j. (2016). the innovation commons: why it exists, what it does, who it benefits, and how. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2617141 allen, r. c. (1983). collective invention. journal of economic behavior and organization, 4(1): 1–24. arrow, k. (1962a). economic welfare and the allocation of resources for innovation. in richard r. nelson (ed.). the rate and direction of inventive activity: economic and social factors. princeton: princeton university press: 609-626. arrow, k. (1962b). the economic implications of learning by doing. review of economic studies, 29 (3): 155–73. arthur, w.b. (2013). complexity economics: a different framework for economic thought. sfi working paper 2013-04012; http://tuvalu.santafe.edu/~wbarthur/papers/comp.econ.sfi.pdf. arthur, w.b., durlauf, s., & lane, d. (1997). the economy as an evolving complex system ii. redwood city: addisonwesley. baumol, w. (1990). entrepreneurship: productive, unproductive and destructive. journal of political economy, 98(5): 893921. bertalanffy, l. von (1968). general systems theory. new york, ny: braziller. beinhocker, e. (2006). the origins of wealth: evolution, complexity, and the radical remaking of economics. new york, ny: random house business. bertacchini, e., bravo, g., marrelli, m., & santagata, w. (2012). cultural commons: a new perspective on the production and evolution of cultures. cheltenham: edward elgar. bleda, m., & del rio, p. (2013). the market failure and the systemic failure rationales in technological innovation systems. research policy, 42: 1039–52. borrás, s., edquist, c. (2013). the choice of innovation policy instruments. technological forecasting and social change 80(8): 1513-1522. bowles, s., & gintis, h. (2011). a cooperative species: human reciprocity and its evolution. princeton: princeton university press. buchanan, j. (1990). the domain of constitutional economics. constitutional political economy, 1(1): 1-10. brennan, g., & buchanan, j. (1985). the reason of rules. cambridge: cambridge university press. chesbrough, h. (2003). the era of open innovation. mit sloan management review, 44(3): 35–41. colander, d., holt, r., & rosser, j.b. (2011). the complexity era in economics. review of political economy, 23(3): 357-69. colander, d. & kupers, r. (2014). complexity and the art of public policy: solving society’s problems from the bottom-up. princeton: princeton university press. davidson s. & spong h. (2010). positive externalities and r&d: two conflicting traditions in economic theory. review of political economy, 22(3): 355–72. http://dx.doi.org/ http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2617141 http://tuvalu.santafe.edu/%7ewbarthur/papers/comp.econ.sfi.pdf 33 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press davidson, s. & potts, j. (2016). the new institutional approach to innovation policy. australian economic review policy forum: research and innovation, (forthcoming) davis, j. (2006). the turn in economics: neoclassical dominance to mainstream pluralism? journal of institutional economics, 2(1): 1–20. dodgson, m., hughes, a., foster, j. & metcalfe, j.s. (2011). systems thinking, market failure, and the development of innovation policy: the case of australia. research policy, 40: 1145–56. dolphin, t., & nash, d. (2012). complex new world. translating new economic thinking into public policy, london, uk: institute for public policy research. dopfer, k., foster, j. & potts, j. (2004). micro–meso–macro. journal of evolutionary economics, 14: 263–79. dopfer, k. & potts, j. (2004). evolutionary realism: a new ontology for economics. journal of economic methodology, 11: 195–212. dopfer, k. & potts, j. (2008). the general theory of economic evolution. london, uk: routledge. durlauf, s. (2012). complexity, economics and public policy. politics, philosophy and economics, 11(1): 45–75. earl, p.e. (1986). lifestyle economics. aldershot, uk: harvester wheatsheaf. elsner, w. (2015). policy implications of economic complexity and complexity economics. https://mpra.ub.unimuenchen.de/63252/1/mpra_paper_63252.pdf. elsner, w., heinrich, t. & schwardt, h. (2015). the microeconomics of complex economies, amsterdam: elsevier academic press. flanagan, k., uyarra, e., laranja, m. (2011). reconceptualising the ‘policy mix’ for innovation. research policy 40(5): 702713. foster, j. (2005). from simplistic to complex systems in economics. cambridge journal of economics 29(6), 873–92. foster, j. (2006). why is economics not a complex systems science? journal of economic issues, 40(4): 1069–91. freeman, c. (1995). the ‘national system of innovation’ in historical perspective. cambridge journal of economics 19(1): 5–24. friedman, m. (1953). essays in positive economics. chicago, il: university of chicago press. frischmann, b. (2013). infrastructure: the social value of shared resources. new york, ny: oxford university press. fullbrook, e. (2003). the crisis in economics: the post autistic economics movement: the first 600 days. london, uk: routledge. hayek, f.a. (1960). the constitution of liberty. chicago, il: university of chicago press. hendrickson, l. & mckelvey, b. (2002). foundations of a new social science. proceedings of the national academy of sciences, 99: 7288-97. herrmann-pillath, c. (2013). foundations of economic evolution: a treatise on the natural philosophy of economics. cheltenham, uk: edward elgar. hess, c. (2008). mapping the new commons. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1356835. hodgson, g. (1998). the approach of institutional economics. journal of economic literature, 36(1): 166-192. hodgson, g. (2007). evolutionary and institutional economics as the new mainstream? evolutionary and institutional economics review, 4(1): 7-25. hofstadter, d. (1979). godel, esher, bach: an eternal golden braid. new york, ny: basic books. holt, r., rosser, jb. & colander, d. (2011). the complexity era in economics. review of political economy, 23(3): 359–71. kauffman, s. (1993). origins of order. oxford, uk: oxford university press. kauffman, s., koppl, r., felin, t. & longo, g. (2015). economics for a creative world., journal of institutional economics, 11(1), 1–31. kirman, a. (2011). complex economics: individual and collective rationality, the graz schumpeter lectures. london, uk: routledge. lindsay c, dougan, w (2013) ‘efficiency in the provision of pure public goods by private citizens’ public choice, 156, 31–43. loasby, b. (1999). knowledge, evolution and institutions in economics. the graz schumpeter lectures. london, uk: routledge. madison, m., frischmann, b. & strandburg, k. (2010b). the complexity of commons., cornell law review 95(4): 839– 50. maki, u. (2001). the economic worldview: studies in the ontology of economics. cambridge, uk: cambridge university press. martin, s. & scott, j. (2000). the nature of innovation market failure and the design of public support for private innovation, research policy, 29 (4-5), 437–47. mccloskey, d. (2016). bourgeois equality: how ideas, not capital or institutions, enriched the world. chicago, il: university of chicago press. mckelvey, b. (2004). toward a complexity science of entrepreneurship. journal of business venturing, 19(3): 313-41. http://dx.doi.org/ https://mpra.ub.uni-muenchen.de/63252/1/mpra_paper_63252.pdf https://mpra.ub.uni-muenchen.de/63252/1/mpra_paper_63252.pdf http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1356835 34 complexity, governance & networks special issue: complexity, innovation and policy (2017) 22-34 jason potts: complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy doi: http://dx.doi.org/10.20377/cgn-40 university of bamberg press mirowski, p. (1989). more heat than light: economics as social physics, physics and nature’s economics. new york, uk: cambridge university press. mirowski, p. (2001). machine dreams: economics becomes a cyborg science. new york, ny: cambridge university press. mirowski, p. (2015). never let a serious crisis go to waste: how neoliberalism survived the financial meltdown. london, uk: verso. mitchell, m. (2009). complexity: a guided tour. oxford, uk: oxford university press. nelson, r. (1959). the simple economics of basic scientific research. journal of political economy, 67: 296–307. nelson, r. & winter, s. (1982). an evolutionary theory of economic change. cambridge, ma: harvard university press. nicolis, g. & prigogine, i. (1989). exploring complexity. new york, ny: wh freeman. north, d.c. (1990). institutions, institutional change and economic performance. cambridge, ma: cambridge university press. nuvolari, a. (2004). collective invention during the british industrial revolution: the case of the cornish pumping engine. cambridge journal of economics 28: 347–63. ostrom, e. (1990). governing the commons: the evolution of institutions for collective action. cambridge, ma: cambridge university press. ostrom, e. (2005). understanding institutional diversity. princeton, ny: princeton university press. pelikan, p. & wegner, g. (2003). the evolutionary analysis of economic policy. cheltenham, uk: edward elgar. popper, k. (1963/2014). conjectures and refutations: the growth of scientific knowledge. london, uk: routledge. poteete, a., janssen, m., & ostrom, e. (2010). working together: collective action, the commons, and multiple methods in practice. princeton, ny: princeton university press. potts, j. (2014). innovation is a spontaneous order. cosmos and taxis, 2(1): 1–10. potts, j. (2000). the new evolutionary microeconomics: complexity, competence and adaptive behaviour. cheltenham, uk: edward elgar. potts, j. (2016). innovation policy in a global economy. journal of entrepreneurship and public policy, forthcoming. prigogine, i., & stengers, i. (1984). order out of chaos. new york, ny: bantam books. robbins, l. (1932). an essay on the nature and significance of economic science. london, uk: macmillan. room, g., (2011). complexity, institutions and public policy. cheltenham, uk: edward elgar. roth, a. (2015). who gets what and why. london, uk: harpercollins. simon, h. (1955). a behavioral model of rational choice. quarterly journal of economics, 69(1): 99–118. simon, h. (1996). the sciences of the artificial (3rd ed., 1st edn. 1969). cambridge, ma: mit press. stigler, g. (1971). the theory of economic regulation.. bell journal of economics and management science, 2(1): 3–21. thaler, r., & sunstein, c. (2008). nudge. new york, ny: penguin books. thierer, a. (2014). permissionless innovation. mercatus centre, george mason university, fairfax. velupillai, k. (2000). computable economics. oxford, uk: oxford university press. von hippel, e. (2005). democratizing innovation. cambridge, ma: mit press. von hippel, e., & von krogh, g. (2003). open source software development and the private-collective innovation model: issues for organization science. organization science, 14(2): 208–23. von hippel, e. (1986). lead users: a source of novel product concepts. management science 32(7): 791–805. wilson, e.o. (1998). consilience: the unity of human knowledge. new york, ny: knopf. williamson, o. (2000). the new institutional economics: taking stock, looking ahead. journal of economic literature 38: 595-613. wolfram, s. (2002). a new kind of science (vol. 5). champaign: wolfram media. ziman, j. (2000). technological innovation as an evolutionary process. cambridge, ny: cambridge university press. endnotes i pluralism in economics (davis, 2006) has been a recent touchstone for movements such as the post-autistic and real world economics (fullbrook, 2003). ii davidson and spong (2010) argue that this notion of externalities in innovation is widely confused with the marshallian conception of an external economy. iii the innovation commons overlaps with concepts from strategy such as von hippel’s (1986, 2005) ‘user innovation’ or ‘democratized innovation’ and chesbrough’s (2003) ‘open innovation’. economic historian robert allen (1983) explored this as ‘collective invention’ (also nuvolari, 2004). http://dx.doi.org/ complexity, economics, and innovation policy: how two kinds of science lead to two kinds of economics and two kinds of policy 35 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press a complexity-theoretic perspective on innovation policy author: koen frenken innovation studies, copernicus institute of sustainable development, utrecht university, the netherlands k.frenken@uu.nl, @kfrenken it is argued that innovation policy based on notions of market failure or system failure is too limited in the context of current societal challenges. i propose a third, complexity-theoretic approach. this approach starts from the observation that most innovations are related to existing activities, and that policy’s additionality is highest for unrelated diversification. to trigger unrelated diversification into activities that contribute to solving societal challenges, government’s main task is to organize the process of demand articulation. this process leads to clear and manageable societal objectives that effectively guide a temporary collation of actors to develop solutions bottom-up. the combination of a broad coalition, a clear objective and tentative governance are the means to cope with the inherent complexity of modern-day innovation. keywords: innovation policy, complexity theory, evolutionary economics, entrepreneurial state, temporary innovation system, societal challenges, tentative governance 1. introduction with the fall of keynesianism in the 1980s and the rise in global competition in the 1990s, innovation policy has become the cornerstone of economic policy in every high-income country or region. indeed, since solow’s (1957) seminal study, ample empirical evidence has been collected showing that innovation accounts for a large part of economic growth. by stimulating innovation in firms, governments hope to increase productivity, employment growth and prosperity. though the main objectives of innovation policy may be widely shared politically, the understanding of innovation processes in the economy, and how these can be influenced by policy, is illunderstood. in this essay, i will first summarize the two dominant perspectives on innovation policy (the market-failure approach and system-failure approach), and i will then develop a more historical perspective based on complexity theory. while the complexity-theoretical perspective incorporates key elements from the two existing paradigms, it opens up a new perspective in which institutions are seen as co-evolving with radical innovation, rather than simply enabling incremental innovation processes. using this framework, we critically assess mazzucato’s (2013) notion of the ‘entrepreneurial state’ and end with a plea for a more bottom-up, challenge-led innovation policy. 2. market failure academic thinking about innovation, and innovation policy, has long been dominated by the economics profession. from the perspective of mainstream economy theory, innovation – defined here as the successful introduction of new products, services and production processes – can be complexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ mailto:k.frenken@uu.nl 36 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press viewed as a knowledge production activity. similar to any kind of production activity, inputs are transformed into output where inputs are mainly knowledge and research equipment and output are new products and production processes. this linear view on innovation is well suited for econometric analysis once inputs are proxied by r&d expenditures and outputs by patents. using such a ‘knowledge-production-function’, one can measure the return to r&d investments at regional or national levels as well as the extent to which regions/countries benefit from r&d invested by other regions/countries, otherwise known as spillovers. from the perspective of the linear model, the economic question holds whether the r&d investments is below the socially optimal level. theoretically, it can be expected that firms underinvest in r&d because part of their outputs of their investments spillover to other firms, who imitate innovation without remunerating the originator, a case of market failure (arrow, 1962). this is especially the case in industries where innovations cannot be patented. hence, the conditions that allow firms to appropriate the returns form their r&d investment are important. a second reason why firms may under-invest related to the inherent uncertainty, in the knightian sense, of the outcomes of r&d investments. in many cases, firms cannot foresee at all whether r&d investment will lead to a successful innovation. consequently, firms will find it difficult to raise funding. these theoretical arguments provide a clear policy rationale for r&d subsidies for firms as well as for universities, and for tight patent protection. indeed, if firms invest less in r&d than that would be socially desirable, r&d subsidies and strengthening patent protection are standard ways to repair such market failures. for long, r&d subsidies – mainly in the form of r&d tax incentives such as wage subsidies for r&d personal and lower profit tax for r&d performing firms – were uncontroversial given the strong theoretical support from economics and the strong political support from left to right given their shared objectives in promoting growth and employment. however, the more recent insights cast some doubts about r&d subsidies. first, the effect of r&d subsidies on private r&d investments is rather small. at the micro-level, a meta-study found that the elasticity is 0.21 meaning that if subsidies reduce the cost of r&d by 1 percent, firms only invest 0.21 percent more in r&d (gaillard-ladinska, et al., 2015). there is also some evidence that this effect is even lower for larger firms. at the macro-level, there is no relation between a country’s innovativeness and the share of government support to r&d that is allocated via r&d tax incentives (gaillardladinska & straathof, 2015). furthermore, it may matter a lot how innovation is defined in the context of r&d subsidies. some countries subsidize any r&d activity including innovations that are new-to-the-firm but have already been developed elsewhere, while others are more restricted and subsidize only firms who develop innovation that are new-to-the-country and new-to-theworld. in the former case, r&d subsidies may actually discourage true innovations to some extent, as governments subsidize firms to imitate innovations of others. in all, this suggests that restricting r&d tax incentives to smaller firms that developed new-to-the-world innovations would be most effective. a second critique to the market-failure rationale for innovation policy focuses on the patent system. as stressed by boldrin and levine (2013), there exists a ‘patent puzzle’: in spite of the enormous increase in the number of patens and the strength of their legal protection from the 1990s onwards, productivity growth has slowed down in almost all high-income counties in the same period. this suggests that patents may do as much good as they do harm. for example, many technologies that are patented are not exploited commercially; rather, these are patented to prevent competitors to enter into particular innovation paths. furthermore, as a piece of knowledge can be both an output of one firm’s r&d process and an input of another firm’s r&d process later on, tight patent protection may block promising innovations, especially of the recombinant type. more generally, the importance of patents in stimulating innovation is should complexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ 37 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press not be overstated, because firms dispose of other strategies as well to protect their innovations, including secrecy, branding and being first to the market (arundel, 2001; cohen, et al., 2002). 3. system failure a second influential approach in innovation policy starts from the notion of a national innovation system. freeman was first to introduce this concept, and defined it as “the network of institutions in the public and private sectors whose activities and interactions initiate, import, modify and diffuse new technologies” (freeman, 1987, p. 1), where the word institutions here refers to actors such as firms, universities, financial organizations, public research organizations, government, consumer organizations, etc.. the concept of the ‘triple helix’ of university-industrygovernment relations proposed by etzkowitz and leydesdorff (2000) also stressed the benefits of collaboration and coordination between actors. both notions of innovation system and triple helix stemmed from the recognition that many innovation processes involve a multitude of actors who each contribute and jointly collaborate to the shaping and success of new technologies and services. in this more holistic perspective, the functioning of an innovation system can thus be assessed by looking for system failures in analogy to market failures. such failures may refer to underdeveloped actors within the innovation system or underdeveloped linkages between actors that make up the innovation system (klein woolthuis, et al., 2005, p. 611). the notion of system failure was already implicit in policy circles in the mid-1990s when the ‘european paradox’ was identified, which came to dominate european innovation policy thinking for at least 15 years. this paradox was formulated as: “compared with the scientific performance of its principal competitors, that of the eu is excellent, but over the last fifteen years its technological and commercial performance in high-technology sectors such as electronics and information technologies has deteriorated.” (european commission, 1995, p. 5). the key to solving the paradox was to increase the interaction between the key actors in the innovation, identified as firms, universities and, to a lesser extent, financial organizations. this observation was the more important because it ran against the linear model – dominant at the time – because the paradox indicates that inputs, such as a strong scientific knowledge base, does not automatically leads to outputs, such as competitive advantage in new industries. the notion of the european paradox, and in particular its emphasis on the need to strengthen the commercialization of university research, was later questioned by dosi et al. (2006) and bonaccorsi (2008), who provided evidence that the loss in technological strength of europe compared to the u.s. is not due to the insufficient translation of otherwise excellent scientific research, but to the lack of excellent research as such. indeed, the u.s. does not only excel technologically but also scientifically. this claim was further confirmed by lissoni et al. (2008) who showed, contrary to common belief, that european university professors are almost as active as their american counterparts in translating their research into patents, albeit through other institutional channels. where in the u.s. many patents on findings from public research are taken by universities, in europe these tend to be taken by firms as measured by authorship of university professors on firm patents. this is part of the evidence that europe’s relative lack of innovation compared to the u.s. may not be due to insufficient translation of scientific knowledge into innovation, but rather in producing top science in the first place. this perspective has been echoed more recently by mazzucato (2013) who emphasized that the innovation success of the u.s. lies much more in patient government funding for missionoriented research, especially in the health and military domains, than in anything else. the longterm commitment of u.s. government to particular areas in science and technology, allows for complexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ 38 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press risky endeavors in large research programs. she pleas for a more proactive innovation policy where government provides visions and commits long-term resources for particular technologies, for example, in stimulating nanotechnology or green energy technologies. however, it remains unclear whether policies that are effective in u.s. with a well-integrated internal market, top private universities and a large military sector, will be equally effective in european countries where such conditions are not met. the debate about the european paradox illustrates that the innovation system concept provides a useful policy frame to go beyond market failure reasoning alone, but also that the concept remains theoretically underdeveloped and, hence, lack explanatory power. in a fundamental sense, the theoretical weakness of the innovation system notion is inherent to its static perspective. national differences in the nature (product, process) and rate (more or less) of innovation are attributed to rather sketchy macroscopic descriptions without a theoretical model how generic institutions – here meaning the rules and laws guiding interactions between actors – would govern collaborative innovation processes. instead, historical case studies tend to show that a major part of innovation process – especially concerning the more radical types – concerns the intentional efforts to change and adapt institutions in ways supportive of particular innovations. put simply, institutions should be thought of not only as an independent variable, but also as dependent variable. that is, institutions ‘co-evolve’ with technical change and market conditions (van den belt & rip, 1987; nelson, 1995; murmann, 2003).1 4. a complexity-theoretic perspective in recent years, a third perspective on innovation policy is emerging that has its roots in evolutionary economics, and links to complexity theory in two fundamental ways. first, it advocates a different notion of knowledge than the ones underlying market-failure and systemfailure approaches. second, it conceives of a complex and country-specific space of opportunities and barriers for innovation. combining these insights, one can argue that the additionality of any innovation policy is not just uncertain but likely to be low, with the exception of innovation policy for societal challenges. let us start by recalling the theories of knowledge underlying contemporary innovation policies. in the market and system failure perspectives, such a theory of knowledge remains rather implicit. in the neoclassical framework underlying the market-failure arguments, the notion of knowledge is primarily one of codified knowledge. indeed, if knowledge is codified, it can easily spillover to competing firms. writing this knowledge down into a legal document such as a patent, would then be in principle an effective instrument to protect new knowledge. the innovation-system framework underlying the system failure argument has a more subtle notion of knowledge. first, it is recognized that knowledge is distributed among many actors that all may contribute critically to innovation processes. second, it is also recognized that a lot of knowledge relevant to innovation is tacit in nature and resides in people in the form of experience knowledge as well as organizations in the form of organizational capabilities. hence, to combine tacit knowledge in people and organizations with other people and organizations, collaboration within and across organizations is key as well as stable institutions such as laws, norms and public policies to support such collaborations. the evolutionary perspective, and later its complexity-theoretic elaboration, follows the innovation-system framework in its view of knowledge as distributed and consisting of tacit and codified knowledge (nelson & winter, 1982). where it differs from the innovation-system approach, however, is that it views innovation primarily as a process where cognitive distances complexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ 39 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press between individuals are to overcome rather than institutional distances between organizations within an innovation system. cognitive distances are most easily overcome if individuals have related knowledge base, for example, work in the same or technologically related industries as well as if individuals share the same norms and values (nooteboom, 2000; frenken, et al., 2007). by associating and recombining pieces of knowledge and technology in new ways, new products and services are being conceived that may, or may not, be successful. collaboration central to the innovation-system perspective is one way of overcoming such cognitive distances, but this can also be done in other ways. for example, some individual firms or entrepreneurs are perfectly capable of re-combining on their own, while in other context recombination critically hinges on collaboration and collaboration within the innovation system. furthermore, informal social networks as these exist in (on-line) communities of practices often crossing national boundaries, can be very effective in organizing innovation as well. the central mechanism in the growth of knowledge is to recombine existing artefacts, knowledge and capabilities in new and more complicated ways (arthur, 2009). following this, a complexity-theoretic perspective emphasizes that, first and foremost, innovation patterns are highly cumulative and path dependent. although the notions of cumulativeness and path dependence go back to the 1980s (dosi, 1982; david, 1985), systematic evidence on how national and regional growth trajectories are constrained by the past, is only of recent date. in particular, hidalgo et al. (2007) were able to show that countries develop their economy by diversifying into export products that are closely related to products they already export. that is, if a country mainly exports shoes and trousers, it is very likely to diversify into related goods (e.g. shirts or socks) rather than into unrelated goods (e.g. toys or televisions). this strongly path dependent nature of diversification in national economic development as shown by hidalgo et al. (2007) also holds true empirically at the sub-national levels (neffke, et al., 2011; boschma, et al., 2013) and equally so for technological and scientific development as measured by patents (kogler, et al., 2013; tavassoli & carbonara, 2014; rigby, 2015; tanner, 2015) and publications (boschma, et al., 2014; heimeriks & balland, 2015). if we think of products as being more or less related, and that this greatly affects the opportunities for countries to diversify and develop, the structure of the ‘product space’ and a country’s position in it, matters. the relatedness of products in the product space can be mapped in various mays, for example by counting how many countries are specialized in the same two products or how many people change jobs between two product industries. whatever method is used, the resulting product space can be thought of as a network with links between related products (hidalgo, et al., 2007). such a network is fundamentally ‘complex’ meaning that some products are very central (i.e. related to many other products) while other are more peripheral (i.e. related to only few products). furthermore, there are modules of products that are interrelated but rather unrelated to all other products. relatedness can even be thought of as asymmetric: it may be easier to diversify from product a into product b, than vice versa (e.g. diversification from airplanes into cars is easier than from cars into airplanes). given the complex nature of the product space, countries located in the center and have ample diversification opportunities while other may be located at the periphery or being locked in into a module. complexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ 40 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press figure 1 : an example of the product space. [source: http://www.chidalgo.com/productspace/network.htm] product relatedness is strongly predictive: the chances that a country starts exporting a product closely related to existing products in a country’s portfolio are much higher than the chances that a country diversifies in more distant products. the interpretation of this evolutionary pattern holds that once a country has developed the capabilities to export particular products, it can easily diversify in related products that require very similar capabilities to produce them. in basic economic terms, the reasoning here is one of economies of scope: diversification into related products allows a country to exploit the knowledge it already has, while diversification into unrelated products would necessitate a country to invest heavily as to acquire new knowledge. it is therefore not surprising that the empirical research just cited showed that most countries and regions have developed new activities closely to existing activities. this is not only a likely pattern but also a sensible pattern, as regions and countries can leverage the knowledge and competencies already present. complexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ http://www.chidalgo.com/productspace/network.htm 41 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press the innovation policy implications of a complexity-theoretic framework are not necessarily in conflict with those motivated by market failures and system failures. indeed, a relateddiversification strategy can be supported by generic r&d subsidies, as well as by strong collaboration within a country’s innovation system. in the case of generic r&d subsidies, those trying to explore new activities are compensated for the risk they take as well as the spillovers they are likely to generate for other firms. firms receiving such subsidies are existing firms producing products making use of an existing knowledge base, and can thus be expected to diversify in closely related products. similarly, system failure policies will strengthen interactions between existing actors who build on existing strengths within an innovation system. thus, a government pursuing market and system failure policies, will almost automatically promote related rather than unrelated diversification. thus, although the outcome is predictable, generic policies do not have to pre-specify this direction and thus have the advantage that the information problem (which technology to choose) is circumvented. however, there are two fundamental weaknesses of market and system failures policies from a complexity perspective. first of all, though the market and system failure policies are in essence generic, large firms benefit more than small firms, as large firms do on average more r&d and have formal r&d departments to which tax benefits apply. similarly, manufacturing will profit more than services, as the latter has little formal r&d activities and hence typically fall outside the scope of r&d subsidies. hence, generic r&d subsidies are most likely to reinforce existing manufacturing specializations dominated by large firms. in the case of policies geared to repair system failures, again larger firms will profit most, as they are better embedded in innovation system and can bear the sunk costs involved in participating in such systems. furthermore, with much emphasis on improving university-industry relations, science-based sectors (mainly hightech manufacturing) are privileged over creative industries and service sectors, though demand as well as employment is much higher in the latter sectors than in the former. though r&d activities coordinated within an innovation system may well lead to true new innovations, most likely these innovations will be closely related to existing strengths of incumbents in such a system, simply because innovation systems are built up by incumbent actors largely setting agenda’s on their own. put differently, a government pursuing solely market and system failure policies will generally de facto lead to a stimulus for large manufacturing firms to diversify into related industries. innovation policy, thus, de facto is rather directed towards certain actors and sectors, rather than truly generic including startups and small firms and including creative industries and service sectors. this also explains the poor additionality of such innovation policies. large manufacturing firms have both the resources, appropriability conditions and organizational capabilities to set-up innovation programs on their own, and with relevant partners within the innovation system. and since marketand system-failure policies will mainly lead to diversification into related products, the inherent risks associated with r&d are actually quite low, which suggests that most innovations would also been pursued in the absence of these policies. a second fundamental weakness of market and system failures policies holds that a country that continues to develop only by diversifying into related activities runs a serious risk of running out of opportunities. it will become increasingly harder to find related activities to exploit, the more related activities have already been explored in the past (saviotti & frenken, 2008). hence, a related diversification strategy should go hand in hand with an unrelated-diversification strategy as well. in other contexts, one also speaks of the exploration versus exploitation and the need for ‘ambidexterity’. if the objective of innovation policy is to help actors diversify into unrelated activities, market and system failures arguments still remain relevant. on the one hand, the market failure argument applies a fortiori given that the risk involved in unrelated diversification is much higher complexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ 42 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press compared to related diversification. that is, actors (firms, consumers, universities, financial institutions) will be even more reluctant to invest in unrelated activities than in related activities, which would imply that r&d subsidies have a stronger rationale in an unrelated diversification policy than in a related diversification policy. however, this risk only applies to a small subset of firms that actually pursue such a strategy, so one would have to abandon a generic r&d subsidy logic (rodrik, 2004). in practice, however, an agency granting r&d subsidies would have a hard time to assess which firms pursues an unrelated or related r&d strategy. for unrelated diversification, the system failure argument also holds a fortiori because it is precisely when innovations deviate from existing knowledge, institutions and user practices, that new institutions have to be built. in a sense, one can say that a new technology-specific innovation system has to be built more or less from scratch (hekkert, et al., 2007; bergek, et al., 2008). however, an unrelated-diversification strategy calls for a different systemic policy than the one that is usually advocated. a policy is needed that aims to change institutions which generally requires involving new actors as well as risk-taking on the part of government itself, rather than to reinforcing existing actor positions, institutional arrangements and government roles. such a policy is difficult, as (parts of) governments may have their own stakes in continuing existing institutional arrangements or be subject to strong lobbies by incumbent actors, and are generally reluctant to experiment with new rules and regulations. one way to allow for more experimentation then would be to decentralize innovation policy and powers to cities and regions, so that a larger variety of new techno-institutional arrangements can be built and evaluated.2 another one would be to create new institutional spaces by socio-technical experiments or introducing the ‘right to challenge’ existing rules by actors who meet public interests by alternative means (european council, 2016). yet another way is to invest more resources into learning how to transplant successful institutional arrangements from one territory to another, or from one sector to another, or from one technology to another (binz, et al., 2014; dawley, 2014). since any economy will benefit in the long run from a mix of related and unrelated diversification, the difficult challenge for governments is to come up with market and system failure policies that on the hand support existing companies and sectors in their quest for the exploitation of their existing competencies in related activities, while on the other hand support those few who wish to break with establish practices and explore more risky paths of potentially much bigger social returns. in stimulating related diversification, the state can rely on generic policies, while in stimulating unrelated diversification, the state has to focus much more on specific policies. though generic and specific market-failure policies may be rather easily combined – for example, by granting all firms doing r&d a (small) generic subsidy, and firms with a very risky investment specific financial support through an investment fund so that potential returns feed back into state finances – the inherent contradictions between related and unrelated diversification systemic policies are harder to solve. a well-functioning innovation system – in the sense of strong coordination and consensus on r&d activities among major actors – is probably also a system that is more reluctant to change the institutions and accept radical innovations that potentially threaten the competitive strength of incumbents. innovation systems that are capable of providing a continuous stream of incremental innovations with firms as active agenda-setting actors and universities as responsive knowledge providers, is obviously not a system where radically new agenda’s will spontaneously emerge advocating to break with past specializations and to diversify into new and unrelated territories. a further problem for any unrelateddiversification policy is how to make an informed choice and politically defend specific priorities given that so many other priorities can be chosen. by contrast, a related-diversification policy does not face these problems as opportunities emerge endogenously in a path dependent manner and complexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ 43 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press political support is almost automatic given that vested interests can perfectly seize the opportunities of related diversification. 5. societal challenges our previous discussion makes clear the rationale for innovation policy is greatest in contexts of unrelated diversification. at the same time, the direction of unrelated diversification is by nature undefined. the notion of cross-specialization introduced by janssen (2015) is a first attempt to solve this apparent contradiction. he advocates an innovation policy aimed at radical innovation by promoting crossovers between unrelated technologies or industries already present in an economy (castaldi, et al., 2015). in this way, one can still build on existing artefacts, knowledge, and capabilities residing in an economy while explicitly aiming to new a very new specialization. such true ‘new combinations’, if successful, may provide a long-term source of competitiveness as other countries that do not share the same specialized knowledge fields being recombined, will find it hard to copy such a success. and given that such a policy still continues to build on an economy’s strongholds, but in a radically new combinations, one expect political support for such policies to exist. a second way to guide radical innovation and unrelated diversification is to move away from existing capabilities viz. interest at the supply-side at the starting point, and alternatively to articulate new needs at the demand side (boon & edler, 2015). the current frame within which such demand-led innovation policies are being discussed, is the european-commission frame of ‘societal challenges’ such as climate change, air quality, sustainable food, aging, smart mobility, internet crime, alienation, obesity, diabetes, analphabetism, mental health, et cetera. there are many missing markets related to these challenges due to the fact that most societal challenges stem from negative externalities or a public-good nature of the problem. demand articulation does not only increases the chances of innovation being accepted and embraced by user publics, it also provides legitimacy and guidance for the innovation process itself. in line with this reasoning, weber and rohracher (2012) mentioned legitimacy, guidance and demand articulation as key ‘transformation failures’, in analogy to market and system failures. in the context of the need for transformation given societal challenges, market and system failure policies lose some of their rationale. these generic policies will have little additionality in solving societal challenges for three reasons. first, it is unlikely that generic policies miraculously lead to solutions for specific societal challenges. second, as the markets associated with societal challenges are either imperfect or completely missing, firms find it difficult and highly risky to try to serve these markets in the first place, as it is uncertain if, and if so how, future markets for such problem will be institutionalized. and, finally, the societal challenges are new, ‘grand’ and complex. hence, solutions will be to a large extent unrelated to technologies and institutions already existing. given our earlier conclusion that generic policies will foster mainly related diversification, it is unlikely to trigger the more radical innovations required to solve these challenges. calls for what mazzucato (2013) called a more ‘entrepreneurial state’ seem indeed justified, given both the urgency and uncertainty of societal challenges. in the context of societal challenges, it is particularly insightful to recall an old text by nelson (1974) who asked the simple question: “if we can land a man on the moon, why can’t we solve the problems of the ghetto?” nelson stressed that the challenge of putting a man of the moon was very different from that of solving problems in ghetto’s or, for that matter, most other societal challenges including climate change, ageing, depression, malaria, floods, congestion, pollution, human trafficking, internet fraud, and so on. the objective to land a man on the moon was wellcomplexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ 44 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press defined and guided technological and scientific research in a clear direction. moreover, the objective (at the time) was broadly supported by the public. through dedicated and specific investments in research and development, the challenge could be met. however, ghetto-like problems are wicked and ill-defined, and how such problems should be defined is even contested (bijker, et al., 1987). furthermore, any progress in solving such problems is hard to monitor rendering experimental intervention and ex post policy evaluation difficult. finally, ghettoproblems are multi-faceted and mainly result from behavioral and social processes that are hard to understand, let alone, to influence by policy. given that most societal challenges are clearly ghetto-type problems, mazzucato’s (2013) call for massive public investment in particular sciences and technologies, or in public enterprises, is questionable. only in some cases can societal challenges be easily translated in clear technological projects with the government as a strong driver. such exceptions include for example battery technology, drugs for rare diseases, or public infrastructures to protect citizens against flooding. however, most of the problems that can be subsumed under societal challenges will not be solved by technological fixes nor by advances in scientific research or state-led policies. rather, a new wave of innovation policies is needed that starts from those parts of society where the challenge is actually present and partial knowledge about it is available, and then translates such a challenge into a well-defined and politically supported problem (boon, et al., 2011). only then, a ‘temporary innovation system’ can be organized effectively, which is dedicated to solving such a problem by addressing both the economic, technological and institutional dimensions of the problem at hand.3 the elaboration of a societal challenge into a concrete innovation objective and policy is a key process. without a concretization of a ‘ghetto-like’ challenge, the problem remains elusive hereby de facto de-politicizing the challenge. by making a challenge operational by formulating a clear objective, the choice for supporting one or another innovation policy objective is made part of a normative process. at the same time, once such a simple objective has been formulates, (absence of) progress can easily be monitored ensuring politicians’ accountability. examples of such objectives are zero deaths in traffic, no homeless sleeping on the streets, no children with obesity, all vehicles zero-emission, et cetera. in this search process, finding an operational objective that finds supports from diverse political parties (often for different reasons) would nevertheless be preferable, because a challenge-led policy is often taking 5-10 years and should be able to survive changes of government. by agreeing on a concrete objective, a temporary innovation can be legitimately organized as the relevant stakeholders can only be identified as a challenge is further specific into a concrete objective. at the same time, the objective sets an agenda – or an ‘innovation trajectory’ if you will – for stakeholders to jointly pursue and a yardstick for evaluation and learning during the innovation process. in this way, a shared and concrete objective greatly reduces the inherent cognitive complexity stemming from the ghetto-type nature of societal challenge as well as the organizational complexity stemming from the distributed nature of expertise among stakeholders. the formulation of a clear objective mobilises existing organisations to work together in findings ways to meet it, through collaboration and coordination of actions fitting in their respective roles and competence. in this context, kuhlmann and rip (2014) speak of tentative governance rather than working from a master plan. that is, the objective is there but the way(s) to achieve is an emergent outcome. a temporary innovation system thus cuts across existing, more institutionalised innovation at sectoral, technological and territorial levels. given the a single guiding objective and its temporary nature, there is no need to institutionalise the temporary innovation system itself, avoiding possible clashes between already existing institutional logics, sunk investments and interests. yet, if particular solutions that emerge from complexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ 45 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press temporary innovation systems are successful, the solutions can effectively become adopted in existing institutional frameworks and may also be the basis as a new export product to countries dealing with similar challenges (dewald & fromhold-eisebith, 2015). finally, note that temporary innovation systems can take many forms, including urban innovation programs, sectoral voluntary agreements, monitoring and labelling instruments, social enterprises, government task forces, citizen movements, online communities, et cetera. the plurality of governance structures is unavoidable and probably necessary, given the variety of technological, institutional and political contexts in which such ‘pop-up’ innovation systems emerge. in some arrangements, governments can take the lead and be directive and financially supportive, while in other contexts the absence of government (and possibly even lack of law enforcement) is most productive. 6. concluding remarks in this essay, i have summarized market-failure and system-failure arguments that constitute the dominant frames in current innovation policy. next, i have proposed a third complexity-theoretic perspective stressing the path dependence of economic development resulting from the tendency to diversity into related rather than unrelated activities. as a consequence, i formulated the need for unrelated diversification policies, especially in light of the grand challenges that societies are facing today. i ended by concluding that given the inherent complex, contested and behavioral nature of societal challenges, a return to technology-push strategies in the tradition of ‘man-on-the-moon’ projects is undesirable except in rare cases. instead, i suggested to translate challenges into concrete objectives and then to build dedicated ‘temporary innovation systems’. for innovation scholars, then, the research challenge regarding temporary innovation systems will be twofold. first, we will have to understand why certain governance structures emerge in certain contexts, and what renders them effective and durable in their intended outcomes. this links to the broader question of how institutions change which is taken up inter alia in the institutional-sociology literature under the label of institutional entrepreneurship (battilana, et al., 2009). second, we may have to transform current innovation systems into ‘meta’ innovation systems that facilitate the setting up of temporary innovation systems including ambitious societal objectives as well as effective structures to diffuse contextualized solutions across territories and sectors. this links to the broader political-science question of policy learning and institutional transplantation. for both questions, it holds that innovation studies as a field has to loosen its traditional prime focus on the economics discipline and has to open up much more to the fields of institutional sociology and political science. acknowledgements: the author wishes thank wouter boon, matthijs janssen and bernhard truffer for comments. the usual caveat applies. this work has been funded under nwo’s vici program. references arrow, k. j. (1962). economic welfare and the allocation of resources for invention, in: r.r. nelson (ed.) the rate and direction of inventive activity. princeton: princeton university press. complexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ 46 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press arthur, w.b. (2009). the nature of technology: what it is and how it evolves. new york: free press. arundel, a. (2001). the relative effectiveness of patents and secrecy for appropriation. research policy 30(4): 611–624. battilana, j., leca, b. & boxenbaum, e. (2009). how actors change institutions: towards a theory of institutional entrepreneurship. the academy of management annals 3: 65–107. bergek, a., jacobsson, s., carlsson, b., lindmark, s., & rickne, a. (2008). analyzing the functional dynamics of technological innovation systems: a scheme of analysis. research policy 37: 407–429. bijker, w.e., hughes, t.p., & pinch, t.j. (1987). the social construction of technological systems: new directions in the sociology and history of technology. cambridge, ma: mit press. binz, c., truffer, b., & coenen, l. (2014). why space matters in technological innovation systems mapping global knowledge dynamics of membrane bioreactor technology. research policy 43(1): 138–155. boldrin, m., & levine, d.k. (2013). the case against patents. journal of economic perspectives 27: 3–22. bonaccorsi, a. (2008). explaining poor performance of european science: institutions versus policies. science and public policy 34: 303–316. boon, w., & edler, j. (2015). the missing links – demand based policy making and instruments in the context of mission orientation: concepts, impacts, governance challenges. paper presented at the annual eu-spri conference, helsinki. boon, w. p., moors, e.h., kuhlmann, s., & smits, r.e. (2011). demand articulation in emerging technologies: intermediary user organisations as co-producers? research policy 40(2), 242–252. boschma, r., heimeriks, g., & balland, p.a. (2014). scientific knowledge dynamics and relatedness in bio-tech cities. research policy 43: 107–114. boschma, r., minondo, a., & navarro, m. (2013). the emergence of new industries at the regional level in spain. a proximity approach based on product-relatedness. economic geography 89: 29–51. castaldi, c., frenken, k., & los, b. (2015). related variety, unrelated variety and technological breakthroughs. an analysis of us state-level patenting. regional studies 49: 767–781. cohen, w. m., goto, a., nagata, a., nelson, r. r., & walsh, j. p. (2002). r&d spillovers, patents and the incentives to innovate in japan and the united states. research policy 31(8): 1349–1367. david, p. (1985). clio and the economics of qwerty. american economic review proceedings 75: 332–337. dawley s. (2014). creating new paths? offshore wind, policy activism, and peripheral region development. economic geography 90(1), 91–112. dewald, u., & fromhold-eisebith, m. (2015). trajectories of sustainability transitions in scale-transcending innovation systems: the case of photovoltaics. environmental innovation and societal transitions 17: 110-125. dewald, u., & truffer, b. (2012). the local sources of market formation: explaining regional growth differentials in german photovoltaic markets. european planning studies 20: 397-420. dosi, g. (1982). technological paradigms and technological trajectories: a suggested interpretation of the determinants and directions of technical change. research policy 11: 147–162. dosi, g., llerena, p., & labini, m.s. (2006). the relationships between science, technologies and their industrial exploitation: an illustration through the myths and realities of the so-called 'european paradox'. research policy 35: 1450–1464. etzkowitz, h., & leydesdorff, l. (2000). the dynamics of innovation: from national systems and "mode 2" to a triple helix of university-industry-government relations. research policy 29: 109–123. european commission (1995). green paper on innovation, brussels: european commission, directorate xiii/d (december). european council (2016). better regulation to strengthen competitiveness, press release, 26 june, brussels. freeman, c. (1987). technology policy and economic performance: lessons from japan. london: pinter. frenken k., van oort f.g., & verburg t. (2007). related variety, unrelated variety and regional economic growth. regional studies 41: 685–697. gaillard-ladinska, e., non, m., & straathof, b. (2015). more r&d with tax incentives? a meta-analysis, cpb discussion paper 309. the hague: cpb netherlands bureau for economic policy analysis. gaillard-ladinska, e. & straathof, b. (2015). will r&d tax incentives get europe growing again? voxeu 20 january 2015. gibbons, m., limoges, c., nowotny, h., schwartzman, s., scott, p., & trow, m. (1994). the new production of knowledge: the dynamics of science and research in contemporary societies. london: sage. hall, p.a., & soskice, d. (2001) (eds.) varieties of capitalism. the institutional foundations of comparative advantage. oxford: oxford university press. hardeman, s., frenken, k., nomaler, ö., & ter wal, a.l.j. (2015). characterizing and comparing innovation systems by different ‘modes’ of knowledge production: a proximity approach. science and public policy 42(4): 530–548. heimeriks, g., & balland, p.a. (2015). how smart is specialisation? an analysis of specialisation patterns in knowledge production. science and public policy, advance access. hekkert, m.p., suurs, r.a.a., negro, s.o., kuhlmann, s., & smits, r.e.h.m. (2007). functions of innovation systems: a new approach for analysing technological change. technological forecasting and social change 74: 413–432. hidalgo, c.a., klinger, b., barabasi, a.l., & hausmann, r. (2007). the product space and its consequences for economic growth. science 317: 482–487. janssen, m. (2015). cross-specialization and structural holes: the case of the dutch top sectors, papers in evolutionary economic geography 15.19, utrecht university. complexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ http://dx.doi.org/10.1093/scipol/scu070 http://dx.doi.org/10.1093/scipol/scu070 47 koen frenken: a complexity-theoretic perspective on innovation policy doi: http://dx.doi.org/10.20377/cgn-41 university of bamberg press klein woolthuis, r., lankhuizen, m., & gilsing, v. (2005). a system failure framework for innovation policy. technovation 25: 609–619. kogler, d.f., rigby d.l., & tucker, i. (2013). mapping knowledge space and technological relatedness in us cities. european planning studies 21: 1374–1391. kuhlmann, s., & rip, a. (2014). the challenge of addressing grand challenges, mimeo, university of twente, january. lissoni, f., llerena, p., mckelvey, m., & sanditov, b. (2008). academic patenting in europe: new evidence from the keins database. research evaluation 16: 87–102. mazzucato, m. (2013). the entrepreneurial state: debunking public vs. private sector myths. london uk: anthem press. murmann, j.p. (2003). knowledge and competitive advantage: the coevolution of firms, technology, and national institutions. new york: cambridge university press. neffke f., henning m., & boschma, r. (2011). how do regions diversify over time? industry relatedness and the development of new growth paths in regions, economic geography 87: 237–265. nelson, r.r. (1974). intellectualizing about the moon-ghetto metaphor: a study of the current malaise of rational analysis of social problems. policy sciences 5: 375–414. nelson, r.r. (1995). co-evolution of industry structure, technology and supporting institutions, and the making of comparative advantage. international journal of the economics of business 2: 171–184. nelson, r.r., & winter, s.g. (1982). an evolutionary theory of economic change. cambridge ma: belknap press. nooteboom, b. (2000). learning and innovation in organizations and economies. oxford: oxford university press. rigby, d. (2015). technological relatedness and knowledge space: entry and exit of us cities from patent data. regional studies 49(11), 1922–1937. rodrik, d. (2004). industrial policy for the 21st century, kennedy school of governance working paper 04-047, harvard university. saviotti, p.p., & frenken, k. (2008). trade variety and economic development of countries. journal of evolutionary economics 18: 201–218. solow, r.m. (1957). technical change and the aggregate production function. review of economics and statistics 39(3), 312–320. tanner, a.n. (2015). the emergence of new technology-based industries: the case of fuel cells and its technological relatedness to regional knowledge bases. journal of economic geography, advance access. tavassoli, s., & carbonara, n. (2014). the role of knowledge variety and intensity for regional innovation. small business economics 43(2), 493–509. van den belt, h. & rip, a. (1987). the nelson-winter-dosi model and synthetic dye chemistry, in: w. bijker, t.p. hughes, & t. pinch, (eds.), the social construction of technological systems. cambridge, ma: mit press; 135-158. weber, m., & rohracher, h. (2012). legitimizing research, technology and innovation policies for transformative change: combining insights from innovation systems and multi-level perspective in a comprehensive ‘failures’ framework. research policy 41(6): 1037–1047. endnotes 1 compared to the national innovation system approach, the varieties of capitalism approach (hall & soskice, 2001) is more grounded in the theoretical notions of institutional complementarities and dynamic capabilities. however, as for the national innovation system approach, this approach still has to incorporate the institutional change in the overall framework. 2 in this context, the rise of solar energy technology in germany is a case in point. the national development of the industry was prompted by very localized and bottom-up processes of users and housing corporation pressuring local governments for new policies and regulation (dewald & truffer, 2012). only later, such initiatives were aggregated into national policies. 3 my notion of temporary innovation system fits with an earlier knowledge of mode 2 knowledge production, contrasting the traditional mode 1 notion of independent ‘ivory tower’ university research with new notions of academic engagement, open innovation and temporary network organization (gibbons, et al., 1994; hardeman, et al., 2014). complexity, governance & networks (2017) special issue: complexity, innovation and policy (2017) 35-47 http://dx.doi.org/ http://link.springer.com/journal/11077 conducting content analysis of documents in network research: a review of recent scholarship complexity, governance & networks-vol 02, issue 01 (2015) 83–102 83 doi: 10.7564/14-cgn11 conducting content analysis of documents in network research: a review of recent scholarship qian hu, ph.d. assistant professor, school of public administration, university of central florida, hpa ii, suite 238 email: qian.hu@ucf.edu in this study i examine how researchers have conducted content analysis of documents to collect and analyze network data in public administration. content analyses of documents help us understand the contexts and meanings of network ties. researchers can also analyze documents to examine the changes of network ties over time. the types of documents that researchers have used to draw network data are diverse, but the ways they are used are limited. the majority of articles reviewed for this study focused on drawing interorganizational collaboration data, whereas in a few articles content analysis was conducted to draw policy network data. in this study i found that most of the existing network research has converted qualitative data extracted from documents to numerical data for quantitative analysis and has not fully taken advantage of the qualitative data. researchers can also benefit from providing explicit coding rules and procedures to ensure the quality of network data collection. keywords: documents, content analysis, network analysis, network scholarship 1. introduction social network analysis (sna) refers to a method and set of tools used to analyze social structures and interactions within social structures, relational processes, and outcomes of these processes (borgatti, everett, & johnson, 2013; scott, 2013; wasserman & faust, 1994). over the past few decades, sna has garnered increasing attention from researchers in public administration (provan & lemaire, 2012). researchers applied sna to examine a wide array of management problems and policy issues. a number of authors reviewed the development of network scholarship in public administration and identified areas for further research (e.g., berry et al., 2004; isett, mergel, leroux, mischen, & rethemeyer, 2011; lecy, mergel, & schmitz, 2013; provan, fish, & sydow, 2007). a few researchers have called for the use of a mixed-methods approach to enhance the validity and reliability of network research (berry et al., 2004; hollstein, 2011; isett et al., 2011). yet relatively few studies reviewed the use of specific methods for network data collection and analysis in the public administration literature. 56317_14-11.indd 83 05/10/15 5:09 pm 84 qian hu / conducting content analysis of documents the studies show that use of quantitative approaches such as survey methods still dominates network research, and the use of qualitative approaches remains limited (hollstein, 2011). documents, especially government documents such as emergency management plans, can serve as important guidelines for public management practices. in network research, scholars can use documents as primary data sources or supplementary data sources. in this study my aim was to examine the status and potential of using content analysis of documents in network research. two research questions were addressed: how did researchers conduct content analysis of documents to collect network data? how can researchers better utilize documents and content analysis to contribute to existing network research? i identified 18 articles in public administration journals that had used content analysis of documents as their primary network data collection methods. through a content analysis of the 18 articles, i found that only few researchers have tapped into new forms of documents such as social media data for network data collection. i also found that interorganizational collaborations was the main focus of these studies. in only few studies, the researchers collected longitudinal network data from documents. with regard to the content analysis process, i found that there was not sufficient information about the coding process of content analysis in existing network research. after presenting the findings, i  make suggestions about conducting content analysis of documents for network research. the following section discusses the use of documents and content analysis in public administration, in general, and in network research. the data and methods section details the process of selecting the 18 articles and the analyses conducted. the last section reports the key findings and discusses the advantages and caveats of using content analysis of documents for network research. 2. content analysis of documents in network research both quantitative and qualitative approaches are used in network analyses. quantitative approaches have strengths in investigating large-scale network formation and structures, testing theories, and explaining networking patterns. qualitative approaches allow for exploring new or understudied network phenomena and observing actual networking behaviors, as well as studying network patterns and processes. qualitative approaches can also help researchers understand actors’ interpretations and perceptions of networks, and explore behind-the-scenes reasons for the formation of network ties and the dynamics of networks (hollstein, 2011). qualitative data go beyond numerical network data to provide information about why actors participate in networks and about how they interact with other actors and engage in network activities. qualitative approaches have strengths in providing contextual information and in assisting the interpretation of quantitative network measures. qualitative approaches can also provide more in-depth information about the “content,” “quality,” “meaning,” and “changes” of ties (edwards, 2010, p. 18). some researchers suggest that both approaches should be used in mixed-methods designs. they have also shown that mixed-methods designs allow for data triangulation to improve data, explanatory power, and generalizability (hollstein, 2011). 56317_14-11.indd 84 05/10/15 5:09 pm qian hu / conducting content analysis of documents 85 there is an array of qualitative research strategies for network data collection and analysis. for instance, researchers can use observations, interviews, and documents to collect network data (hollstein, 2011). in this study i focused on the use of documents and content analysis for network research. documents, including government plans and reports, organizational publications, journal articles, newspaper articles, and websites, can provide useful information for researchers to address a variety of questions. content analysis refers to a technique that systematically analyzes textual data for “making replicable and valid inferences from texts (or other meaningful matter) to the contexts of their use” (krippendorff, 2012, p. 24). content analysis is an investigative tool that includes the use of explicitly formulated rules and procedures for the coding process (bowen & bowen, 2008). when they conduct content analysis of texts, researchers need to develop their coding schemata by clarifying their research purposes and by defining their coding units, coding themes, coding categories, and coding rules. many public administration scholars analyzed documents to address a variety of questions (bowen & bowen, 2008), and many conducted content analyses of publications to examine key research topics and the evolution of research in the field. for example, bingham and bowen (1994) conducted a content analysis of a sample of publications from public administration review from 1940 through 1991 to identify “mainstream” public administration topics. lan and anders (2000) conducted a content analysis of 634 articles in eight public administration journals from 1993 through 1995 to examine key subjects, academic backgrounds of public administration scholars, epistemological approaches, data collection and analysis methods, and identifiable paradigms. in addition to examining the development of research in the field, scholars also have analyzed government documents to understand management practices and policy implementation. yang and melitski (2007) analyzed 10 state information technology strategic plans to examine the value orientation embedded in these strategic plans. to evaluate the implementation of public management reform, moynihan (2006) conducted a systematic content analysis of managing for results documents, including government budgets, strategic plans, and performance reports. yet, these content analyses were conducted to draw conventional management or policy implementation data, which do not involve interactions between actors or relational data. in network research, documents can also serve as primary sources of information or can complement existing network data. documents and archival data are relatively easy to access and cost less than a large-scale survey approach for data collection (hollstein, 2011). the availability of large stores of social media data presents new opportunities for researchers to draw unique network data and to explore new research questions (xu & li, 2013). although public administration has seen an increase in the use of documents and content analyses in network research, their uses remain limited, especially compared with the use of surveys and interviews. the content analyses of documents can help researchers identify lists of participants for conducting surveys or qualitative interviews, draw network connections between actors, and examine network structures (e.g., kapucu, 2006; kapucu & demiroz, 2011). for instance, an analysis of a situation report 56317_14-11.indd 85 05/10/15 5:09 pm 86 qian hu / conducting content analysis of documents of a disaster response may detail interorganizational interactions in response to a disaster (e.g., kapucu, 2006). an analysis of an inter-local service delivery report may document an intergovernmental collaboration in delivering public services (e.g., andrew, 2009). an analysis of an organizational directory may present information about the interlocking boards (vidovich & currie, 2012). despite the well-established need for more qualitative network research and the availability of relevant documents in various formats (berry et al., 2004; hollstein, 2011), documents remain an underused source of information for network research. few articles have evaluated the use of documents and content analysis in network research. to fill this gap, in this study i present a systematic review of the existing network research that used documents and content analysis. i do not only evaluate the quality of network data drawn from documents but also identify the potential and challenges in conducting content analysis of documents for network research. 3. data and methods the selection of journals and the search for relevant network articles was a collaborative effort. to identify the network articles that used content analysis of documents and focused on management problems or policy issues, we searched 39 public administration journals (for a detailed description of the selection of the journals and the search process, see kapucu, hu, and khosa, 2014). we adopted the list of 39 public administration journals based on bernick and krueger’s (2010) and forrester and watson’s (1994) comprehensive rankings of public administration journals. the journals we selected publish articles on broad topics in public administration or public policy, as well as those on the subfields of public administration, including public budgeting and finance, public personnel administration, and public organization studies (forrester & watson, 1994). the list of the 39 journals is presented in appendix a. in four steps we identified relevant network research articles that analyze substantive management, policy, and governance issues and excluded those that used networks only as a metaphor. first, we searched the titles, abstracts, and keywords of publications in all the 39 journals published from 1997 to 2012, using words “networks,” “network analysis,” “collaboration,” and “collaborative.” in this first step we found 1,279 articles in relevant journals. second, we read the abstracts of the 1,279 articles to identify those articles that focused on networks. the list was narrowed down to 677 articles focusing on networks, excluding articles that discussed the broader themes of collaborative governance and collaboration. third, we identified those articles that used sna as part of their methodology. this step yielded a total of 136 articles. the last step was to identify those articles that discussed networks related to the contexts of public administration and public policy. we reviewed the articles carefully to ensure that each article had used sna as part of its methods and discussed topics that were relevant to the public organizational or policy context. through the systematic search, we identified a total of 76 articles that used sna in their methods (kapucu, hu, & khosa, 2014). 56317_14-11.indd 86 05/10/15 5:09 pm qian hu / conducting content analysis of documents 87 to select network articles for further analysis in this article, i read all the 76 articles to identity articles in which the researchers used documents and content analysis in their data collections and analyses. i found 18 articles out of the 76 articles. to be included in this study, documents had to have been used to generate primarily network data. the articles that used documents to generate a list of participants for their primary data collection, such as interviews and surveys, were excluded. articles that analyzed documents to produce supplementary organizational attributes data were also excluded. table 1 lists the journals and the number of network articles that used documents and content analysis to collect primarily network data. the journal of public research and theory has three articles, followed by administration & society, the american review of public administration, and the journal of urban affairs, each with two articles. i read each of the 18 articles carefully and coded through an open coding process in combination with pre-established coding themes (bowen & bowen, 2008). i examined the types of documents considered in each article, the kinds of network data that researchers drew from the documents, and the analysis of the network data that researchers conducted. i conducted open coding to identify the research topics and the types of documents covered in each article by examining research questions, methods, and key findings. during the open coding process, i paid special attention to the title, abstract, and introduction of each article to identify its research focus. as i read the articles, i created tentative labels that summarized what the research was mainly about. for instance, article a may have focused on economic development issues, while article b may have examined emergency management. a diverse range of research foci emerged as the coding process continued. i read the data collection sections of each article to identify the data source and the type of table 1 list of journals and the number of network articles journals number of network articles that used documents and content analysis administration & society 2 the american review of public administration 2 journal of management studies 1 journal of urban affairs 2 journal of policy analysis and management 1 journal of public research and theory 3 nonprofit management and leadership 1 nonprofit and voluntary sector review 1 organization studies 1 public administration review 1 public performance and management review 1 policy studies journal 1 urban affairs review 1 total 18 56317_14-11.indd 87 05/10/15 5:09 pm 88 qian hu / conducting content analysis of documents network data that it discussed. i organized the coded data and saved them in spreadsheets for further comparisons, categorization, and analyses. i used the categories and concepts from existing network research that were already established (e.g., borgatti et al., 2013; isett et al., 2011; provan et al., 2007) to analyze and reorganize the key topics and document types into different categories. i used borgatti, everett, and johnson’s (2013) book, not only because it was the most frequently cited social network analysis book in public administration, but, more important, their book discussed the types of data sources and methods of network data collection. following borgatti and his colleagues’ categorizing of oneand two-mode network data, i also coded the network data in the 18 articles into different categories. similarly, i examined whether the network data involved multiplex ties, whether the data were longitudinal, and whether mixed-methods were used in data collection. these were the pre-established categories used in the coding process to further understand the network data collected through content analysis of documents. i used provan, fish, and sydow’s (2007) review article to help further examine the management and policy domains that the network data cover. i used their definition of “interorganizational networks” as another pre-established category to be differentiated from “policy networks.” interorganizational networks focus on interorganizational interactions in delivering public services to pursue public values (isett et al., 2011; provan et al., 2007), whereas policy networks examine the power relationships between policy makers and the relationships between the structural arrangements of policy networks and policy outcomes (keast, 2014; rethemeyer & hatmaker, 2008). although the research foci for these two types of networks differ, the two types of networks are related to and influence each other. in addition, i used the evaluation criteria from krippendorff (2012) as the coding themes to evaluate the quality of existing content analysis in network research. the evaluation criteria included whether the articles presented information about coding units, coding themes, coding categories, and other coding rules (krippendorff, 2012), which helped evaluate the reliability and validity of the content analysis. 4. findings and discussion in this section i identify the types of documents that were analyzed in the existing network research, present the research foci of these studies, and discuss the types of network data drawn from these documents. then, i examine the challenges of drawing network data from documents. 5. types of documents scholars used various kinds of organizational documents to collect network data, as seen in table 2. government documents or records served as primary sources of information in 10 out the 18 articles examined here. in six articles, researchers presented network 56317_14-11.indd 88 05/10/15 5:09 pm qian hu / conducting content analysis of documents 89 table 2 types of documents and data drawn from the documents types of documents number of articles data drawn from the documents organizational documents interlocal service delivery reports, contract records, service catalog; joint service database 3 • contractual arrangements in law enforcement (andrew, 2009) • local government collaboration in service delivery (leroux & carr, 2010) • contractual relationships in public transportation (ponomariov, et al., 2011) comprehensive emergency management plans 6 • lead agencies according to the emergency management plan (choi & brower, 2006) • key actors and organizational interactions in the planned emergency management network (kapucu, 2006; kapucu et al., 2010; kapucu et al., 2009; kapucu & demiroz, 2011; kapucu & van wart, 2006) after-action reports, situation reports from fema 5 • key actors and organizational interactions in the actual emergency response (kapucu, 2006; kapucu et al., 2010; kapucu et al., 2009; kapucu & demiroz, 2011; kapucu & van wart, 2006) organizational publications, reports, directory, and database, and other documents 6 • presence of collaborations, formal meetings, strategic alliances, and joint projects among colleges (akkerma et al., 2011) • organizational involvement in public health cooperation on the united states-mexico border (collins-dogrul, 2012) • policy actors’ policy preference about swiss climate policy (ingold & varone, 2012) • nongovernmental organizations’ network affiliations and other attributes data in mexico (neal, 2008) • interlocking ties among companies in swiss (ruigrok, peck, & keller, 2006) • professional membership among danish mayors (villadsen, 2011) newspaper articles new york times, other newspapers 6 • key actors and organizational interactions in the actual emergency response (kapucu, 2006; kapucu et al., 2010; kapucu et al., 2009; kapucu & demiroz, 2011; kapucu & van wart, 2006) • policy actors in germany during the economic reform between 1990 and 1994 (raab, 2002) websites federal election commission website, company websites 3 • donation and voting patterns (jasny, 2012) • interlocking boards between nonprofit and private organizations in perth, australia (vidovich & currie, 2012) • professional membership among danish mayors (villadsen, 2011) 56317_14-11.indd 89 05/10/15 5:09 pm 90 qian hu / conducting content analysis of documents information in other types of organizational documents, such as publications, reports, directory information, and databases. the researchers analyzed newspaper articles along with other organizational documents to collect information about interorganizational collaborations in emergency management. compared with the use of organizational official documents, there is limited use of other sources of documents such as websites and social media data. in only two articles the researchers analyzed websites to collect network information. despite the rapid growth of social media data, researchers still rely heavily on the content analysis of conventional forms of documents for network data collection in public administration. 6. research foci and network data drawn from documents in recent years, researchers drew various types of network data from documents in public administration to address a variety of questions. as table 2 shows, there are various types of network data drawn from documents. the 18 articles included in this study addressed various management and policy issues, ranging from emergency management to economic development, transportation, education, and public health. a wide range of management and policy domains. as figure 1 shows, among the 18  articles, 6 (33%) were on emergency management. in the field of emergency management, scholars analyzed governmental documents such as comprehensive emergency management plans (cemps) of local governments, after-action reports, and situation reports from the federal emergency management agency (fema) to collect network data (e.g., choi & brower, 2006; comfort, 2007; kapucu, 2006; kapucu, arslan, & collins, 2010; kapucu, augustin & garayev, 2009; kapucu & demiroz, 2011; figure 1. types of network data. 56317_14-11.indd 90 05/10/15 5:09 pm qian hu / conducting content analysis of documents 91 kapucu &  van wart, 2006). scholars examined cemps to identify lists of primary or and support agencies in the face of emergencies. furthermore, in the analyses of cemps researchers examine the planned network of emergency management and the interorganizational interactions defined in the documents (choi & brower, 2006). analysis of afteraction reports and situation reports can capture the actual response networks, involving the interactions between organizations during and after disasters (comfort & kapucu, 2006; kapucu, 2006). the analysis of newspaper articles such as those from the new york times can help capture cross-sector collaboration in emergency management, especially the engagement of nongovernmental actors in response to disasters. scholars examined not only the lead organizations but also the structural characteristics of the actual response networks (kapucu & demiroz, 2011). in addition, networks may operate differently in urgent situations than in routine environments. using content analysis to capture how networks respond to external environmental uncertainty can help better measure the performance of networks under stress. the relatively high number of articles on emergency management networks can be attributed partly to the availability of government emergency management plans and other government records, as well as the fast-growing research interest in this topic, especially after the terrorist attacks of september 11, 2001. interlocal public service delivery and intergovernmental collaboration is another area where scholars applied content analysis to collect data. scholars analyzed governmental service records such as interlocal service delivery reports, contract records, service catalogs, and joint service databases to examine intergovernmental and cross-sector collaborations in a wide variety of areas (andrew, 2009; leroux & carr, 2010; ponomariov, kingsley, & boardman, 2011). andrew (2009) examined interlocal service delivery reports to collect information about contractual arrangements in law enforcement among 66 local governments in the orlando-kissimmee metropolitan area. he examined the contractual ties in various formats—service agreements, contracts and lease agreements, memoranda of understanding, mutual-aid agreements, and informal letters of agreement. he analyzed the patterns of contractual arrangements over five time periods from 1986 to 2003. leroux and carr (2010) analyzed interlocal government agreements among the 44  local governments of wayne county, in michigan, in system maintenance functions (e.g., road and bridge construction and repair) and four lifestyle services (e.g., economic development and public safety). in this study the existence of interlocal government agreements was considered as a network tie between local governments. similarly, ponomariov, kingsley, and boardman (2011) examined the contract records of a transportation agency and analyzed the contracting relationships among a state transportation agency, contractors, and subcontractors in engineering design services between 1992 and 2003. these three articles by andrew (2009), leroux and carr (2010), and ponomariov et al. (2011) each collected longitudinal network data about interorganizational collaborations. scholars used other organizational documents such as organizational reports or directories to draw network data on organizational engagement in interorganizational collaborations. for instance, akkerma, torenvlied, and schalk (2011) examined the written reports and meeting minutes of colleges to identify the presence of collaborations, formal meetings, strategic alliances, and joint projects among five danish colleges. collins-dogrul (2012) 56317_14-11.indd 91 05/10/15 5:09 pm 92 qian hu / conducting content analysis of documents examined organizational databases, publications, and records to investigate organizational involvement in public health cooperation on the united states-mexico border. neal (2008) examined the directory of 173 nonprofit organizations in oaxaca, mexico, to identify the nonprofits’ connections with government authorities and other nongovernmental organizations and the impact of party politics on the nonprofit organizations’ network connections. this study did not specify the types of interactions that nonprofit organizations have with others. scholars have analyzed documents to examine the interlocking ties among organizations. ruigrok, peck, and keller (2006) examined the annual reports of 217 swiss companies to understand the interlocking directorships. vidovich & currie (2012) researched interlocking boards between nonprofit and private organizations in perth, australia, by examining the websites of nonprofit organizations and companies. to examine danish mayors’ affiliations with professional organizations, villadsen (2011) examined professional journals and other sources “including municipality associations, the political parties, municipality homepages, ministries, local media, article databases, and directories of business information” (p. 585). policy is another important domain where researchers used documents to generate network data. raab (2002) identified the key policy actors in germany during the economic reform between 1990 and 1994 by examining newspaper articles and other publications. ingold and varone (2012) examined official documents (not specified) to understand the preferences of the actors involved in the formation of swiss climate policies. jasny (2012) analyzed the websites of the federal election commission and other documents to examine donation behaviors and voting patterns among congressional representatives. compared with research on interorganizational collaboration, the research on policy networks is limited. variety of network data drawn from documents. through content analyses of documents, researchers can draw various kinds of network data and pull both one-mode network data as well as two-mode network data. one-mode data refers to those datasets that include relational data for the same set of actors, whereas two-mode data involve two different sets of actors. as table 3 shows, the majority of the 18 articles i analyzed focused on one-mode network data, whereas only seven articles collected two-mode data. although the types of network data are diverse, the study of interorganizational collaboration is the focus of many studies. in eleven of the 18 articles the researchers presented interorganizational collaboration data in a wide range of management and policy domains. researchers examined the existence of collaboration ties and organizational involvement in collaborative efforts. the examination of documents allowed researchers to study interorganizational collaborations in different formats, ranging from formal contractual relations to informal collaboration agreements, strategic alliances, and joint projects (akkerman, torenvlied, & schalk, 2011; andrew, 2009). there are fewer studies on the formation and development of policy networks, which may be partially explained by the unavailability of appropriate policy documents and the challenge of coding policy documents. selecting the appropriate documents might not be easy, as researchers need to justify the selection of particular policy documents instead 56317_14-11.indd 92 05/10/15 5:09 pm t ab le 3 a n a na ly si s of d at a d ra w n fr om t he d oc um en ts a ut ho rs a re a of f oc us m od es o f d at a c on te nt o f n et w or k m ul ti pl ex it y of n et w or k t ie s l on gi tu di na l da ta m ix ed -m et ho ds a kk er m a, e t al . ( 20 11 ) in te rc ol le ge n et w or ks an d gr ad ua te s at is fa ct io n o ne m od e an d t w om od e p re se nc e of “ co ll ab or at io ns , f or m al m ee ti ng s, s tr at eg ic a ll ia nc es , an d jo in t p ro je ct s, ” bo ar d an d co m m it te e m em be rs hi p (p . 6 61 ). y es n o y es , c ro ss -v al id at ed w it h in te rv ie w d at a a nd re w ( 20 09 ) in te rl oc al g ov er nm en t co ll ab or at io n in p ub li c sa fe ty o ne -m od e p re se nc e of c on tr ac tu al t ie s in va ri ou s fo rm at s n o y es n o c ho i & b ro w er , 2 00 6; p la nn ed , p er ce iv ed , a nd ac tu al e m er ge nc y m an ag em en t ne tw or ks o ne -m od e in te ro rg an iz at io na l in te ra ct io ns de fi ne d on t he e m er ge nc y m an ag em en t pl an s n o n o y es , c om pa re d w it h su rv ey d at a c ol li ns -d og ru l (2 01 2) c ro ss -b or de r co op er at io n in p ub li c he al th o ne -m od e “a ss oc ia ti on al a nd b eh av io ra l in te ra ct io n of m ex ic an a nd u s or ga ni za ti on ” (p . 1 01 0) . n o y es y es , c om pl em en te d w it h in te rv ie w d at a in go ld & v ar on e (2 01 2) p ol ic y br ok er s, th ei r be ha vi or s, a nd im pa ct on p ol ic y m ak in g w it h th e co nt ex t o f w it hi n th e s w is s cl im at e po li cy t w om od e p ol ic y pr ef er en ce s of p ol ic y m ak er s n o y es y es , c om pl em en te d w it h in te rv ie w d at a ja sn y (2 01 2) r el at io ns hi p be tw ee n vo ti ng p at te rn s an d do na ti on t w om od e v ot in g be ha vi or s of co ng re ss pe rs on s y es n o n o k ap uc u (2 00 6) ; k ap uc u et a l. ( 20 10 ); k ap uc u et a l. (2 00 9) ; k ap uc u & d em ir oz (2 01 1) ; k ap uc u & v an w ar t (2 00 6) ; in te ro rg an iz at io na l co lla bo ra ti on s in e m er ge nc y m an ag em en t o ne -m od e t he p re se nc e of i nt er or ga ni za ti on al i nt er ac ti on s in t he pl an ne d ne tw or ks , a nd t he pr es en ce o f in te ro rg an iz at io na l co ll ab or at io ns i n re sp on se t o di sa st er s n o y es n o 93 (c o n ti n u es ) 56317_14-11.indd 93 05/10/15 5:09 pm a ut ho rs a re a of f oc us m od es o f d at a c on te nt o f n et w or k m ul ti pl ex it y of n et w or k t ie s l on gi tu di na l da ta m ix ed -m et ho ds l er ou x & c ar r (2 01 0) in te rl oc al g ov er nm en t co ll ab or at io n in p ub li c se rv ic e de li ve ry o ne -m od e t he p re se nc e of c ol la bo ra ti on i n ei gh t se rv ic e ar ea s y es y es y es , c om pl em en te d w it h in te rv ie w d at a n ea l (2 00 8) n on pr ofi ts ’ co nn ec ti on s w it h go ve rn m en t au th or iti es a nd o th er n on go ve rn m en ta l or ga ni za ti on s o ne -m od e t he p re se nc e of c ol la bo ra ti on o n pr oj ec ts , m em be rs hi p, i nv ol ve m en t in s oc ia l m ov em en t, a nd di al og ue . y es n o n o p on om ar io v et a l. ( 20 11 ) c on tr ac ti ng r el at io ns hi ps in t ra ns po rt at io n o ne -m od e t he e xi st en ce o f co nt ra ct ua l re la ti on sh ip s be tw ee n a go ver nm en t ag en cy a nd o th er or ga ni za ti on s n o y es n o r aa b (2 00 2) p ol ic y ne tw or ks i n g er m an y du ri ng t he e co no m ic r ef or m t w om od e k ey p ol ic y ac to rs n o y es y es , c om pl em en te d w it h in te rv ie w d at a r ui gr ok e t al ., 20 06 in te rl oc ki ng d ir ec to rs hi ps in s w is s co m pa ni es a nd bo ar ds ’ st ra te gi c de ci si on m ak in g t w om od e b oa rd m em be rs ’ af fi li at io n w it h or ga ni za ti on s n o n o y es , c om pl em en te d w it h su rv ey d at a v id ov ic h & c ur ri e (2 01 2) in te rl oc ki ng d ir ec to rs hi ps t w om od e b oa rd m em be rs ’ af fi li at io n w it h or ga ni za ti on s n o n o n o v il la ds en , 2 01 1 d an is h m ay or s’ p ro fe ssi on al m em be rs hi p t w om od e m ay er ’s a ffi li at io n w it h pr of es si on al o rg an iz at io ns n o n o y es , c om pl em en te d w it h su rv ey d at a 94 t ab le 3 a n a na ly si s of d at a d ra w n fr om t he d oc um en ts ( c o n ti n u ed ) 56317_14-11.indd 94 05/10/15 5:09 pm of others. furthermore, unlike emergency management plans and interlocal government service agreements, thorough content analysis of documents related to the formation of policy networks may be more complicated. the policy documents can include a large number of policy stakeholders and involve intricate interactions among policy actors, which may require a labor-intensive coding process in order to draw useful network data. longitudinal network data. compared with survey methods and qualitative interviews, content analysis of documents has strengths in producing longitudinal network data at a reasonable cost, which allows for the examination of the changes in network ties. in table 4 the strengths and weaknesses of content analysis and survey methods are compared. collecting longitudinal network data through surveys and interviews can be expensive and time consuming. during a survey study or an interview, respondents may face challenges recalling the history of an event or a collaborative experience. moreover, it might be difficult to ensure the participation of the same participants in a longitudinal survey or interview study (marsden, 2011). in contrast, researchers can take advantage of archival data and pull network data for needed time periods. researchers have collected longitudinal network ties among organizations by examining organizational documents for certain periods of time (e.g., andrew, 2009; leroux & carr, 2010; ponomariov et al., 2011). for instance, by examining the interlocal service delivery agreements or reports, researchers can pull interlocal collaboration data from each time period and merge the data into panel network data (andrew, 2009). although seven of the 18 articles considered here collected longitudinal data, most of these articles aggregated the data at different points of time, and only few conducted network analysis to examine the change of network ties. qualitative and quantitative analysis. data drawn from documents can be analyzed using both quantitative and qualitative analysis methods. the data allow researchers to conduct statistical network analysis. the data can also provide rich contextual information and the meaning and content of the relational data captured in the documents (edwards, 2010), though the data are not directly reported by the network participants. researchers can apply descriptive network measures such as degree centrality and network density to identify the key actors and examine the network structural characteristics table 4 a comparison between content analysis of documents and survey methods content analysis of documents survey methods strengths • provide information about the contexts, the content, quality, and meaning of network ties (edwards, 2010) • easy to access and cost less (hollstein, 2011). • longitudinal data (edwards, 2010) • draw social media data (groenewegen & moser, 2014) • large-scale datasets, and whole-networks (marsden, 2011) • easy to analyze (marsden, 2011) • standard process and easy to be implemented (marsden, 2011) • address a variety of network questions (such as intensity of collaborations) in an efficient manner weakness • subjective coding process (edwards, 2010) • generalizability issues (edwards, 2010) • concern about response rates • lack of contexts (marsden, 2011) • attrition for longitudinal study 56317_14-11.indd 95 05/10/15 5:09 pm 96 qian hu / conducting content analysis of documents (e.g., kapucu & demiroz,  2011). researchers can also apply sophisticated statistical models such as actor-oriented models and quadratic assignment procedure regression to understand the collaboration patterns among local governments in public safety (e.g., andrew, 2009; leroux & carr, 2010). using qualitative analysis methods, researchers can focus on describing the context, the meaning of network ties, and the process of network formation. for instance, after reviewing newspaper articles and other documents, raab (2002) described the germany economic reform between 1990 and 1994 and explained the involvement of key policy actors in great detail. or, researchers can combine both a quantitative analysis approach with a qualitative approach to analyze network data to provide a quantitative description of the actors and the network as well as detailed contextual background information. most of the existing research has focused more on converting data to numerical forms for quantitative analysis and could take better advantage of qualitative data by providing more contextual information and by interpreting the content or meaning of network ties, as table 4 suggests. 7. challenges of drawing network data from documents applying content analysis to drawing net data from documents can be as challenging as using other quantitative approaches, depending the quantity and types of documents that researchers need to analyze. the meaning of network ties may be not accurately captured through the documentation process, which may lead to differences between what the documents suggest and what the network participants actually mean or would interpret. with survey or documents data, researchers still need to carefully address potential cognitive and perceptual biases of third-party interpreters. furthermore, without systematic coding rules and procedures, the qualitative approach to collecting and analyzing data, including conducting content analysis of documents, may face critiques about its reliability and generalizability (bowen & bowen, 2008). the following will identify some of the areas that deserve more attention to improve the use of documents and content analysis in network research, based on the analysis of 18 instances of network scholarship in public administration. document sources. among the 18 articles included in this study, a few did not specify the types of documents that the authors had used to draw their data, which makes it difficult to examine the validity of using the documents for particular research topics and makes it extremely challenging to replicate the study in other contexts. in addition, the majority of the 18 articles used either government documents or other types of traditional publications, including organizational directories, records, and meeting minutes. few articles used information from websites, let alone social media data. yet, many scholars believe that the era of big data presents both “theoretical and methodological opportunities” for network researchers (groenewegen & moser, 2014, p. 463). social media data can be timely and suitable for social network research, as social media data, by nature, involve relational data. on the other hand, the formation and development of network ties in the online community may differ from their counterparts in offline contexts. this 56317_14-11.indd 96 05/10/15 5:09 pm qian hu / conducting content analysis of documents 97 presents new opportunities to test whether existing theories are still valid in explaining the formation and development of online network ties (groenewegen & moser, 2014). social media data, however, exist in vast quantities and can be rife with unwanted information. computer-aided qualitative data analysis software packages (including various software tools such as nvivo, atlas ti, and qda), known as caqdas, may become necessary for analyzing the large volume of online data. to take advantage of voluminous amounts of text data, including social media data on the internet, researchers may need to learn new tools such as nodexl and python programming language for data-mining purposes. coding process. most of the 18 articles examined in this study did not discuss the coding process they had used in detail. there is insufficient information in these articles about the coding units, coding themes, coding categories, and other coding rules. the authors do not provide enough information about whether they coded the documents by words, phrases, sentences, paragraphs, or whole texts; neither do they provide information about whether the researchers established their coding schema before or during coding documents. moreover, few authors mention whether they conducted coding manually or through some computer software. they do not mention r whether the coding was finished by one author or by collaborative efforts either. even fewer authors reported any intercoder reliability results. all of these factors may make it difficult for researchers with the same research interests to validate the results or replicate the studies in other contexts. to improve the reliability of the coding process and to strengthen the validity of the study designs, researchers may consider providing more explicit coding rules and procedures for closed-ended content analysis in the text or the appendices. for open-ended coding processes, researchers may provide explicit information about how the open-coding was conducted so that other researchers may better understand the process and replicate the process in similar studies. furthermore, the availability of a variety of network analysis tools may help researchers reduce labor needed for manual coding. pfeffer and carley (2012) applied rapid ethnographic assessment (rea), which originated in anthropology, to social network analysis. rea can be used to “quickly create a socio-cultural profile of a group or region,” using “semi-automated computer-based text-mining and machine learning techniques” (carley, bigrigg, & diallo, 2012, p. 330). software such as automap can extract network information from unstructured documents through text preprocessing, generating thesauri, formulating linkage rules, and revealing the system structure of texts (diesner & carley, 2005). for instance, morçöl, vasavada, and kim (2013) used automap to analyze 354 newspaper articles published between 1990 and 2009 to examine the key policy players in an urban governance network. network data extracted using automap can also be analyzed by other network analysis tools such as organization risk analyzer (ora) and ucinet. quality of the network data. content analysis of the documents in the 18 articles produced relational data, most of which focused on the existence of interorganizational collaboration ties or affiliation ties. in seven out of the 18 articles the researchers collected longitudinal network data. a few researchers delved into the network ties and provided more information about the meaning or content of the network ties. 56317_14-11.indd 97 05/10/15 5:09 pm 98 qian hu / conducting content analysis of documents for instance, andrew (2009) examined the formation of collaboration ties to provide 14  kinds of public services related to public safety. and the format of collaboration ties is not limited to formal contractual ties. this level of detail on the intergovernmental collaboration provides useful contextual information for the network analysis. this type of data would be difficult to collect using quantitative methods such as a survey design. document data have more potential to be explored to enrich the conversation about the content and quality of network ties. more diverse types of network data such as semantic network data (a network of concepts) and metanetworks data can be extracted from documents (diesner & carley, 2005; van atteveldt, 2012). furthermore, with the advancement of computer software, archival data could be better utilized for collecting panel network data. in addition, in eight studies the researchers used mixed-methods designs to collect network data, as shown in table 3. most of the eight articles used other methods such as surveys or interviews to collect different kinds of data and to complement the network data drawn from documents. only two articles actually used the interview data or survey data to cross-validate or compare with the network data from documents (e.g., akkerma et al., 2011; choi & brower, 2006). to take advantage of a mixed-methods design, researchers need to better integrate content analysis with other methods. 8. conclusion this article presents an examination of the use of documents and content analysis in existing network research in the field of public administration. the number of studies using documents and content analysis to draw network data is relatively small. yet the content analysis of documents can not only produce quantitative network data but also provide unique contextual information about the formation of network ties. furthermore, documents, especially archival data, can allow researchers to explore the change of network ties. scholars have used documents to extract interorganizational collaboration data and affiliation data and to identify the key policy actors and their policy preferences. more diverse types of network data can be drawn from documents, such as longitudinal network data, to examine the development or evolution of network ties. most of the existing research used formal organizational documents or newspaper data, while social media data remain underexplored. to strengthen the validity and the reliability of the qualitative approach to analyze documents, future researchers may need to provide more explicit coding procedures and rules. this study has some limitations. the results are preliminary and descriptive based on the analysis of 18 network articles. it may raise more questions than it has answered. as the number of studies using documents as the primary source of information continues to grow, a more comprehensive review of qualitative network research will be needed. 56317_14-11.indd 98 05/10/15 5:09 pm qian hu / conducting content analysis of documents 99 references akkerman, a., torenvlied, r., & schalk, j. (2011). two-level effects of interorganizational network collaboration on graduate satisfaction: a comparison of five intercollege networks in dutch higher education. the american review of public administration, doi: 10.1177/0275074011425418. andrew, s. a. (2009). regional integration through contracting networks: an empirical analysis of institutional collection action framework. urban affairs review, 44(3), 378–402. bernick, e., & krueger, s. (2010). an assessment of journal quality in public administration. international journal of public administration, 33(2), 98–106. berry, f. s., brower, r. s., choi, s. o., goa, w. x., jang, h. s., kwon, m., & word, j. (2004). three traditions of network research: what the public management research agenda can learn from other research communities. public administration review, 64(5), 539–552. bingham, r. d., & bowen, w. m. (1994). “mainstream” public administration over time: a topical content analysis of public administration. public administration review, 54(2), 204–208. borgatti, s. p., everett, m. g., & johnson, j. c. (2013). analyzing social networks. los angeles, ca: sage. bowen, c. c., & bowen, w. m. (2008). content analysis. in g. j. miller & k. yang (eds.). handbook of research methods in public administration (2nd ed., pp. 689–704). boca raton, fl: crc press. carley, k. m., bigrigg, m. w., & diallo, b. (2012). data-to-model: a mixed initiative approach for rapid ethnographic assessment. computational and mathematical organization theory, 18(3), 300–327. choi, s. o., & brower, r. s. (2006). when practice matters more than government plans: a network analysis of local emergency management. administration and society, 37(6), 651–678. collins-dogrul, j. (2012). tertius iungens brokerage and transnational intersectoral cooperation. organization studies, 33(8), 989–1014. comfort, l. k. (2007). crisis management in hindsight: cognition, communication, coordination, and control. public administration review, 67(1s), 189–197. comfort, l. k., & kapucu, n. (2006). inter-organizational coordination in extreme events: the world trade center attack, september 11, 2001. natural hazards, 39(2), 309–327. diesner, j., & carley, k. m. (2005). revealing social structure from texts: meta-matrix text analysis as a novel method for network text analysis. in v. k. narayanan & d. j. armstrong (eds.), causal mapping for research in information technology (pp. 81–108). harrisburg, pa: igi global. edwards, g. (2010). mixed-methods approach to social network research. retrieved from http://eprints.ncrm. ac.uk/842/1/social_network_analysis_edwards.pdf forrester, j. p., & watson, s. s. (1994). an assessment of public administration journals. public administration review, 54(5), 474–482. groenewegen, p., & moser, c. (2014). online communities: challenges and opportunities for social network research. in d. j. brass, g. labianca, a. mehra, d. s. halgin, & s. p. borgatti, (eds.), research in the sociology of organizations (vol. 40). bradford, uk: emerald publishing. hollstein, b. (2011). qualitative approaches. in j. scott & p. j. carrington (eds.). the sage handbook of social network analysis (pp. 404–416).thousand oaks, ca: sage. ingold, k., & varone, f. (2012). treating policy brokers seriously: evidence from the climate policy. journal of public administration research and theory, 22(2), 319–346. isett, k. r., mergel, i. a., leroux, k., mischen, p. a., & rethemeyer, r. k. (2011). networks in public administration scholarship: understanding where we are and where we need to go. journal of public administration research and theory, 21(s1), i157–i173. jasny, l. (2012). baseline models for two-mode social network data. policy studies journal, 40(3), 458–491. kapucu, n. (2006). interagency communication networks during emergencies boundary spanners in multiagency coordination. the american review of public administration, 36(2), 207–225. kapucu, n., & demiroz, f. (2011). measuring performance for collaborative public management using network analysis methods and tools. public performance and management review, 34(4), 551–581. 56317_14-11.indd 99 05/10/15 5:09 pm 100 qian hu / conducting content analysis of documents kapucu, n., arslan, t., & collins, m. l. (2010). examining intergovernmental and interorganizational response to catastrophic disasters toward a network-centered approach. administration & society, 42(2), 222–247. kapucu, n., augustin, m. a., & garayev, v. (2009). interstate partnerships in emergency management: emergency management assistance compact in response to catastrophic disasters. public administration review, 69(2), 297–313. kapucu, n., hu, q., & khosa, s. (2014). the state of network research in public administration. administration & society. doi:10.1177/0095399714555752 kapucu, n., & van wart, m. (2006). the evolving role of the public sector in managing catastrophic disasters: lessons learned. administration & society, 38(3), 279–308. keast, r. (2014). network theory tracks and trajectories: where from, where to? in r. keast, m. mandell, & r. agranoff (eds.). network theory in the public sector: building new theoretical frameworks (pp. 15–30). new york, ny: routledge. krippendorff, k. (2012). content analysis: an introduction to its methodology (3rd ed.). thousand oaks, ca: sage publications, inc. lan, z., & anders, k. k. (2000). a paradigmatic view of contemporary public administration research an empirical test. administration & society, 32(2), 138–165. lecy, j. d., mergel, i. a., & schmitz, h. p. (2013). networks in public administration. public management review, 15(1), 1–23. leroux, k., & carr, j. b. (2010). prospects for centralizing services in an urban county: evidence from eight self-organized networks of local public services. journal of urban affairs, 32(4), 449–470. marsden, p. v. (2011). survey methods for network research. in j. scott & p. j. carrington (eds.). the sage handbook of social network analysis (pp. 370–388). thousand oaks, ca: sage. morçöl, g., vasavada, t., & kim, s. (2013). business improvement districts in urban governance: a longitudinal case study. doi:10.1177/0095399712473985 moynihan, d. p. (2006). managing for results in state governments: evaluating a decade of reform. public administration review, 66(1), 77–89. neal, r. (2008). the importance of the state: political dimensions of a nonprofit network in oaxaca, mexico. nonprofit and voluntary sector quarterly, 37(3), 492–511. pfeffer, j., & carley, k. m. (2012). rapid modeling and analyzing networks extracted from pre-structured news articles. computational and mathematical organization theory, 18(3), 280–299. ponomariov, b., kingsley, g., & boardman, c. (2011). searching for contracting patterns over time: do prime contractor and subcontractor relations follow similar patterns for professional services provision? journal of policy analysis and management, 30(3), 581–597. provan, k. g., fish, a., & sydow, j. (2007). interorganizational networks at the network level: a review of the empirical literature on whole networks. journal of management, 33(3), 479–516. provan, k. g., & lemaire, r. h. (2012). core concepts and key ideas for understanding public sector organizational networks: using research to inform scholarship and practice. public administration review, 72(5), 638–648. raab, j. (2002). where do policy networks come from? journal of public administration research and theory, 12(4), 581–622. rethemeyer, r. k., & hatmaker, d. m. (2008). network management reconsidered: an inquiry into management of network structures in public sector service provision. journal of public administration research and theory, 18(4), 617–647. ruigrok, w., peck, s. i., & keller, h. (2006). board characteristics and involvement in strategic decision making: evidence from swiss companies. journal of management studies, 43(5), 1201–1226. scott, j. (2013). social network analysis. (3rd ed.). los angeles, ca: sage. van atteveldt, w. (2012). semantic network analysis techniques for extracting, representing, and queryingmmedia content. retrieved from https://www.cs.vu.nl/en/images/whvanatteveldt14-10-2008_ tcm75-259647.pdf 56317_14-11.indd 100 05/10/15 5:09 pm qian hu / conducting content analysis of documents 101 vidovich, l., & currie, j. (2012). governance networks: interlocking directorships of corporate and nonprofit boards. nonprofit management and leadership, 22(4), 507–523. villadsen, a. r. (2011). structural embeddedness of political top executives as explanation of policy isomorphism. journal of public administration research and theory, 21(4), 573–599. wasserman, s., & faust, k. (1994). social network analysis: methods and applications. cambridge: cambridge university press. xu, g., & li, l. (eds.). (2013). social media mining and social network analysis: emerging research. hershey, pa: igi global/information science reference. yang, k., & melitski, j. (2007). competing and complementary values in information technology strategic planning: observations from ten states. public performance and management review, 30(3), 426–452. 56317_14-11.indd 101 05/10/15 5:09 pm 102 qian hu / conducting content analysis of documents appendix a list of 39 public administration journals administration & society the american review of public administration australian journal of public administration canadian public administration evaluation review financial accountability and management human relations international journal of public administration international review of administrative sciences journal of accounting and public policy journal of health politics, policy, and law journal of management studies journal of policy analysis &management journal of public administration research &theory journal of public budgeting, accounting and financial management / public budgeting and financial management journal of public policy journal of urban affairs municipal finance journal national tax journal nonprofit management & leadership nonprofit & voluntary sector quarterly organization studies policy sciences policy studies journal policy studies review /review of policy research political psychology political science quarterly public administration & development public administration: an international quarterly public administration quarterly public administration review public budgeting and finance public finance quarterly / public finance review public performance and management review publius review of public personnel administration social science quarterly state & local government review urban affairs review 56317_14-11.indd 102 05/10/15 5:09 pm teaching complexity by means of problem-based learning: potential, practice, and pitfalls complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 32 university of bamberg press teaching complexity by means of problem-based learning: potential, practice, and pitfalls lasse gerritsa, martin wirtzb abstract: this paper examines the use of problem-based learning (pbl) in teaching the complexity sciences to students in public administration, policy, and management. we will discuss the background of the method, as well as demonstrate how it is used in some of our courses. we conclude that, while the method is not applicable in every situation, it does perform very well in bridging the gap between concepts and theories from complexity on the one hand, and more mainstream theories in public administration on the other. it performs equally very well in bridging the gap between theory and practice, as such preparing students in developing a complexity-informed approach to policy issues. keywords: teaching; complexity; problem-based learning; public administration 1. introduction for about two decades, scholars have argued for a natural fit between concepts derived from the realm of the complexity sciences and public administration, public policy and public management, because such concepts adequately capture the nature of policy problems they are facing (oecd, 2009). as mary lee rhodes and elizabeth eppel note in the introduction to this special issue, however, while there is an existing body of literature on teaching public administration (see e.g. the journal ‘teaching public administration’), there is virtually no literature on how complexity theory is or should be taught to students in public administration, public policy and public management. it is self-evident that any teaching approach should help students to learn the content of a discipline, as well as enable them, eventually, to apply their knowledge to real, practical cases (van der steen, martijn, van twist, & frissen, 2017). beyond that, the teaching method in question should also display features that make it particularly suitable for teaching a. complexity thinking to b. future public administrators and managers. classic ex-cathedra teaching is arguably not the best answer to these requirements. first, the form and content of teaching would not appear compatible (van der meer, fransbauke & marks, 2015). if we expect future administrators to engage efficiently in low-hierarchical networks, teachers shouldn’t instil a passive stance in students by doing the actual teaching in a hierarchical manner (e.g. alford & brock, 2014). second, the teaching approach should make use of the fact that students, in particular in post-experience programs, bring considerable preknowledge (hall, 2015). third, related to both of the foregoing points, ex-cathedra teaching with pre-defined solutions to clearly identified problems does not appear adequate to the complex, interdisciplinary challenges public administrators face in daily practice (van der waldt, gerrit, 2014). a lasse gerrits, otto-friedrich university bamberg, germany, e-mail: lasse.gerrits@uni-bamberg.de b martin wirtz, otto-friedrich university bamberg, germany, e-mail: martin.wirtz@uni-bamberg.de http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 33 university of bamberg press fortunately, chalk-and-talk is not the only option for teaching in public administration, even though we can tell from personal experience that this remains a popular teaching method. we also witness many dedicated instructors striving to include more innovative elements in their teaching efforts. this could encompass the inclusion of innovative elements within classic teaching methods, but also more encompassing approaches such as case teaching (andrews, 1950; golich, 2000) that serve to restructure the overall format of teaching. the present paper discusses problem-based learning (pbl) (barrows & tamblyn, 1980; schmidt, 1983) as a superior alternative to chalk-and-talk in teaching complexity. to our knowledge, the diffusion of pbl in the social sciences has so far been limited (craig & hale, 2008, p. 165), though it should be noted that some of its elements are also present in other approaches. maastricht university (the netherlands) started using pbl in its medical education in 1976 but gradually extended its usage to include the social sciences and the european studies curriculum amongst others (maastricht university, 2017; maurer & neuhold, 2014). pbl was also adopted by the erasmus university rotterdam (the netherlands), where it was first used in the psychology curriculum, and later on established as the main teaching doctrine for all programs in all social sciences including public administration and policy, and public management (erasmus university rotterdam, 2016). in addition, the university of huddersfield (uk) uses pbl in political science (craig & hale, 2008; hale, 2006). the limited number of social science applications known to us notwithstanding, the experience of the european studies curriculum at maastricht university since 2002 and our own experience at two other universities indicates that pbl is a powerful educational tool in our field. pbl shows considerable parallels to, and sometimes is even used synonymously with, the case teaching method (hale, 2006). nevertheless, we focus on pbl here for four reasons. first, pbl can be used for most classes of problems, e.g. practical, theoretical, and some methodological ones. second, it offers a distinct, clear and effective way of structuring a course and the sessions within the course. third, it is the method we personally have experience with and have dwelled on also in more theoretical terms. fourth and most important, we are not intending or claiming to revolutionise teaching in public administration per se, but first and foremost want to explore a teaching method particularly suitable for teaching complexity within the field. this paper argues that pbl can be instrumental in teaching complexity to students in public administration, public policy and public management. in doing so, the remainder of this paper proceeds as follows. first, we will present the potential of pbl for teaching complexity by pointing out where and how pbl and complexity match. second, we will present how pbl is applied in practice using the so-called seven-step-approach (maurer & neuhold, 2014, pp. 204– 207; van til & heijden, francy van der, 2000). we will provide examples from our own experience to demonstrate the approach. third, we discuss some pitfalls when using pbl, again based on our experiences. we will conclude the paper by identifying pbl as a one promising way of addressing the gap between the awareness of complexity in public administration on the one hand and integrating the respective theories into teaching practice on the other hand. for only if teaching manages to bridge this gap effectively can policy and administration become complexityinformed. 2. potential we will first highlight the potential of pbl for teaching complexity theory, largely by elaborating on a number of parallels between pbl on the one hand and complexity theory and/or public administration on the other hand. naturally, complexity is multidimensional and appears in many shapes to public administrators. while it is useful to dissect the various dimensions (see gerrits, http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 34 university of bamberg press 2012), we acknowledge that complexity’s primary face in the practice of public administration and management comes in the shape of a concrete but wicked problem that begs to be solved. solving the problems requires one to understand the nature of the problem as expressed in terms of the complexity sciences, i.e. it being systemic, dynamical, emergent, etc. ideally, knowledge acquired at the academy helps in understanding the problem but this is not as straightforward as it seems: there is a considerable gap between academic knowledge and professional practice (van der steen, martijn et al., 2017), and students are often trained to apply abstract and rarefied solutions that – naturally – don’t work in a specific case they are facing. a second gap is between the standard concepts and theories in public administration and the ones deployed in the complexity sciences. pbl provides an approach with which these gaps and shortcomings can be remedied. first of all, pbl was specifically developed in answer to the gap between state-of-the-art knowledge taught at universities and professional practice. h.g. schmidt recognized such problems both within and beyond his own discipline of medical education leading him to the puzzling conclusion that “people can possess knowledge which they seem unable to apply” (schmidt, 1983, p. 11). indeed, this is particularly acute in the medical sector. the theory-vspractice gap ties in specifically with the complexity of the matter at hand: after all, the human body as the subject of medical studies constitutes a complex system (sturmberg & topolski, 2014), and pbl was “established to come to terms with complex problems in the domain of medical studies” (ibid. 2014, p. 199). in his 1983 article, schmidt discussed how problem-based learning (pbl) as developed by barrows and tamblyn and evolving “out of the needs of professional practice” (ibid. 1983, p. 12) might enable students to turn the knowledge acquired into successful practice. the value of pbl in medical education as established mainly theoretically by schmidt in 1983 has been questioned within the empirical literature (kirschner, sweller, & clark, 2006), but more recent meta-analyses arrive at an overall positive result (koh, khoo, wong, & koh, 2008; schmidt, van der molen, te winkel, & wijnen, 2009). as we have already noted earlier, pbl can generally be applied in the social sciences, but its success in medical science, as a field characterised by complexity, might imply that it can be transferred even more smoothly to teaching complexity in public administration. in order to substantiate this point, we can further elaborate on the core features of making pbl successful, namely the activation of prior knowledge, elaboration of knowledge and encoding specificity (ibid. 1983, p. 12; see also albanese, & mitchell1993; bridges, 1992; gijselaers, 1996; schmidt et al., 2009). as we summarise point by point, all of those conditions are generally key to effective learning. they are particularly valuable, however, when applying pbl to teaching complexity in public administration. the activation of prior knowledge, for instance, generally facilitates the memorization and storage of new information that builds on that existing knowledge (bridges, 1992; gijselaers, 1996; maurer & neuhold, 2014). pbl achieves this by involving all students in a brainstorm on the (as yet undefined) problem situation. the resulting collective formulation of learning objectives then further assures that students are neither bored by having to repeat all too selfevident explanatory components, nor overburdened with questions requiring a degree of knowledge that the students do not yet possess. for teaching complexity, activation of preknowledge is important no matter what the composition of the student group: students with professional experience have a rich storage of practical experience, while students following a classic academic curriculum will rarely enter a specific course on complexity without theoretical insights from other classes in the curriculum. in fact, these links have to be established thoroughly in order to ensure that complexity is not understood as yet another intellectual container unrelated to chunks of knowledge already acquired. http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 35 university of bamberg press pbl then forces students to actively elaborate on their knowledge in group discussions. through these discussion, students often only then start to truly understand what they have just read (gijselaers, 1996; maurer & neuhold, 2014; schmidt, 1983). involving students to this extent – both in the activation and elaboration processes – has been considered a paradigm shift (cf. maurer & neuhold, 2014, p. 203): whereas chalk-and-talk follows a positivist epistemology, pbl treats learning as “an active process and social phenomenon that is highly dependent on the context in which learning is taking place” (maurer & neuhold, 2014, p. 203). this echoes morçöl’s stance that knowledge generated in the complexity sciences must be understood as contextual. at the same time, complexity science does not deny the possibility of any objective truth (as postmodernists and post-structuralists would). a similar attitude is reflected in the practice of pbl: while “[t]he tutor’s [i.e. instructor’s] main task is to promote both the learning process of the students and their mutual co-operation” (van til & heijden, francy van der, 2000, p. 16) and to hold himself back, it is still part of his or her role to control the discussion, for example by asking questions “if ideas are incorrect” (ibid.). this suggests that pbl, like complexity science, follows a post-positivist, but not (necessarily) a post-modernist philosophy. in less philosophical terms, pbl involves students by accepting their inputs as relevant and even handing over the organisation of the discussion to a student in the role of a ‘chair’ (maurer & neuhold, 2014; savin-baden & major, 2004; van til & heijden, francy van der, 2000), which by definition is much less hierarchical than ex-cathedra teaching. addressing complex problems in a non-hierarchical manner is also much more in line with the recommendations for governance that follow from a complexity perspective on public policy making (klijn & koppenjan, 2014; cf.klijn & koppenjan, 2016). it is the last of the three conditions for effective learning – encoding specificity – where applying pbl in the teaching of complexity seems to carry the highest potential. encoding specificity concerns the resemblance between the situation in which something is learned and the situation in which it is applied. the higher the resemblance, the better the performance of the students (schmidt, 1983). leaving the definition of the problem to students is not just less hierarchical than chalk-and-talk, it is also reasonable to assume that students have to deal with undefined (wicked) problems in their later professional life. within the medical context where pbl was developed, a problem definition might follow from a patient describing his/her vague symptoms (e.g. schmidt, 1983, p. 13). similarly, much of public administration is already concerned with dissecting and then solving what is considered a (societal) problem – be it an untameable budget deficit of a public agency, public roads in disarray after a botched tender, or the introduction of new curricula at public schools not leading to expected results. once (back) in professional contexts, encoding specificity will ease the application of the knowledge acquired (gijselaers, 1996; maurer & neuhold, 2014; schmidt, 1983). note that this doesn’t imply that all problems raised need to be real, empirical cases (as applies to the case teaching method, cf. golich, 2000, p. 12) or purely practical ones. there is also much utility in putting a theoretical problem upfront, provided the instructor can demonstrate how such a question could have relevance in the real world. also the ways in which the students are then asked to address the problem resembles administrative practice: rather than being reliably presented with a clear-cut solution by some authority further up the hierarchy, students first have to muddle their way through the information available. in doing so they are – as frequently in real-life – given the opportunity to discuss their approach to a solution in the team, i.e. the pbl group. at the same time, discussing the problem in a team will necessarily involve a degree of disagreement about the ‘right’ solution. this resembles the kind of substantive complexity (klijn & koppenjan, 2014, 2016) public administrators will have to handle in daily practice as well: a rather non-hierarchical network within which knowledge needs to be organised is equally similar to the situation in a pbl class. http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 36 university of bamberg press finally, by taking on the role of a ‘chair’ leading a discussion to a productive end, students probe the role of a network manager. in short, teaching complexity by means of pbl reveals both the content and the process of dealing with complex problems. 3. practice we will now present the concrete steps of the pbl process in more detail (for an overview, see table 2). throughout the section, we draw from our own experiences in using pbl to teach complexity in various courses within academic programs in public administration, policy, and politics, in germany and the netherlands, serving more than 250 course participants in total. the full list of courses is given in table 1 below. we will use these experiences to substantiate our argument that pbl is suitable for teaching complexity in public administration, as well as provide some guidelines for those interested in applying the method themselves. table 1: overview of courses using pbl to teach complexity as used for this paper. note that ‘confronting the urban crisis’ and socio-technological evolution’ focus on a specific policy domain (urban planning and technology, respectively) but still spend considerable attention on complexity theory and its application to the persistent questions in those domains. title period level program university dynamics in complex systems 2010-2013 ma public administration erasmus university rotterdam (ned) theories of complex systems 2014-current ma political science otto-friedrich university bamberg (ger) governance: theory and praxis 2014 ba political science otto-friedrich university bamberg (ger) why things are complex 2014-current ba political science otto-friedrich university bamberg (ger) confronting the urban crisis 2014-current ma political science otto-friedrich university bamberg (ger) social-technological evolution 2015-current ma political science otto-friedrich university bamberg (ger) preparation course preparations focus on designing the ‘problem’ students will have to address (gijselaers, 1996, p. 20; hale, 2006, p. 90; maurer & neuhold, 2014, p. 208; schmidt, 1983, p. 15; sockalingam, 2010). the problem should be presented in the form of a “neutral description of event or a set of phenomena that are in need of explanation in terms of the underlying process, principles or mechanisms” (schmidt, 1983, p. 15). the problem needs to be formulated as concretely as possible and needs to fit the degree of complexity students can be expected to handle (cf. willems, 1981). in practice, there is a wide range of ways for doing it: from problems reported by people working in public administration (craig & hale, 2008; hale, 2006) to entirely selfwritten descriptions of academic puzzles, such as a text juxtaposing various definitions of ‘policy’ http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 37 university of bamberg press as a starting point for a course (maurer & neuhold, 2014, p. 211). this raises the question which approach is preferable. there are pros and cons attached to each approach. the former takes the real-life linkage of pbl most seriously. the latter approach seems much more pragmatic not only in terms of workload (craig & hale, 2008, p. 168) but also in terms of flexibility regarding the level of studies, institutional contexts, study programme structure etc. fortunately, it is not necessary to make a clear-cut decision. first, between those two extremes, there is a lot of space in-between, constituting for instance in the careful selection of relevant news coverage. newspaper articles often combine a (more or less neutral) description of a real-life situation, while already skipping minor details. second, there is no reason not to vary the type of problem definitions between various sessions within the same course. in fact, variation of this sort is recommended to avoid monotony (maurer & neuhold, 2014, p. 211). in our own practice, we found it helpful to largely rely on self-written problems (also referred to as ‘assignment texts’, cf. maurer, neuhold 2014), so as to maintain coherence between the problem and the materials to be covered for that particular issue. in the following example from the ba course ‘why things are complex’, however, we simply drew on a passage from a foreword: the destruction brought down and across the japanese coast in march 2011 has returned to us our fear and our uncertainty. we had assumed mastery where misunderstanding and negligence lay. above all, we have seen the world, and ourselves inside it, brought to cold reflections: how have our systems failed? how have we failed to prevent? how have we failed to imagine? the third charge, our most severe and accusatory, encompasses the whole of our missteps and malpractice towards this disaster and all before it: we may have failed to imagine. once, requiring only a crude understanding of possibilities and outcomes, wielding technologies of risk was a seemingly simple affair. their behaviour we assumed to be known, controlled and local. we managed the benefits and risks of these instruments with predictive laws, and thus, predictive remediation when we experience failure. the uncertainties of our technology and its scientific basis were to be uncovered through progress, never beyond the limits of investigation and never included as a feature of science. [source: y. fujigaki (2015). lessons from fukushima. japanese case studies on science, technology and society (foreword by r. chhem). heidelberg (et al.): springer. e-book available at the university library.] the introductory text then serves as a concrete entry point for reading and discussing academic work. understanding the literature – here, perrow’s normal accident theory (perrow, 1999, first published in 1984) – is made easier for students by linking it to a real-world problem they can directly or indirectly relate to. the example here, the fukushima nuclear disaster of march 2011, is something which most students will personally recall. such a connection is always necessary to get to grips with the more abstract literature on complexity. it will also help them connect it to previous topics and other literature they may have read, e.g. on the nature complex systems. a problem such as described above will also be used to structure the ensuing group discussion. new roles of instructors and students while preparation of the session in terms of problem design and selecting learning materials is the task of the instructor, he/she should focus on facilitating the learning process during what is called the preand post-discussion. instead of the instructor, a student – appointed as chair for a given session – should organise the discussions (for more detailed discussion and/or practical guidance on the roles of instructors and students in pbl, see savin-baden & major, 2004 and van til & heijden, francy van der, 2000, respectively). the learning process is now handed to the students. this corresponds to the non-hierarchical nature of pbl mentioned earlier. in addition to http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 38 university of bamberg press the advantages of pbl in terms of effective learning that we have noted above, we have found in our own teaching that this shift of responsibility is highly appreciated by many – albeit not all – students: “this feels more like a think-tank than like a normal class”, as one student remarked. we concur and see this as a situation that approaches daily practice of administrators. in addition to the aforementioned encoding specificity – indeed think tanks might be where complexityminded students of public administration might end up – students realise that their inputs are taken seriously and retrieve additional motivation from it. this change of role doesn’t mean that the instructor can remain passive. indeed, the instructor should ask provocative questions, help in terms of organisational issues and, quite importantly, must pay attention to group dynamics and keep the group from “going off the track”, as maurer and neuhold (2014, p. 208) put it. this may involve the correction of factually wrong statements as well as the thematic detours. the seven-step pbl cycle we will now focus on the actual pbl cycle (see table 2). before even hoping to eventually ‘solve’ the problem within the classroom, the description of the problem as such must be fully understood by all students (step 1). amongst others, it must be assured that concepts used in the description provided by the instructor are clarified. this also gives the instructor the opportunity to highlight certain concepts that students may overlook. next, the students have to formulate a concise problem statement that becomes the title for the session (step 2). for example, we discussed the intellectual history of system’s theories in the course ‘theories of complex systems’. a session about the work of e.g. forrester or parsons could focus on the following problem statement: ‘why did structural functionalism in systems theory fall from grace?’ naturally, the statement differs from semester to semester as different students will come up with different suggestions, depending on their reading of the introductory text, pre-knowledge etc. having done this, students need to come up with their first possible explanations of the problem, or at least on potentially relevant explanatory elements. this is done using brainstorming techniques involving e.g. a whiteboard (step 3). note that this draws primarily from existing knowledge, as well as very tentative – one might even say common sense – ideas from the students. in the example mentioned above, students would often feel that functionalism in systems theory conjured up a rather mechanistic understanding of any complex system – as if such systems could be governed with the push of the button (see klijn & snellen, 2009; for a discussion) – and so they would focus on the jargon or normative statements in literature. inasmuch as possible, those ideas should then be categorised, structured or set in a relation to each other (step 4), for example by assigning them to the category of conceptual issues or issues with real-world applications. it can be expected that students will struggle with this because they have yet to delve deeper in the learning materials. however, this struggle has utility as it will make them acutely aware of their own knowledge gaps. the fifth step, then, is the identification of smaller but more manageable research questions or learning goals. students can’t be expected to formulate these goals according to the standards for academic courses and the quality will vary accordingly. an example comes from the course ‘confronting the urban crisis’. here, the topic of complex systems, and functionalism in such systems, as mentioned above resurfaces in the session on modernism in urban planning. the chair of the session let the students read a text by lecorbusier. they then formulated the research questions / learning goals as follows: http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 39 university of bamberg press – is planning alone enough to change the behaviour of people? – what is the empirical base for planning in complex cities? – how can modernist design be implemented properly? – what explains the popularity of modernism? not all possible questions might be equally relevant and it might not be necessary that everyone addresses all of the learning objectives. the group should discuss, therefore, who should address which questions, and, depending on the exact course design, which learning resources are required. in some cases, it may be decided that everyone should address all questions. the instructor could enforce this if there is a risk that students will distribute the work among them with the risk of ignoring some important elements in the learning materials. http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 40 university of bamberg press table 2: overview of the seven-step approach and its underlying logic (source: maurer, neuhold 2014 pp. 205-206), added by complexity-relevant aspects step what to do? what to do in detail? why? value added when teaching complexity 1 clarification of terms and concepts ask for explanations of words or concepts that are not understood if illustration: discuss what picture shows any sentences/passages that are difficult to understand? provide common starting point, i.e. every group member should understand the assignment text as it stands linking abstract notions of complexity theory and thinking to real-life experiences of students facing an unstructured problem 2 formulation of problem statement provide title for the session or formulate wider research question, i.e. what is it about students dive into topic and grasp the underlying problem of the assignment by discussing in the group, students establish a common ground of the problem – they not only name it but discuss it and also examine its wider relevance 3 brainstorming everything is allowed: collection of ideas, potential explanation in regard of problem statement, etc. to establish and contrast: what does the group already know – what does the group want to find out students spontaneously name aspects that they consider as interesting and relevant activation of prior knowledge and real-world experiences – students should link the problem statement to existing knowledge non-linear, free-floating process of problem solving 4 categorising and structuring of brainstorming keywords from brainstorming are put into similar categories (according to question type; why, how, what consequences, etc.) structuring first creative collection of ideas to find patterns and facilitate the formulation of few learning objectives (partial) structuring of complexity 5 formulation of learning objectives use categories of structured brainstorming to formulate single questions or research tasks (e.g. look for x) provide clear focus in reading the literature by having smaller research questions guiding the learning process clear and guided assessment of what is needed to answer the posed question no top-down assignment of tasks 6 self-study students read literature, look for additional sources, prepare answers to the formulated learning objectives students as self-directed and responsible learner imperfect match between problem (including learning goals) and ‘solutions’ (information in the literature) 7 post-discussion and reflection on learning process students report back on how they answered the learning objectives, compare results but also exchange arguments self-assessment of students in learning process and peer assessment, especially in roles of chair and discussant by formulating acquired knowledge in own words and by exchanging arguments with peers, deeper understanding is facilitated in contrast to pure memorising students become aware of potential misinterpretations of (empirical) material being confronted with reports from peers by becoming aware of what works well and what could be improved, first step to improve learning process not all experiences students have to make themselves, but they can learn tremendously by observing and providing feedback to each other confronting and managing substantive complexity resulting from differences in the readings of the literature http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 41 university of bamberg press the fifth step completes the so-called ‘pre-discussion’ (cf. maurer & neuhold, 2014). it is followed by self-study of learning materials in preparation of the next classroom session. this is step 6. typically, the materials are provided by the instructor. alternatively, students could be instructed to identify materials themselves. in the course ‘theories of complex systems’ we asked students to identify and summarize the key elements from certain authors in complexity, such as holland (e.g. 1995; 2006). this requires that students not only read a text from those authors but also make an attempt to separate the main themes from the minor ones. this is a daunting task for anyone but the students seemed to cope quite well. they would also use secondary materials such as wikipedia or academic reviews to get a better understanding of the original texts. authors such as holland have not only written a lot themselves, others have written about them or used their models in other research projects. students can get a better understanding of the primary materials through such secondary sources. authors such as gerrits (2012) and morçöl (2012) have given useful overviews of the main concepts and theories in the complexity sciences for public administration and students can use them as reference. finally, in the last step, namely the so-called ‘post-discussion (cf. maurer & neuhold, 2014), students report in the classroom setting how they have addressed the learning objectives, compare their own results and discuss potential differences (step 7). this is important in science in general but especially important in the complexity sciences where theories can often appear to contradict, overlap or explain the same mechanism using different terminology (see e.g. morçöl, 2012; on this issue within public administration and public policy). 4. pitfalls there are some conditions that are hard to influence for individual instructors, while at the same time they do play a role for the effective application of pbl – as an example, think of the time students can dedicate to a course in the face of obligations for other courses. moreover, our experience has also provided us with a number of lessons learned concerning some pitfalls when using pbl. we would like to briefly discuss these general conditions and lessons learned here. there seems to be little doubt that pbl cannot be used effectively in larger groups (>20 students) because such a group size doesn’t allow for much direct and interactive discussion among all students. even below this number, some students might try to ‘hide’ in the group, thereby avoiding activating their pre-knowledge as well as elaborating on what they have learned in self-study. generally, smaller groups seem to entail an increased pressure to participate, as students realise quickly that everything depends on them. it should also be noted that we have had good experiences in groups even as small as three students. nevertheless, it seems that groups of about ten students are ideal, since it increases the variety of insights and opinions without fostering free riding too much. the institutional setting matters in the performance of pbl, too. for example, the incentive structure in terms of grading may play a role. ideally, all students participate fully and at high levels without any external incentive. in reality, there will always be a need for an incentive as some students simply want or even need to have their performance graded, or the institutional rules may demand that students can earn grades or credits by deploying certain activities. while some universities have rules that oblige students to participate in sessions and allow for rewarding good performance of both ‘chairs’ and regular participants in the discussion, others do not even force students to even be present in regular sessions. nevertheless, there is often some room for manoeuvre within a given institutional structure that can make it more conducive to pbl. for example, we decided to hand out grades for the student chairing a session, leading the discussion and ensuring its quality in terms of form (e.g. involving as many other students as http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 42 university of bamberg press possible) and content (e.g. covering the learning goals, being led by them rather than the contents of the texts). students also need to get acquainted with pbl, which is equally challenging in both form and content. this is independent from the composition of student groups. students – at least the ones we encounter in our classrooms – are primed for chalk-and-talk, either through high school experiences or because of lectures followed elsewhere at the university. even post-experience students can be difficult to guide at first, given that their work experience has primed them in certain ways and has created some mental path-dependency. here, the instructor must take care to keep the discussion open, e.g. by asking critical questions. generally, it helps students if the instructor does not only formally present some ‘rules of the game’ for pbl to which students can refer, but if he or she also leads by example in taking over the role of a chair in the first session. in addition, whatever the external conditions, instructors should seek to avoid a number of pitfalls. the most basic pitfall is to assume that pbl sessions would be less demanding for instructors because students are supposed to take control of the sessions. holding yourself back as an instructor doesn’t mean leaning back. balancing between holding back and trying to keep control is one of the main challenges (hale, 2006, p. 92; maurer & neuhold, 2014). it may be tempting, for example, to intervene in the learning process by leading the pre-discussion as an instructor and to shape learning goals in order to make them match the assigned literature. here, as well as in the post-discussion, it is necessary to be careful not give students the impression that their involvement is just window-dressing. in one of our own courses, providing pre-defined learning goals after having students read the ‘problem’ led the students to conclude that the text outlining the problem was just some way of linking two sessions. this is not undesirable per se, but a far cry from the activation and usage of prior knowledge. last but not least: our experiences with fitting pbl into a given institutional setting have taught us that the seven-step-design of pbl should not be changed – it is designed to be used in this way and we believe that it is the best way to structure the sessions. for example, the course ‘dynamics of complex systems’ originally used elements from pbl but not in the exact order as prescribed in the seven-step approach. while this could still make a viable course, it doesn’t benefit from the activation of the mechanisms pbl is aimed at. after all, it is not just a way of organising a session, but a way of making sure that the benefits of pbl in terms of activation of pre-knowledge, encoding specificity and elaboration are actually reaped. similarly, in the course ‘why things are complex’ we adopted most but not all elements from pbl in the expectation that this would ease ba students into this type of learning. instead, some students got confused about the actual purpose of the method. in the following semester, we added even more elements of pbl, but left the instructor in control of the pre-discussion. this, in turn, made students feel that they were supposed to guess the learning goals the instructor wished to hear. in most cases, it is possible to integrate pbl within one’s teaching environment. when deciding in favour of pbl, our experience shows that it is then better to go for it wholeheartedly, this means, first, to adopt as much of the framework and mind-set presented above, and, of course, not to give up after the first round, as group dynamics will change from seminar to seminar. 5. conclusion our experience with pbl in teaching complexity has been very positive so far. student surveys, which have been collected after each course, show that a majority of the students value the format. they report that they have experienced a steep but manageable learning curve, and that they can relate the new knowledge to knowledge already obtained. in addition, they report that they see the relevance of the understanding knowledge for solving complex real-world issues. asked in the http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 43 university of bamberg press qualitative section of the evaluation form what should be kept the same about the course, “pbl” is regularly named. most importantly, students agree or even fully agree to the statement that “teaching methods employed are well suited to the subject matter”. naturally, it is very difficult to measure the long-term impact of using pbl in these courses. whether it helps students to later apply complexity thinking in their work after graduation is something that needs to be researched. in any case, our practical experience so far does not contradict the theoretical arguments made in the first sections of this paper. hence there is reason to believe that pbl helps in bridging the gaps identified in the introduction to this paper, and that it helps students to comprehend theories and concepts from the complexity sciences, and to relate it to more mainstream concepts and theories from public administration. teaching complexity theory to students in public administration, policy and management is highly relevant, but there is little guidance on how to do it. without excluding all other innovative teaching approaches, we have suggested problem-based learning (pbl) as deserving consideration. we discussed the origin as well as applications of pbl and presented the main steps of the approach using examples from our own course. while we will not claim that pbl is always the best method in any setting, we do believe that it is a powerful way of overcoming the gaps between the complexity sciences and mainstream approaches in public administration, and between theory and practice. based on the match we perceive in terms of the mind-sets behind pbl and the complexity sciences, respectively, as well as the feedback we receive from using pbl in practice, we hold that that it is one promising method in teaching of complexity in public administration. references albanese, m., & mitchell, s. (1993). problem-based learning: a review of literature on its outcomes and implementation issues. academic medicine, 68(1), 52–81. alford, j., & brock, j. (2014). interactive education in public administration (1). teaching public administration, 32(2), 144–157. doi:10.1177/0144739413515491 andrews, k. r. (1950). case method of teaching human relations and administration. cambridge: hup. barrows, h. s., & tamblyn, r. m. (1980). problem-based learning: an approach to medical education. springer series on medical education: v. 1. new york: springer pub. co. bridges, e. (1992). problem based learning for administrators. retrieved from http://www.eric.ed.gov/pdfs/ed347617.pdf craig, j., & hale, s. (2008). implementing problem-based learning in politics. european political science, 7(2), 165–174. doi:10.1057/eps.2008.6 erasmus university rotterdam. (2016). problem based learning system. retrieved from https://www.eur.nl/fsw/english/information_for/current_students/services_and_facilities /problem_based_learning_system/ gerrits, l. (2012). punching clouds. litchfield park, az: emergent publications. gijselaers, w. h. (1996). connecting problem-based practices with educational theory. in l. wilkerson & w. h. gijselaers (eds.), bringing problem-based learning to higher http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 44 university of bamberg press education: theory and practice. san francisco: jossey-bass publishers (pp. 13–22). san francisco: jossey-bass publishers. golich, v. l. (2000). the abcs of case teaching. international studies perspectives, 1(1), 11–29. doi:10.1111/1528-3577.00002 hale, s. (2006). politics and the real world: a case study in developing case-based learning. european political science, 5(1), 84–96. doi:10.1057/palgrave.eps.2210060 hall, m. (2015). knowledge transfer and teaching public administration. teaching public administration, 33(3), 241–254. doi:10.1177/0144739415593338 kirschner, p. a., sweller, j., & clark, r. e. (2006). why minimal guidance during instruction does not work: an analysis of the failure of constructivist, discovery, problem-based, experiential, and inquiry-based teaching. educational psychologist, 41(2), 75–86. doi:10.1207/s15326985ep4102_1 klijn, e.-h., & koppenjan, j. f. m. (2014). complexity in governance network theory. complexity, governance & networks, 1(1), 61–70. retrieved from http://ubp.unibamberg.de/ojs/index.php/cgn/article/view/20 klijn, e.-h., & koppenjan, j. f. m. (2016). governance networks in the public sector. london, new york: routledge. koh, g. c.-h., khoo, h. e., wong, m. l., & koh, d. (2008). the effects of problem-based learning during medical school on physician competency: a systematic review. cmaj : canadian medical association journal = journal de l'association medicale canadienne, 178(1), 34– 41. doi:10.1503/cmaj.070565 maastricht university. (2017). problem-based learning education maastricht university. retrieved from https://www.maastrichtuniversity.nl/education/why-um/problem-basedlearning maurer, h., & neuhold, c. (2014). problem-based learning in european studies. in s. baroncelli, r. farneti, i. horga, & s. vanhoonacker (eds.), teaching and learning the european union (pp. 199–215). dordrecht: springer netherlands. oecd. (2009). applications of complexity science for public policy: new tools for finding unanticipated consequences and unrealized opportunities. retrieved from http://www.oecd.org/science/sci-tech/43891980.pdf perrow, c. (1999). living with high-risk technologies (new ed.). new jersey: princeton university press; [distributor] john wiley and sons ltd. savin-baden, m., & major, c. h. (2004). foundations of problem-based learning. berkshire: open university press. schmidt, h. g. (1983). problem-based learning: rationale and description. medical education, 17(1), 11–16. doi:10.1111/j.1365-2923.1983.tb01086.x schmidt, h. g., van der molen, h.t., te winkel, w. w. r., & wijnen, w. h. f. w. (2009). constructivist, problem-based learning does work: a meta-analysis of curricular comparisons involving a single medical school. educational psychologist, 44(4), 227–249. sockalingam, n. (2010). characteristics of problems in problem-based learning. rotterdam: eur publishing. http://dx.doi.org/10.20377/cgn-46 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 32-45 doi: http://dx.doi.org/10.20377/cgn-46 45 university of bamberg press sturmberg, j., & topolski, s. (2014). for every complex problem, there is an answer that is clear, simple and wrong: and other aphorisms about medical statistical fallacies. journal of evaluation in clinical practice, 20(6), 1017–1025. doi:10.1111/jep.12156 van der meer, frans-bauke, & marks, p. (2015). research and teaching pa. teaching public administration, 34(1), 109–116. doi:10.1177/0144739415613655 van der steen, martijn, van twist, m., & frissen, p. (2017). learning from experience. teaching public administration, 35(1), 105–125. doi:10.1177/0144739416670701 van der waldt, gerrit. (2014). public administration teaching and interdisciplinarity. teaching public administration, 32(2), 169–193. doi:10.1177/0144739414523285 van til, c., & heijden, francy van der. (2000). pbl study skills: an overview (2nd print). maastricht: department of educational development and research (o&o), universiteit maastricht. willems, j. (1981). problem-based (group) teaching: a cognitive science approach to using available knowledge. instructional science, 10(1), 5–21. doi:10.1007/bf00124562 http://dx.doi.org/10.20377/cgn-46 how to educate navigators in a complex world: making a case in higher professional education in the netherlands complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 19 university of bamberg press how to educate navigators in a complex world: making a case in higher professional education in the netherlands paul van der cingel abstract: this article focuses on the increased relevance and urgency to pay explicit attention to complexity-informed perspectives on real-world issues in dutch higher professional education. moreover, it describes lessons learned from experiments with socalled embedded complexity workshops. these workshops, comprised of activities like network visualization, yield promising results and can serve as a starting point for further development. keywords: professional education; network visualization; connectivity; systemic awareness; ambiguity; dynamism introduction governments and other future employers of today’s students, experience an increasing relevance and urgency to call on educational institutes to prepare students to deal with complex issues. in this article, the call to action is addressed by discussing the utility and use of so-called plug-in workshops in dutch higher professional education. the workshops aimed at offering students alternative and complexity-informed perspectives on complex problems which they were already studying. in this article, we call this type of education embedded complexity workshops, ecw’s. in the first section, we outline our working definition of the concept of complexity. the second section provides some background information about tertiary, or higher, education in the netherlands. in the third section, we will argue why it is necessary to develop ecw’s. section four covers the design of learning objectives that pave the way for developing and testing the plug-in workshops. the workshops themselves are described in section five. section six draws conclusions and poses some questions for future experiments. a working definition of complexity due to the multidisciplinary nature of complexity science, scientists, professionals and policymakers still lack a consensus definition of the concept of complexity (heydari & wade, 2014). in this article, our working definition of complexity contains three aspects: connectivity, ambiguity and dynamism. paul van der cingel, windesheim university of applied sciences, the netherlands, e-mail: p.van.der.cingel@windesheim.nl http://dx.doi.org/10.20377/cgn-63 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 20 university of bamberg press the first aspect is connectivity, by which we mean that in a complex issue, various agents and factors are in some way related, linked, and interdependent. taking a systemic perspective, one could say that this aspect of complexity is all about becoming aware that getting to know the whole requires knowledge of the system’s components as well as their interrelationships. the second aspect is ambiguity, by which we mean that relations between agents and factors can be multi-dimensional. for example, the relation between two agents can be financial, legal, organizational or social, effectively introducing potential tensions in relationships. the third aspect is dynamism, by which we mean that in a complex system of interacting and adapting agents, change is certain to occur. for example, two agents may switch from a competitive to a collaborative relationship, adpating to the context of a changing situation. however, this does not mean that the system as a whole is in total flux. rather, it is the dynamics of alternating between stability and change that constitutes complexity (gerrits, 2012). we label real-world systems, networks, problems, challenges and developments as “complex” if they match these three aspects. the relationship between two human beings would fit the bill, providing us with the example of complexity that is arguably the easiest to grasp. in this article, we will encounter real-world issues that show considerably more complexity, because of a larger number of interrelated agents and factors. the context of higher professional education in the netherlands tertiary, or higher, education in the netherlands is characterized by two distinct paths. the first path offers scientific research-oriented education, embedded in the common structure of bachelor and master programs. currently, about 260,000 students take this path at some 20 universities. the second path offers higher professional education, embedded in a four-year bachelor program. currently, over 440,000 students follow this track at 37 universities of applied sciences. unlike other contributions to this special issue, this article focuses on education for students who still lack working experience in the relevant professional context. http://dx.doi.org/10.20377/cgn-63 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 21 university of bamberg press table 1 hightlights some of the differences: table 1 selected differences between professional and research-oriented education in the netherlands adapted from: http://www.rug.nl/feb/education/premaster/hbo professional education at universities of applied sciences research-oriented education at universities higher vocational education emphasizes training for a specific profession. the training is directed towards the acquisition of competences. the emphasis is on learning to look analytically and critically at the way a certain field can be approached. you learn to present convincing oral and written arguments and to draw conclusions from them. professions usually clear in advance: after you graduate, you will usually end up in a white-collar job. professions less clear in advance: personnel ads often ask for an “academic level of thinking” rather than for specific knowledge. more supervision: the contacts with students are often more intensive. there are usually more contact hours, including compulsory ones, such as lectures and work groups. less supervision: own initiative, self-discipline, independence. as can be glanced from table 1, setting up an academic course, e.g. “introduction to complexity sciences”, will not work in professional education. here, ecw’s should be developed bearing in mind that one single priority: the purposeful application of scientific insights within contexts of a specific professional field. moreover, setting up a course which requires students to master literature (mostly in english) through self-study, will not work either. here, ecw’s should be developed in a framework of lectures and supervised work groups, in dutch (or any other local language, depending on where this is taught). windesheim university of applied sciences has been ranked as one of the dutch top institutes since a number of years (windesheim, september 27, 2017). with a student population of over 25,000 and a staff of 2,000, it hosts 50 bachelor programs. these bachelor programs are preparing students for a great variety of professions. table 2 lists examples of typical four-year bachelor programs from various domains at windesheim university. http://dx.doi.org/10.20377/cgn-63 http://www.rug.nl/feb/education/premaster/hbo complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 22 university of bamberg press table 2 examples of bachelor programs at windesheim university domain examples of bachelor programs and professional fields technology engineering, industrial design, it media journalism health applied gerontology, speech therapy business financial services management, business & information management social work socio-pedagogical care, child care management education teaching in primary and secondary education, subjects ranging from sports to theology multidisciplinary honours programs global project and change management windesheim university of applied sciences provided us with resources to develop and test new ways to implement ecw’s. why? making a case for ecw’s in dutch higher professional education in this section, we will argue why there is a need for new ecw’s in dutch higher professional education. in order to make a credible case, we will examine two questions: [1] how can we pinpoint the complexity that students in higher professional education will encounter? [2] what are the deficiences in the regular curricula in teaching students how to deal with this complexity? the question of pinpointing complexity just a few years ago, complexity-informed perspectives were almost exclusively offered in academic, research-oriented curricula. within the context of professional education, complexity was no more than a buzz word, fashionably placed in forewords and introductions, but rarely taken seriously and recognised as having implications. for example, in 2014, an advisory committee wrote a report on the future of business programs in dutch higher professional education. the report was commissioned by the netherlands association of universities of applied sciences (verkenningscommissie hoger economisch onderwijs (2014). the word “complex” was mentioned 20 times in the report, a stunning 9 of which stood in the foreword. a quote from the foreword read (translated from dutch) “first and foremost, reflection on higher economic professional education means: teaching students that they have to accept that complexity and uncertainty are fundamental to our society…” (verkenningscommissie hoger economisch onderwijs 2014, p.3) http://dx.doi.org/10.20377/cgn-63 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 23 university of bamberg press the report did not live up to the expectations of the foreword’s words. in the rest of the report, no advice was given as to how exactly students had to be taught the acceptance of complexity and uncertainty. this is typical for the “buzz word era”, but indications are that this era is rapidly drawing to an end for reasons of relevance and urgency. there are ample indications that complexity in society is increasing (heydari & wade 2014; gosselin & tindemans 2016; coronges, barabási & vespignani 2016). looking back to the aspects of our working definition of complexity, we can see that connectivity, ambiguity and dynamism are applicable to a growing range of real-world systems, problems, challenges and developments. this trend has been strengthening for a number of years, helped by a combination of rapid advances in internet technology and globalization. the expansion of the internet of things, and the anticipated rise of the internet of everything provides just one example of fast growing connectivity dynamism. the world economic forum’s global agenda council on complex systems (2013) recalls the metaphore of a web to describe connectivity in wicked problems like global climate change: “many of the grand challenges that confront humanity often seem to entail impenetrable webs of cause and effect.” (world economic forum’s global agenda council on complex systems 2013, p.2) it is definitely not only on a global scale that complexity is on the rise. on the contrary, complexity is also increasing on the national, regional and local level. tabel 3 provides an example from the author’s professional deskresearch. http://dx.doi.org/10.20377/cgn-63 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 24 university of bamberg press table 3 an example of a real-world complex development in the netherlands aspect of complexity working definition example connectivity as of january 2015, some 390 local governments became responsible for the provision of healthcare services for their citizens. they had to close deals with suppliers from a group of 6,000 healthcare organizations. the deals could be one-year contracts or longer term commitments. healthcare services were categorized into 120,000 different productcodes. the actual provision of healthcare was often organized into new, multidisciplinary teams, with teams typically covering parts of the cities or communities. ambiguity connections within this new ‘ecosystem’ were multi-dimensional. governments and organizations could be linked legally and financially, while local healthcare teams operated as self-organizing networks of multidisciplinary professionals, personally linked through trust to citizens. dynamism connections could change over time, e.g. as one-year contracts came to an end or had to be renegotiated. this change in context caused some local governments to adapt their behaviour towards healthcare organizations, and incorporate new terms in contracts. this, in turn, triggered new actions in the healthcare sector, with some trying to respond to the new terms, and some leaving the arena. second, there is what we could call the argument of urgency. organizations, businesses, institutes and governments are experiencing difficulties in coping with the consequences of growing complexity. for example, the 2010 ibm global chief executive officers study revealed a gap between the level of complexity anticipated by ceos versus their perceived ability to deal with complexity successfully (ibm 2010). as the study noted “our interviews revealed that ceos are now confronted with a “complexity gap” that poses a bigger challenge than any factor we’ve measured in eight years of ceo research.” (ibm 2010, p.19) a possible explanation is, that businesses, governments and other organizations are still mainly hierarchically structured into stable and discrete departments and silos. thus, they lack the ability to mirror the complexity of their environment. metaphorically speaking, one could say that they fail to comply with ashby’s law of requisite variety (boisot & mckelvey 2011). one indication that complexity is getting more ‘hurtful’, is the need to attend to evermore unintended consequences of planned projects, policies or strategies. in our example of table 3, one unintended consequence was the emergence of local governments becoming overly cautious, thus creating budget surpluses instead of using the full budget for healthcare. http://dx.doi.org/10.20377/cgn-63 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 25 university of bamberg press it seems clear, then, that complexity has transcended the “buzz word era” and has reached the priority lists for attention. for instance in 2016, the world economic forum asked: what are the 21st-century skills every student needs? respondents in its future of jobs survey put complex problem solving skills on pole position (world economic forum 2016, pp.21-22). the question of deficiences in regular curricula the relevance and urgency to better cope with complexity that is felt in public organizations and businesses, feeds back to higher professional education. looking back at table 1 this should not come as a surprise, as it states that ”[…] higher vocational education emphasizes training for a specific profession”. therefore, all programs at universities of applied sciences should be based on a consensus profile of the profession at hand: the body of knowledge and skills required for the specific profession. input from organizations, businesses and governments from the professional field is considered important, because they are the future employers of the students. despite this obvious need to carefully monitor changing needs of the students’ future employers, educational institutes find themselves struggling to integrate the changing needs into their curricula. the aforementioned advisory report commissioned by the netherlands association of universities of applied sciences, is a typical case in point (verkenningscommissie hoger economisch onderwijs 2014). it all boils down to the question exactly what needs to be changed in the existing curricula, and what can remain unchanged because it is still functioning satisfactorily. fred janssen, in his recent inaugural lecture as professor of science education at leiden university, stresses education’s deep-rooted focus on teaching students how to solve “carefully structured puzzles” as opposed to teaching how to navigate in unstructured complex situations, which he calls “swamps” (janssen 2017). his chair will explicitly focus on scientific research which aims to bridge the gap. the struggle is certainly not confined to higher education. in april 2017, the inspectorate of education, part of the dutch ministry of education, culture and science, states in its yearly review of primary and secondary education: “, dutch pupils find that their teachers stimulate them to tackle complex problems, less than in other countries” (inspectie van het onderwijs 2017, p.42). the main conclusion of this section is, that future employers are explicitly and persistently calling on educational institutes to enhance complex problem solving skills with their pupils and students. in the next section we venture our approach on how to do just that. from why to what: learning objectives the experiments at windesheim university of applied sciences aimed to enhance students’ ability to deal with complex issues. note how we consciously broadened the scope from problems to the wider concept of issues: problems, challenges, systems, networks and societal developments. note also, that we stepped away from the perspective of problem solving, because we didn’t want to go back to janssens’ frame of carefully constructed puzzles. students are confronted with realworld complex issues, and they have to learn to deal with them, which differs from actually being able to solve them. http://dx.doi.org/10.20377/cgn-63 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 26 university of bamberg press we started out by trying to construct learning objectives that captured the essence of student skills needed to deal with complexity. we let the following design principles guide us: • learning objectives should cover at least one (and, ideally all) of the three aspects of our working definition of complexity, as laid out in the first section: connectivity, ambiguity and dynamism. • learning objectives should mirror the priority of professional education, as laid out in the second section: the purposeful application of scientific insights within contexts of a specific professional field. given these design principles, we decided to develop workshops that were consciously embedded within an existing setting, as opposed to creating something new. in practice, this meant that our new eductional tools had to be “plugged in” as parts of programs where students were already studying real-world complex issues. this ensured that we were going to be working in the right setting: applying insights within a specific professional field. ultimately, we started working with three learning objectives, which we labelled insight, research and dialogue. insight the first learning objective was to create systemic awareness and strengthen students’ synthesizing skills, in order to deepen their insight into the complex issues at hand. our assumption was, that much of the curricula that we were plugged into, focussed on “carefully constructed puzzles”, teaching students to use their analytic skills to deconstruct the issue into discrete, disconnected components. thus, in this traditional approach to problem solving we expected students to forget paying attention to the interconnectedness of agents in the issue. our main objective instead was to get students to focus on connectivity. furthermore, we wanted to nudge them into thinking about the multi-dimensionality of connections and possible tensions that might arise from the existence of various types of relationships (ambiguity). research the second learning objective was to enhance the effectiveness of students’ research into the complex issue, by making use of the insights into connectivity from the first learning objective. here, we assumed that when we were being plugged into the curriculum, students had spent time collecting information about the complex issue at hand. we expected many students to be in some kind of struggle with information overload from a fragmented body of literature, data, etc. our objective here was to let students review the system of their complex issue and assess which of the agents, factors and connections could be underpinned by the information gathered. we called this the known knowns. as students probably weren’t able to underpin everything (yet), we consequently expected known unknowns to emerge. http://dx.doi.org/10.20377/cgn-63 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 27 university of bamberg press dialogue the third learning objective was to foster dialogue within student teams, or between students and real-world stakeholders. here, the idea was to let students share their perspective on the connectivity, ambiguity and dynamism of the complex issue at hand. we expected this dialogue not only to yield new insights, but also to create a movement to a shared vision on the complex issue. having set these objectives, we reached out to lecturers in the current programs at windesheim university. in a variety of departments across the university, we found teachers who were willing to open up their courses for an experimental plug-in workshop. from what to how: experimental workshops by choosing a workshop setting, we created a learning environment of supervised working groups. as noted in the second section, that fitted optimally in the context of higher professional education. as noted in fourth section, by choosing a plug-in setting, we were certain that students were already working on complex issues from within their professional field. table 4 offers an overview of the ecw’s we conducted at windesheim university. http://dx.doi.org/10.20377/cgn-63 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 28 university of bamberg press tabel 4 experimental ecw’s at windesheim university domain bachelor program workshop aspects frequency (example of ) complex issue in specific professional context media journalism 3rd-year students; onesession workshop done twice (april 2016 and september 2016) create a story about house construction, land transactions and local government policy in a specific city or region, using the concept of constructive journalism health applied gerontology 2nd-year students; twosession workshop done once (jan 2017) development of new products in cocreation with elderly in a specific city district, using the good lives model business business & information management 3rd-year students; onesession workshop done twice (november 2015 and november 2016) map out the customer journey of patients in short-stay clinics of a specific hospital, as a part of the design process of a new health care education teaching in primary and secondary education 1st-year students; two-session workshop done twice (both may 2017) make a documentary about the value of testing in primary and secundary education, specifically using the context of schools that hosted your internships multidisciplinary honours programs global project and change management 2nd-year students; threesession workshop done once (may 2017) map out existing youth health care initiatives in a specific city, in public, private commercial and private charitable contexts. give advice on how to increase effectiveness of the city’s program on combating obesity. as can be seen in table 4, we were able to work in quite different programs at the university. it shows that the need for new ecw’s clearly crosses departmental boundaries. a single workshop session took 2.5 to 4 hours. depending on how much time we were given, we were able to conduct one-session workshops, but also multi-session workshops. the main tool we used in the ecw’s, was network visualization. being a configuration of interconnected components, a network is particularly suited to represent the system of connected agents and factors of a specific complex issue. we used the term network literacy, originally coined by netscied, the network science in education initiative to cover a set of activities in the workshops (van der cingel in press): http://dx.doi.org/10.20377/cgn-63 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 29 university of bamberg press 1. interactive presentation to create systemic awareness. in the first part of the workshop we presented and discussed the concepts of complexity and very basic systems thinking. we showed that connectivity can be visualized by network diagrams. 2. class discussions to relate systemic awareness to the real-world issues that students are currently working on. the goal of discussion was to let students discover that the issues they were currently studying, were actually complex. they were nudged into thinking about the question: how can we better deal with the connectivity, ambiguity and dynamism of the complex issue at hand? this was the moment to actually embed complexity-informed perspectives in the context of the current student activities. 3. making and discussing network diagrams. we presented network literacy as a tool to better deal with the complexity in the issues that students were tackling. we asked them to hand-draw agents (e.g. stakeholders), factors (e.g. facilities, or contextual aspects like regulation) and their relationships in a network diagram. we explicitly encouraged students to think about multiple types of relationships. we then created a first moment for dialogue, as students shared their diagrams, in a round of peer review. students could use the insights from this round to alter their diagram, or make an improved version. when they were working in teams, this was the moment that they were asked to merge the individual sketches into a “team diagram”. the improved diagrams were subjected to second round of peer reviewing. then, we applied some basic concepts from network science, e.g. clustering, core and periphery and small-world properties, to let students study the “topology” of their complex issue. we typically ended with a discussion of the possible impact of the dynamism inherent to complex issues. if time permitted, we asked students to come up with what if-scenarios and we invited them to regard the scenarios as pathways in a landscape of future outcomes. the experiments taught us at least two important lessons. first, we learned that ecw’s could definitely enhance students’ insights into their complex issue. the most obvious indicator was simply witnessing the students while they were creating their network diagrams. network visualization urged them to think about connectivity and ambiguity. moreover, it enabled them to share their thoughts with each other and start a dialogue. on some occasions, the dialogue led to new insights, e.g. missing links. one student stated: “to me, the workshop was a real eye-opener. i am not very good at relating and integrating concepts and research findings. thus, my research mostly results in lots of incoherent lists of findings. by connecting the dots, i gained far more insight.” in all programs except for the honours program, it was clear that this was a complete new way of approaching complex issues. second, we quickly discovered that we would need more workshop sessions if we were to give the aspect of dynamism the attention it deserves. in the majority of the experiments, creating systemic awareness and creating the “topology” network diagrams took up all of our time. evaluating with regular course instructors, most of them expressed the need to properly address aspects of dynamism. http://dx.doi.org/10.20377/cgn-63 complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 30 university of bamberg press conclusion and discussion in this article, we argued that there is a compelling case for designing, testing and implementing embedded complexity workshops (ecw’s) in dutch higher professional education. in the real world, from the local to the global level, problems, challenges, systems, networks and developments are looking less like carefully constructed puzzles. instead, they are looking evermore like intricate webs of related agents and factors. businesses, governments and all other future employers of today’s students, experience an increasing relevance and urgency to call on educational institutes to prepare students to deal with complex issues. it is safe to say that complexity has transcended the “buzz word era” in professional education. in dutch higher professional education, this call to action from students’ future employers has left universities of applied sciences searching for an optimal response. at windesheim university of applied sciences, we were given the opportunity to set up experiments for new ecw’s. we conducted plug-in workshops in various bachelor programs, ranging from journalism to applied gerontology. overall, the workshops clearly fulfilled a need for students. providing the tool of network visualisation helped many students to uncover connectivity and ambiguity in their complex issue. on various occasions, it also helped students to tackle information overload using the network diagram as a tool to order their research. though the workshops enabled students to study the status quo of their complex issue more effectively, the aspect of dynamism, which is inherent to our working definition of complexity, was not covered sufficiently. the topic of dynamism and uncertainty is going to be addressed in future workshops. another topic that could possibly lead to workshop re-design in the near future, is the question of transfer: which learning outcomes, if any, will be used by the students in the weeks after the workshop? do students who experience new perspectives on complex issues change the way they actually operate in complex professional contexts? such questions notwithstanding, the introduction of complexity-informed perspectives in teaching programs in dutch professional higher education is quite promising. references retrieved from https://www.studyinholland.nl/education-system retrieved from http://www.vsnu.nl/f_c_ingeschreven_studenten.html retrieved from http://www.vereniginghogescholen.nl/hogescholen retrieved from https://www.windesheiminternational.nl/ retrieved from: https://sites.google.com/a/binghamton.edu/netscied/ boisot, m., & mckelvey, b. (2011). complexity and organization: environment relations: revisiting ashby’s law of requisite variety. in peter allen, steve maguire & bill mckelvey (eds.), the sage handbook of complexity and management. sage: london. coronges, k., barabási, a.-l., & vespignani, a. (2016). future directions of network science: a workshop report on the emerging science of networks [workshop report]. arlington, va: virginia tech applied research corporation. gerrits, l. (2012). punching clouds: an introduction to the complexity of public decision making. litchfield park az: emergent publications. http://dx.doi.org/10.20377/cgn-63 https://www.studyinholland.nl/education-system http://www.vsnu.nl/f_c_ingeschreven_studenten.html http://www.vereniginghogescholen.nl/hogescholen https://www.windesheiminternational.nl/ https://sites.google.com/a/binghamton.edu/netscied/ complexity, governance & networks – vol. 4, no 1 (2018) special issue: teaching complexity, p. 19-31 doi: http://dx.doi.org/10.20377/cgn-63 31 university of bamberg press gosselin, d., & tindemans, b. (2016). thinking futures: strategy at the edge of complexity and uncertainty. leuven: lannoo campus. heydari, b., & wade, j. (2014). complexity: definition and reduction techniques: some simple thoughts on complex systems. csdm. ibm (2010). capitalizing on complexity: insights from the global chief executive officer study. somers, ny: ibm global business services. inspectie van het onderwijs (2017). de staat van het onderwijs 2015-2016. [onderwijsverslag]. utrecht: inspectie van het onderwijs. janssen, f. (2017). grip krijgen op complexiteit: onderwijs voor het ‘moeras’. [oratie] leiden: universiteit leiden. doi: http://hdl.handle.net/1887/51601 verkenningscommissie hoger economisch onderwijs (2014). wendbaar in een duurzame economie: een externe analyse van het economisch domein ten behoeve van de verkenning hoger economisch onderwijs (heo). den haag: vereniging hogescholen. van der cingel, p. (in press). advances in non-technical network literacy: lessons learned in tertiary education. in c. cramer, h. sayama, l. sheetz, s.m. uzzo, & m. porter (eds.), network science in education: tools and techniques for transforming teaching and learning. cham: springer international publishing. windesheim (2017, september 27). windesheim again high in keuzegids hbo ranking [press release]. retrieved from https://www.windesheim.com/aboutwindesheim/news/2017/september/windesheim-again-high-in-keuzegids-hbo-ranking/ world economic forum (2016). the future of jobs: employment, skills and workforce strategy for the fourth industrial revolution. [executive summary] cologny/geneva: world economic forum. world economic forum’s global agenda council on complex systems (2013). perspectives on a hyperconnected world: insights from the science of complexity. n.p.: world economic forum. http://dx.doi.org/10.20377/cgn-63 http://hdl.handle.net/1887/51601 complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 7 university of bamberg press better than rational: a naturalistic view of economic governance author: j.w. stoelhorst university of amsterdam, the netherlands e-mail: j.w.stoelhorst@uva.nl the purpose of this paper is to present a recently emerging evolutionary approach to the study of human organization that i will refer to as ‘naturalistic’, and to highlight its integrative potential for the social and administrative sciences. this naturalistic approach considers the various forms of human cooperation as products of gene-culture coevolutionary processes, and in doing so goes to the heart of the collective action problems that are central to explaining the human condition. moreover, in building empirically grounded explanations of human behavior and organization, it also offers an alternative to the traditional view of governance in economic theory, with its emphasis on decentralized exchange and rational self-interested choice. the naturalistic approach both explains why human nature has evolved to the point where we often can do ‘better than rational’ in the face of the social dilemmas underlying collective action problems, and why modern forms of social, economic, and political organization are nevertheless always prone to being undermined by these same dilemmas. keywords: collective action; institutional economics; evolutionary theory; multi-level selection; geneculture co-evolution “economic transactions take place not only in markets, but also within firms, associations, households, and agencies. whereas economic theory has comprehensively illuminated the virtues and limitations of markets, it has traditionally paid less attention to other institutional arrangements.” (from the press release announcing the sveriges riksbank prize in economic sciences in memory of alfred nobel 2009)i “while no full-blown theory of collective action yet exists, evolutionary theories appear most able to explain the diverse findings from the lab and the field and to carry the nucleus of an overarching theory.” (ostrom, 2000: 138) 1. introduction much to the surprise of many economists, in 2009 political scientist elinor ostrom earned a nobel prize in economics for helping bring “economic governance research from the fringe to the forefront of scientific attention”.ii by awarding her this prize, the economic sciences prize committee recognized ostrom’s work as a fundamental contribution to ‘institutional economics’: the analysis of the ‘sets of rules that govern human interaction’. both her work on governing the commons and that of her co-recipient economist oliver williamson on transaction cost economics, were seen as addressing “head-on the challenges posed by the 1991 laureate in economic sciences, ronald coase … [w]hat exactly do organizations such as firms and associations accomplish that cannot be better accomplished in markets?” in this paper i argue that this fundamental question is best answered from what i will refer to as a ‘naturalistic’ perspective, that is, by considering human behavior and organization as the result of gene-culture co-evolutionary processes (cf. stoelhorst & richerson, 2013). i believe this perspective is in line with ostrom’s general outlook on economic governance (cf. wilson, ostrom complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 8 university of bamberg press & cox, 2013). as the quote from her work at the beginning of this paper testifies, she saw evolutionary theory as the most promising candidate for the development of a general theory of collective action. by awarding her the nobel, the economic sciences prize committee put ostrom’s work squarely in a long, if quite diverse, tradition of institutional analysis in economics that includes the early 20th century american institutionalism of veblen and commons, the new institutionalism of coase, williamson and north, as well as more recent evolutionary game theoretic work (for recent reviews of these three traditions, see, respectively, hodgson & stoelhorst, 2014; menard & shirley, 2014; hodgson & huang, 2012). ostrom’s work can thus be seen as bridging the interests of political and administrative scientists (the main audience of this journal) and economics research on institutional complexity and evolution (the topic of this special issue). but in addition to this, her work also links in with a recent interdisciplinary stream of research at the intersection of biology, psychology, anthropology, sociology, and economics that studies human cooperation by combining gene-culture co-evolutionary arguments with a multilevel selection perspective (e.g. bowles & gintis, 2011; haidt, 2012; nowak, 2011; wilson, 2012). the main purpose of this paper is to show the relevance of this latter stream of research for institutional analysis and design. as this stream of research is still somewhat fragmented, my first goal is to bring together what i consider to be its main insights into human behavior and organization, and to articulate the naturalistic perspective on economic governance that these insights suggest. my second goal is to detail how this naturalistic perspective differs from the traditional view of governance in economic theory, with its emphasis on rational choice, self-interest, and arm’s length market transactions as the key to economic welfare (cf. gowdy et al., 2013).iii while the fact that the naturalistic perspective takes us away from some cherished notions in traditional economic theory may make it a relatively difficult ‘sell’ to many economists and political scientists, it has the important advantage that it is consistent with the empirical evidence on actual human behavior and organization from both the lab and the field. in particular, a naturalistic perspective explains how evolved behavioral dispositions allow humans to often do ‘better than rational’ (ostrom, 1998) in the face of the social dilemma’s underlying collective action problems. in doing so, it also suggests a very specific answer to coase’s question as posed by the nobel prize committee: what ‘firms, associations, households, and agencies’ accomplish that cannot better be accomplished in markets is to help us create economic welfare by triggering evolved behavioral dispositions that allow us to overcome our short-term self-interest. to substantiate this answer to coase’s question, i will below first articulate a redefinition of the problem of economic governance as it has traditionally featured in economic theory. in particular, i will argue that the social dilemmas underlying collective action problems are much more central than most economic theory assumes, because they do not only affect the provision of public but also of private goods. at the same time, i will also argue that their solution is less problematic than typically thought, because of differences between the behaviors of the stylized homo economicus featuring in traditional economic models and actual homo sapiens. i will subsequently ground these claims in insights into the nature of human cooperation from evolutionary theory. i will argue that the main problem in sustaining human cooperation is not our alleged self-interest, but the evolutionary mismatch between our evolved moral and social psychology, on one hand, and the scale of modern human organization, on the other. more specifically, the main problem of human cooperation is not to have rational self-interested agents play ‘cooperate’ in social dilemmas. it is rather to design institutions that trigger evolved behavioral dispositions that put bounds on pursuing our short term self-interest, even if these dispositions originally evolved to sustain cooperation on a much smaller scale. complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 9 university of bamberg press 2. the problem of economic governance redefined from adam smith, via walras and the marginal revolution, to general equilibrium models and welfare economics, the canonical view of human cooperation in economic theory is of the ‘invisible hand’ of the market guiding the actions of self-interested agents towards an outcome that maximizes the collective interest. this view models markets as perfectly decentralized systems that consist of rational agents interacting through arm’s-length competitive exchanges (demsetz, 1988). one of its undeniable attractions is that it posits a system that will maximize collective welfare on the basis of an automatic, bottom-up, self-organizing process – and that it does so despite the self-interested nature of its constituent agents.iv the policy implications are clear: to maximize economic welfare, institutions should facilitate decentralized exchange, for instance through enforcing individual property rights and promoting free markets. one of the problems with this view on economic governance was recognized relatively early on: markets will not be able to coordinate the actions of rational self-interested agents to provide public goods (samuelson, 1954). in contrast to private goods, which are competitive and (assuming institutions that effectively enforce property rights) excludable, public goods are noncompetitive and non-excludable. given these characteristics, a rational self-interested agent will recognize that the optimal course of action is to free-ride on the contributions of others, which means that a decentralized system will not be able to provide public goods. in other words, for as far as economic welfare also depends on the provision of public goods, a decentralized system of self-interested rational agents will not maximize social welfare. this is particularly troubling because securing property rights, a precondition for perfect decentralization to work, is itself a public good. the solution to the provision of public goods proposed in economic theory is perfect centralization, the provision of the public good by a central authority that imposes a tax on the individual agents in the system to provide the public good. in combination with perfect decentralization, the idea of perfect centralization defines the ‘two goods, two systems’ view of economic governance that forms the theoretical backdrop against which ostrom’s nobel prize in economics can be understood (see figure 1). public goods (centralization) private goods (decentralization) figure 1: the traditional two goods, two systems view as ostrom explained, an important part of her contribution to our understanding of economic governance was to demonstrate that this ‘two goods, two systems’ view is too limited (ostrom, 2010). there are not two but four types of goods, and, as scholars in public administration have long realized, the provision of these additional goods typically involves ‘polycentric’ governance rather than perfect (de)centralization, which is to say that a diverse array of public and private agencies is involved in providing them (see figure 2). subtractability high low excludability low cprs public goods high private goods toll goods figure 2: from two to four goods, after ostrom (2010) complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 10 university of bamberg press what figure 2 does not capture is that the problem with the ‘two goods, two systems’ view of economic governance runs much deeper. figure 2 could be interpreted as proposing that economic governance involves perfectly decentralized markets for the provision of private goods, perfect centralization for the provision of public goods, and polycentric solutions for the provision of common pool resources and toll goods.v but a more realistic picture of economic governance would dispense with perfect centralization and perfect decentralization altogether. first, consider the notion of perfect decentralization as the way to provide private goods. perfect decentralization assumes that individual agents only interact through arm’s-length market transactions. but this is not how most private goods are provided: firms play a central role in this process and they are themselves solutions to a collective action problem that has been referred to as ‘team production’ (alchian & demsetz, 1972).vi in other words, in contrast to the unitary agent view of firms in traditional economic models, firms are in fact cooperative entities consisting of multiple agents with conflicting interests. second, consider the notion of perfect centralization as the way to provide public goods. perfect centralization assumes a single benevolent, omniscient, and omnipotent central agent.vii but, again, this is not how government works. governments are collections of agencies whose existence and interplay throw up collective action problems of their own. figure 3: a naturalistic view (concepts of perfect (de)centralization are useful theoretical benchmarks, but the provision of all goods involves social dilemmas, which are solved in the polycentric context of households, firms, associations, and agencies) the actual landscape of economic governance is more like figure 3. this figure explicitly captures the fact that perfect centralization and decentralization are merely useful theoretical abstractions, and that all economic governance is a matter of polycentric solutions to collective action problems at different levels of analysis. in reality there is no such thing as perfect centralization or perfect decentralization. ‘households, firms, associations and agencies’ are all manifestations of solutions to collective action problems. in other words, it is not just that the actual landscape of economic governance is more complicated than the picture painted in the ‘two goods, two systems’ view. the problem is rather that this view fundamentally misrepresents that landscape. economic welfare is not the result of competition among individual agents facilitated by a benevolent designer imposing the rules of competition from above. economic welfare is the result of bottom-up interactions among agents that allow them to reap the benefits of cooperation by forming functional groups. it is not that competition does not play an important role in this process, but not all of this competition is guided by markets. moreover, when it is, competition is mostly between groups of individuals such as firms, which first need to solve the social dilemma perfect centralization perfect decentralization agencies associations households firms complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 11 university of bamberg press of team production to have products or services to compete with at all. in other words, whether it is the provision of private goods by firms (alchian & demsetz, 1972), the provision of public goods (samuelson 1948) and club goods (buchanan, 1965), or the governance of common pool resources (hardin 1968; ostrom 1990), in all cases the central problem of economic governance is to overcome social dilemmas.viii 3. the ultimate explanation of human cooperation how humans are able to overcome social dilemmas is exactly the question that is at the core of the naturalistic perspective on human behavior and organization that is emerging at the intersection of the natural and social sciences (stoelhorst & richerson, 2013). in the context of evolutionary theory, cooperation in general, let alone the unique nature of human cooperation, poses a fundamental problem for reasons that are similar to why it is a problem for economic theory. in fact, the traditional emphasis on decentralized competitive exchange among rational self-interested agents in economics (i.e. the role of markets in coordinating human action) reflects the baseline model of biological evolution as a process driven by competition for scarce resources among individual organisms: the darwinian variation-selection-retention algorithm underlying biological evolution can only reward behaviors that are in individual organisms’ self-interest (i.e., allowing them to pass on more of their genes to future generations). given this algorithm, how can cooperative behaviors evolve at all? in answering this question, evolutionary theory makes an important distinction between ultimate and proximate explanations (mayr, 1961). a proximate explanation accounts for behaviors in mechanistic terms and explains how behaviors come about. for instance, a proximate explanation of cooperative behaviors in public good games may invoke strong reciprocity – the tendency of human agents to reward cooperative behaviors and punish uncooperative behaviors, even when doing so comes at a personal cost (fehr & gintis, 2007). such a proximate explanation needs to be complemented by an ultimate explanation that accounts for why behaviors such as strong reciprocity, which are at odds with rational self-interest, may have evolved. an ultimate explanation explains behaviors in functional terms: it shows how particular behaviors can increase an individual organism’s fitness. over the last 50 years, evolutionary theory has identified five mechanisms that can explain the evolution of cooperation (nowak, 2006): kin selection, or inclusive fitness theory (hamilton, 1964), direct reciprocity (trivers, 1971; axelrod, 1984), indirect reciprocity (e.g. alexander, 1987), network reciprocity (ohtsuki et al., 2006) and group selection, or multi-level selection theory (e.g. sober & wilson, 1998).ix all five mechanisms are likely to have played a role in the evolution of the human ability to cooperate, but the crux of the ultimate explanation of human cooperation seems to lie in a combination of multi-level selection (sober & wilson, 1998; wilson & wilson, 2007) and geneculture co-evolution (boyd & richerson, 1985; richerson & boyd, 2005). what makes humans unique is that we can sustain very large-scale cooperation among non-kin. this empirical fact rules out explanations in terms of kin selection or reciprocity alone. kin selection cannot explain that human cooperation extends to individuals that are not genetically related. direct reciprocity cannot explain the large scale of human cooperation, because cooperation based on reciprocity quickly breaks down when group size increases (boyd & richerson, 1988). even indirect reciprocity fails in the face of the empirical evidence about human cooperative behaviors, because humans even display cooperative behaviors in one-shot interactions with anonymous strangers without reputation effects (e.g. fehr and fischbacher, 2003). complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 12 university of bamberg press to explain such behaviors, we need to turn to multi-level selection theory, in which groups as well as individuals are units of selection (cf. sober & wilson, 1998; wilson & wilson, 2007). this theory recognizes that individual organisms are often organized in larger functional groups. whenever this is the case, we need to distinguish the selection pressures from within-group competition for scarce resources, which favor behavior that is beneficial to the individual, on the one hand, and selection pressures from between-group competition for scarce resources, which favor behavior that is beneficial to the group, on the other hand. note that this distinction mirrors the structure of a social dilemma, where free-riding is the behavior that is beneficial to the individual, and cooperating the behavior that is beneficial to the group. the crux of multi-level selection theory is that whenever individuals are organized into groups that compete with each other, the net selection effect of within-group competition and between-group competition may actually favor genes that code for within-group cooperative behaviors. although the evolution of cooperation on the basis of multi-level selection acting on genetic evolution alone is possible (sober & wilson, 1998), the explanatory value of the multi-level selection framework is much increased if, in addition to genetic mechanisms, we allow cultural mechanisms to play a role as well (e.g. bell et al., 2009; henrich, 2004). doing so leads to the following explanation of the origins of the unique human ability to cooperate (cf. bowles & gintis, 2003; henrich, 2004; richerson & boyd, 2005). in the between-group competition for scarce resources among our tribal ancestors it was not only the genes of the individuals within the group that were selected for, but also their culturally transmitted ideas and behaviors. ceteris paribus, groups that evolved cultures that supported cooperation could out-compete other groups, and cultures that favored cooperation spread. moreover, these cultures changed within-group selection pressures to favor genes that predispose humans to cooperative behaviors. as a result, humans are a highly group-selected species: over evolutionary history we became social animals with moral instincts conducive to solving social dilemmas and sustaining cooperation among non-kin. 4. proximate explanations of human cooperation if, as the naturalistic view asserts, we are a moral rather than a purely self-interested species, then how do our moral instincts manifest themselves? they do, for instance, in a host of findings in behavioral economics and social psychology demonstrating that humans facing social dilemma games do not, on average, behave as theories of rational self-interested choice would predict (see van lange et al., 2013 and van lange et al, 2014 for reviews). one specific example of this is the finding already mentioned above, that in public good games many people play as ‘strong reciprocators’ rather than as ‘rational self-interested agents’: they reward others’ cooperative behaviors and punish others’ uncooperative behaviors, even if this comes at a personal cost (fehr & gintis, 2007).x strong reciprocity is clearly at odds with maximizing self-interest and is especially relevant for our understanding of economic governance because it occurs in the context of a game that mirrors the basic pay-off structure of team production, common pool resource management, and the provision of public goods. the most likely ultimate explanation for such seemingly irrational behavior is the combination of multi-level selection and gene-culture evolution summarized above. in fact, it has been argued that most, if not all, of our social and moral psychology has evolved to help us sustain norms that facilitate within-group cooperation in the context of between-group competition (greene 2013; haidt 2012; richerson & boyd, 2005). on this view, our social behaviors are the result of adaptations that originally emerged to sustain cooperation in the small-scale societies in which our ancestors lived for most of human evolutionary history. these behaviors not only complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 13 university of bamberg press include strong reciprocity (related to maintaining within-group cooperation), but, for instance, also gossiping (related to cheater detection and reputational mechanisms) and in-group favoritism and out-group aggression (related to between-group competition) (haidt, 2012; greene, 2013). the proximate explanations of such behaviors, in turn, are the neural mechanisms causing the emotions that trigger and reinforce these behaviors (fehr & camerer, 2007). examples of this are guilt and shame (related to observing group norms oneself), anger and contempt (related to punishing others that do not observe group norms), and feelings of reward when punishing noncooperators (greene, 2013). thus, while the ultimate function of our moral psychology is to help us overcome social dilemmas, the proximate effect is to put bounds on the pursuit of our self-interest. theories of the nature of these bounds invoke other-regarding preferences (meaning that individuals do not just care about their own outcomes but also about the outcomes for others) and preferences for fairness (such as reciprocal fairness and inequity aversion).xi what these explanations have in common is that they posit internal pressures that can be thought of as changing the pay-off matrix that a rational self-interested player would perceive to one in which cooperation actually becomes the ‘rational’ strategy (van lange et al., 2014). homo sapiens does not play like homo economicus because we have the ability to internalize moral norms that make us ‘better than rational’ at solving social dilemmas. moreover, our psychology is rich in neural mechanisms that trigger various emotions that make us behave in ways that help us reinforce these norms. while it is, therefore, not the case that the actual behaviors of homo sapiens correspond to our homo economicus typecasting in traditional economic theory, neither is it the case that our behaviors correspond to a view of our species as homo sociologicus, whose internalized norms always lead to norm-regarding behavior (fehr & gintis, 2007). for instance, results from public good games show that 30% of subjects do not contribute to the public good and free-ride on the contributions of others, while 50% of subjects cooperate conditionally, meaning that they contribute to the public good as long as others are doing so (fischbacher, et al., 2001). in other words, a substantial minority of individuals does in fact behave in line with the assumptions of standard economic theory, at least in the context of the anonymous interactions of lab experiments with public goods. but the majority of individuals does not, and, crucially, given the opportunity to punish free riders, even if this punishment is costly and even in the context of anonymous interactions, it is the majority’s tendency to cooperate that prevails (e.g. fehr & gächter, 2000). a suggestion that may offer an additional key to the proximate explanation of these results is that human behavior may be guided by two competing neural pathways (cf. boone, et al., 2008; declerck, et al., 2013) that we could label ‘strategic’ and ‘moralistic’. while the two pathways can be active in parallel, one of the two will ultimately dominate and inform actual behavior. the strategic pathway is more likely to result in self-interested behaviors, while the moralistic pathway is needed for the regulation of social relationships (rai & fiske, 2011) and is more likely to trigger positive and negative reciprocity. if this is indeed how the human mind works, it raises interesting questions about how environmental cues may contribute to the dominance of one or the other pathway, and thus trigger different individual behaviors and, ultimately, collective outcomes. for instance, the finding that labeling a prisoners dilemma game ‘the community game’ or ‘the wall street game’ affects cooperation, with significantly higher cooperation in the community game (liberman, et al., 2004), seems to be consistent with this. findings like this are particularly rich in implications for the administrative sciences, because they suggest that institutions may affect the level of cooperation in social dilemmas by triggering different neural pathways.xii complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 14 university of bamberg press 5. a naturalistic view of economic governance what do the arguments above mean for economic governance? the first three rows of table 1 summarize the main points made so far, namely that social dilemmas are both more central to economic governance and less problematic to solve than traditional economic theory has assumed. they are more central because economic welfare crucially depends on solving collective action problems, even in the provision of private goods. most private goods require firms to produce, and the viability of firms depends on agents solving team production problems. at the same time, social dilemmas are less problematic to solve than traditional economic theory assumes, because for all of human evolutionary history our biological and economic success has depended on solving collective action problems like team production. as a result, human psychology has evolved in ways that help us overcome the social dilemmas that emerge when our individual interest in within-group competition diverges from the collective interest in betweengroup competition. traditional view naturalistic view main cause of welfare arm’s length exchange (independent action) collective action (mutual dependence) nature of competition single-level selection (competition between individuals) multi-level selection (competition between individuals and groups) behavioral assumption self-interest bounded self-interest the function of institutions to facilitate decentralized exchange among selfinterested agents to help boundedly selfinterested agents overcome social dilemmas what institutions are single rule complexes imposed top-down competing rule complexes evolving bottom-up what undermines institutions self-interest evolutionary mismatch policy focus secure property rights and provide other public goods keep competition peaceful and leaders honest table 1: a comparison of the traditional and naturalistic views of economic governance note that the claim that social dilemmas are less problematic than traditional economic theory assumes does not mean that they are unproblematic. in fact, the crux of multi-level selection theory is that it acknowledges the tension between individual and group interest, meaning that functional groups are always vulnerable to being undermined by self-interested behavior (campbell, 1994). nevertheless, a naturalistic view on human behavior and organization does suggest a fundamental redefinition of the problem of social dilemmas in economic governance: the problem of economic governance is not how to get self-interested agents (homo economicus) to play cooperate in the face of social dilemmas, but how to get boundedly selfinterested human beings (homo sapiens) to do so. this is where the role of institutions comes in, as summarized in the last four rows of table 1. the purpose of institutions is not primarily to facilitate competition among individual selfinterested agents, but to help boundedly self-interest humans overcome social dilemmas. institutions, be they formal or informal, achieve this through the ‘rules of the game’ they impose on us. these rules work in one of two ways (alone or in combination): by triggering the bounds on complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 15 university of bamberg press our self-interest that change the pay-off matrix from within (i.e., by appealing to our moral sentiments), and/or by changing the pay-off matrix from without (i.e., by some form of punishment). in contrast to traditional economic theory, when considering the latter mechanism there is no presumption of a benevolent designer that can impose a set of rules from the top down. rules complexes rather evolve from the bottom up, as the result of competition between individuals and, at higher levels of analysis, competition between groups. one immediate implication of this is the explicit recognition that institutions are politically contested, and that the rules they impose are historically and culturally specific phenomena that may favor the interests of particular subgroups. another implication of a naturalistic view of institutions is that it points in a fundamentally different direction when it comes to the problem at the core of designing effective institutions. this problem is not so much humanity’s presumed self-interest, but ‘evolutionary mismatch’. the idea of evolutionary mismatch is that while our social and moral psychology evolved in ways that help us overcome our short-term self-interest, this psychology is the result of solving social dilemmas in the small-scale societies in which our evolutionary ancestors lived. as a species we are particularly good at solving social dilemmas in the context of relatively small groups where face-to-face contact, reputation effects, and group identities allow for ways to discipline possible free riders. but our psychology is not necessarily particularly well-adapted to the large-scale, complex societies with their often anonymous interactions in which we live today. in the popular rendition of this core idea of evolutionary psychology, we are constructing our modern worlds on the back of ‘stone age minds’ – and the two do not always match. two problems, in particular, stand out as obvious effects of evolutionary mismatch. the first is the problem of keeping competition, and particularly between-group competition, peaceful. because our psychology evolved in the context of between-group competition, our cooperative tendencies tend to be much stronger for in-group members, and, in fact, often go hand-in-hand with outright aggression to out-group members (greene, 2013). the second is the problem of keeping leaders honest. while our evolutionary ancestors were very successful in maintaining egalitarian societies (boehm, 1993), this has changed dramatically since, some 10.000 years ago, the agricultural revolution started the process of increasing the scale of our societies. one of the defining features of the tribal cultures with which our psychologies co-evolved was that they were very effective in keeping in check those who would want to emulate the ‘alfa male’ behaviors of many other primate species (boehm, 1993). the fact that the increasing scale of our societies went hand-in-hand with an increasing stratification turned this feature of human societies on its head. these two manifestations of evolutionary mismatch, out-group aggression and the tension between our evolved egalitarian preferences and the tendency towards inequality in large-scale societies, present themselves as even more fundamental issues for institutions to solve than providing public goods such as property rights. attempts to provide public goods can at least appeal to aspects of our psychologies that specifically evolved to help us solve the social dilemmas involved. but in the case of the two evolutionary mismatch problems, our evolved psychologies are actually the root cause of the problem. 6. implications for the administrative sciences what does the emerging naturalistic view on human behavior and organization mean for the administrative sciences? two interrelated claims have been made about the potential value of the naturalistic view, and both are especially relevant for the administrative sciences, which are by nature interdisciplinary and applied. the first claim relates to the interdisciplinary nature of the complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 16 university of bamberg press administrative sciences, and is that a naturalistic view offers an integrative paradigm for the social sciences (e.g. fehr & gintis, 2007; fowler & schreiber, 2008; gintis, 2006; mesoudi, whiten & laland, 2006; stoelhorst & richerson, 2013; wilson and gowdy, 2013). for instance, gintis (2007: 1) argues that a naturalistic view offers the “analytical and empirical bases to construct the framework for an integrated behavioral science” that can reconcile the ‘distinct and incompatible ways’ in which the various behavioral and social sciences have modeled human behavior. reconciling incompatible models of human behavior, such as the homo economicus and homo sociologicus views, is especially important for the coherence of interdisciplinary fields like the administrative sciences that build on a variety of behavioral and social sciences and that are therefore particularly susceptible to importing conflicting assumptions about human behavior into their theories. the second claim relates to the applied nature of the administrative sciences, and is that a naturalistic view offers a foundation for a ‘science of intentional change’ (wilson, et al., 2014). the basic idea underlying this second claim is that “examples of successfully managed behavioral and cultural change at scales ranging from individuals to small groups and large populations” are “examples of managing evolved mechanisms of phenotypic plasticity” (wilson, et al., 2014: 395). ‘phenotypic plasticity’ refers to the ability of organisms to respond adaptively to changes in their environments during their life time. this ability is particularly well-developed in humans because of our capacity for symbolic communication. wilson et al.’s (2014) claim is that, in addition to our genotype, networks of symbolic relations also regulate human behavior.xiii applying this view to the arguments in this paper leads to the suggestion that the applied nature of the administrative sciences can be thought of as a concern with understanding how institutions can be intentionally designed so that the networks of symbolic relations that they represent facilitate human cooperation. some important building blocks for this understanding are brought together in relational models theory (fiske, 1991, 1992). relational models are models that people use (unselfconsciously) “to plan and to generate their own action, to understand, remember, and anticipate others’ action, to coordinate the joint production of collective action and institutions, and to evaluate their own and others’ actions” (fiske, 2004: 3). the four relational models are communal sharing (a relationship of unity, community, and collective identity), authority ranking (a relationship of hierarchical differences, accompanied by the exercise of command and complementary display of deference and respect), equality matching (a relationship among equals manifested in balanced reciprocity), and market pricing (a relationships where people compute cost/benefit ratios and pursue their self-interest) (fiske, 1991). these models are not only cognitive schemata, but also comprise needs, motives, evaluative attitudes and judgments, as well as emotions, all of which trigger different behaviors in social interactions because they make different relational self-representations salient (‘who am i in relation to the other(s)?’) that are associated with different needs and motivations and involve different rules of behavior (‘what is appropriate behavior for myself and the other in this social interaction?’) (fiske, 1991). relational models theory holds that four models suffice to generate the very high diversity in social relationships observed in practice (fiske, 1991). seen within the broader framework of a naturalistic perspective on human behavior, the four relational models may be understood as the elemental building blocks that enable and constrain the symbolic relations that, in addition to our genotype, govern human behavior. they enable them because of their combinatorial potential and flexibility with respect to the specific rules that they allow, but they also constrain them because they are the result of deeply ingrained aspects of our evolved social and moral psychology. if this is indeed true, then we must evaluate institutions in terms of the specific (combination of) relational models that they trigger. and if we want to intentionally guide the evolution of complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 17 university of bamberg press institutions in ways that allow us to create more economic value, then we must ask how triggering each of the four relational models will affect individuals’ behaviors in the context of social dilemmas. bridoux and stoelhorst (2016) is an attempt to answer this question for the case of team production. it argues that of the four relational models, market pricing, the model that traditional economic theory emphasizes as most conducive to furthering our economic welfare, is actually the model that is least conducive to solving social dilemmas. in fact, market pricing exacerbates social dilemmas because it appeals to parts of our psychology that are likely to trigger strategic as opposed to moralistic reasoning. whereas the other three models trigger interpersonal identities and other-regarding motivations, market pricing triggers self-interest. this prediction is borne out by a host of empirical findings (e.g. bowles, 2008). thus, for as far as our economic welfare depends on solving social dilemmas, as a naturalistic perspective on economic governance suggests it does, traditional economic theory, with its emphasis on markets, may be pointing the administrative sciences in the wrong direction with respect to the types of institutions that we should be designing. if our economic welfare fundamentally depends on appealing to the bounds on our self-interest, then institutions that make us frame our social relationships in terms of market pricing would in fact achieve the opposite, and we would do well to design institutions that help us frame our social relations more in terms of authority ranking, equality matching, and especially communal sharing, because it is these frames, as opposed to market pricing, that trigger our moral sentiments. 7. conclusion i hope to have achieved three things. the first is to present the main elements of the naturalistic perspective on human behavior and organization that is emerging from work at the intersection of the biological and social sciences that is grounded in gene-culture co-evolution and multi-level selection. the second is to show that this perspective offers an empirically-grounded alternative to the traditional economic research paradigm with its emphasis on decentralized exchange among rational self-interested agents as the key to economic welfare. the third is to suggest how this alternative paradigm can inform the administrative sciences. social dilemmas are both more central to economic governance and less problematic to solve than traditional economic theory has assumed. they are more central because economic welfare crucially depends on solving collective action problems – not just because these problems play a role in the provision of public goods, but also because they play a role in the provision of most private goods, which typically involve team production. humans solve these collective action problems through establishing functional groups like households, firms, associations, and agencies. this means that the competitive landscape that drives our economic welfare is fundamentally misrepresented if we think of it in terms of single-level competition among individual agents, as traditional economic theory does. the actual landscape is one of both withingroup and between-group competition, and has been for all of human evolutionary history. as a result of this long history, our psychology includes evolved dispositions that put bounds on our self-interest, which help us to overcome the social dilemmas that stand in the way of forming functional groups. commenting on ostrom’s nobel prize, economist paul romer noted the following: most economists think that they are building cranes that suspend important theoretical structures from a base that is firmly grounded in first principles. in fact, they almost always invoke a skyhook, some unexplained result without which the entire structure collapses … a typical conclusion [is] complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 18 university of bamberg press that rules that assign property rights and rules that let people trade lead to good outcomes. what’s the skyhook? that people will follow the rules. why would they respect the property rights of someone else? [economists] may have in mind something like this: police officers will arrest people who don’t follow the rules. but this is just another skyhook. who are these police officers? why do they follow rules? … economists who have become addicted to skyhooks find it hard to even understand what it would mean to make the rules that humans follow the object of scientific inquiry. if we fail to explore rules in greater depth, [we] will have little to say about the most pressing issues facing humans today …cheers to the nobel committee for recognizing work on one of the deepest issues in economics. bravo to the political scientist who showed that she was a better economist than the economic imperialists who can’t tell the difference between assuming and understanding.xiv cheers and bravo as well for ostrom’s (2000) early recognition of evolutionary theory’s potential to develop an overarching theory to address this fundamental issue. as i hope to have shown, there has been substantial progress on developing such a theory since 2000, and as the still somewhat disparate findings from research on human cooperation from an evolutionary perspective are being brought together we are not just beginning to see the contours of an overarching theory, but indeed of an integrative research paradigm for the social sciences. applied to coase’s question as paraphrased by the nobel prize committee, ‘what do organizations such as firms and associations accomplish that cannot be better accomplished in markets?,’ this paradigm leads to the unambiguous answer that firms and associations, as well as other collectives like households and agencies, are vehicles to overcome the social dilemmas underlying collective action problems. what they accomplish is to provide contexts that trigger the moralistic neural pathways that allow us to sustain cooperative social relations. perhaps even more important than this basic insight is the fact that a naturalistic perspective leads to the conclusion that the coasean question is simply misdirected, or at least misdirecting, in presuming the primacy of markets. while evolutionary theory also takes competition as its starting point, a naturalistic perspective on human behavior and organization acknowledges that competition has long, long ago been complemented with, and in many ways been superseded by, cooperative arrangements – and that these cooperative arrangements explain the bulk of our economic progress. the landscape of economic governance is one of multi-level competition in which organizations dominate markets. it is a landscape that, as simon (1991) observed, a martian visiting earth, unaware of reigning economic theory, would no doubt describe as an ‘organizational economy’ rather than a ‘market economy’. moreover, economic theory’s traditional emphasis on markets over organizations as the key to economic welfare is far from neutral. for organizations to function, our institutions need to trigger the moralistic processes that put bounds on our self-interest. the logic of markets, by triggering strategic reasoning, does the opposite. thus, a naturalistic view suggests that traditional economic theory may have been pointing the administrative sciences in the wrong direction when it comes to designing institutions that further our economic welfare. references alchian aa, & demsetz h. (1972). production, information costs, and economic organization. american economic review 62(5): 777-795. alexander, r.d., (1987). the biology of moral systems. hawthorne, ny: aldine de gruyter. axelrod, r., (1984). the evolution of cooperation. new york, ny: basic books. complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 19 university of bamberg press bell, a.v., richerson, p., & mcelreath, r., (2009). culture rather than genes provides greater scope for the evolution of large-scale human prosociality. proceedings of the national academy of sciences of the united states of america 106, 17671–21767. blair, m.m. & stout, l.a. (1999). a team production theory of corporate law, virginia law review, 85(2): 247-328. boehm, c., (1993). egalitarian behavior and reverse dominance hierarchy. current anthropology 34, 227–254. boone, c., declerck, c. h., & suetens, s. (2008). subtle social cues, explicit incentives and cooperation in social dilemmas. evolution and human behavior, 29(3), 179–188. bowles, s. (2008), policies designed for self-interested citizens may undermine "the moral sentiments": evidence from economic experiments, science, 320: 1605-1609. bowles, s., & gintis, h., (2003). the origins of human cooperation. in p. hammerstein (ed.), the genetic and cultural origins of cooperation. cambridge, ma: mit press. 429-444. bowles, s., & gintis, h. (2011). a cooperative species: human reciprocity and its evolution. princeton, ny: princeton university press. boyd, r., & richerson, p.j., (1985). culture and the evolutionary process. chicago, il: university of chicago press. boyd, r., & richerson, p.j., (1988). the evolution of reciprocity in sizable groups. journal of theoretical biology, 132, 33756. bridoux, f.m., & stoelhorst, j.w. (2016), stakeholder relationships and social welfare: a behavioral theory of contributions to joint value creation. academy of management review. 41(2): 229-251. buchanan, j.m. (1965). an economic theory of clubs. economica, 32(125): 1–14. byrne, d., & gallaghan, g, (2014). complexity theory and the social sciences: the state of the art, milton park: routledge. campbell, d.t., (1994). how individual and face-to-face group selection undermine firm selection in organizational evolution. in: baum, j.a., singh, j.v. (eds.), evolutionary dynamics of organizations. oxford, uk: oxford university press, pp. 23–38. declerck, c. h., boone, c., & emonds, g. (2013). when do people cooperate? the neuroeconomics of prosocial decision making. brain and cognition, 81(1), 95–117. demsetz, h. (1988). the theory of the firm revisited. journal of law, economics, & organization, 4(1): 141-161. fehr, e., & camerer, c.f. (2007). social neuroeconomics: the neural circuitry of social preferences. trends in cognitive sciences, 11(10): 419-427. fehr, e., & fischbacher, u. (2003). the nature of human altruism. nature, 425: 785-791. fehr, e., & gächter, s. (2000). cooperation and punishment in public goods experiments. american economic review, 90: 980–994. fehr, e. and gintis, h. (2007), human motivation and social cooperation: experimental and analytical foundations, annual review of sociology, 33:43–64 fischbacher, u., gächter, s., & fehr, e. (2001). are people conditionally cooperative? evidence from public goods experiments, economic letters, 71: 397-404. fiske, a. p. (1991). structures of social life: the four elementary forms of human relations. new york, ny: free press. fiske, a. p. (1992). the four elemental forms of sociality: framework for a unified theory of social relations. psychological review, 99: 689–723. fiske, a. p. (2004). relational models theory 2.0. in n. haslam (ed.). relational models theory: a contemporary overview. mahwah, nj: erlbaum. 3–25 fowler, j.h., & schreiber, d., (2008). biology, politics, and the emerging science of human nature. science, 322: 912– 914. gintis, h., (2006). a framework for the integration of the behavioral sciences. behavioral and brain sciences, 30, 1-61. gintis, h., bowles, s., boyd, r., & fehr, e. (2003). explaining altruistic behavior in humans. evolution and human behavior, 24: 153-172. gowdy, j.m., dollimore, d.e., wilson, d.s., & witt, u. (2013). economic cosmology and the evolutionary challenge. journal of economic behavior and organization, 90 (supplement): s11–s20. greene, j.d. (2013). moral tribes: emotion, reason, and the gap between us and them. new york, ny: penguin. guala, f. (2012). reciprocity: weak or strong? what punishment experiments do (and do not) demonstrate,behavioral and brain sciences, 35 (1): 115. hagen, e.h., & hammerstein, p. (2006). game theory and human evolution: a critique of some recent interpretations of experimental games, theoretical population biology, 69: 339-348. haidt, j. (2012). the righteous mind: why good people are divided by politics and religion. new york, ny: pantheon. hall, p.a., & taylor, r.c.r (1996). political science and the three new institutionalisms. political studies, 44: 936-957. hamilton, w.d., (1964). the genetical evolution of social behavior. journal of theoretical biology, 37: 1-52. hardin, g. (1968). the tragedy of the commons. science, 162: 1243–1248. henrich, j., (2004). cultural group selection, coevolutionary processes and large-scale cooperation. journal of economic behavior and organization, 53: 3-35. http://journals.cambridge.org/action/displayjournal?jid=bbs http://journals.cambridge.org/action/displayjournal?jid=bbs complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 20 university of bamberg press hodgson, g.m., & huang, k. (2012). evolutionary game theory and evolutionary economics: are they different species? journal of evolutionary economics 22: 345–366. hodgson, g.m., & stoelhorst, j.w. (2014). introduction to the special issue on the future of institutional and evolutionary economics, journal of institutional economics, 10(4): 513-540. liberman, v., samuels, s., & ross, l. (2004). the name of the game: predictive power of reputation vs. situational labels in determining prisoner’s dilemma game moves. personality and social psychology bulletin, 30: 1175–1185. mayr, e., (1961). cause and effect in biology, science, 134: 1501-1506. menard, c., & shirley, m.m. (2014). the future of new institutional economics: from early intuitions to a new paradigm? journal of institutional economics, 10(4): 541-565. mesoudi, a., whiten, a., & laland, k.n. (2006). towards a unified science of cultural evolution.behavioral and brain sciences, 29: 329347. nowak, m.a., (2006). five rules for the evolution of cooperation. science, 314: 1560-1563. nowak, m.a., (2011). super cooperators: evolution, altruism and human behavior, or why we need each other to succeed. edinburgh, uk: canongate. ohtsuki, h., hauert, c., lieberman, e., & nowak, m.a. (2006). a simple rule for the evolution of cooperation on graphs and social networks. nature 441: 502-505. ostrom, e. (1990). governing the commons: the evolution of institutions for collective action. cambridge, uk: cambridge university press. ostrom, e. (1998). a behavioral approach to the rational choice theory of collective action. american political science review, 92(1): 1–22. ostrom, e. (2000). collective action and the evolution of social norms, journal of economic perspectives, 14(3): 137-158 ostrom, e. (2010). beyond markets and states: polycentric governance of complex economic systems. american economic review, 100: 1–33 rai, t.s, fiske, a.p. (2011). moral psychology is relationship regulation: moral motives for unity, hierarchy, equality, and proportionality, psychological review, 118(1): 57–75. richerson, p.j., & boyd, r., (2005). not by genes alone: how culture transformed human evolution. chicago, il: the university of chicago press. samuelson, p.a. (1954). the pure theory of public expenditure. review of economics and statistics, 36(4): 387–89. sanderson, s.k. (2007). evolutionism and its critics. boulder, co: paradigm publishers. simon, h.a. (1991). organizations and markets, journal of economic perspectives, 5(2): 25–44 sober, e., & wilson, d.s., (1998). unto others: the evolution and psychology of unselfish behavior. cambridge, uk: harvard university press. stoelhorst, j.w., & richerson, p.j. (2013). a naturalistic theory of economic organization, journal of economic behavior and organization, 90 (supplement): s45-s56. sugden, r. (1986). the economics of rights, co-operation and welfare. oxford, uk: blackwell trivers, r.l., (1971). the evolution of reciprocal altruism. quarterly review of biology, 46: 35-57. van lange, p.a.m., balliet, d., parks, c.d., & van vugt, m. (2014). social dilemmas: the psychology of human cooperation. oxford, uk: oxford university press. van lange, p.a.m., joireman, j., parks, c.d., & van dijk e. (2013). the psychology of social dilemmas: a review. organizational behavior and human decision processes, 120: 125-141. wilson, d.s., & gowdy, j., (2013). evolution as a general theoretical framework for economics and public policy. journal of economic behavior and organization. 90 (supplement): s3–s10. wilson, d.s., hayes, s.c., biglan, a., & embry, d.d. (2014). evolving the future: toward a science of intentional change. behavioral and brain sciences, 37: 395–460. wilson, d.s., ostrom, e., & cox, m.e. (2013). generalizing the core design principles for the efficacy of groups journal of economic behavior and organization. 90 (supplement): s21–s32 wilson, d.s., wilson, e.o., (2007). rethinking the theoretical foundation of sociobiology. the quarterly review of biology, 82, 327-348. wilson, e.o. (2012). the social conquest of earth. ww norton & company http://journals.cambridge.org/action/displayjournal?jid=bbs http://journals.cambridge.org/action/displayjournal?jid=bbs complexity, governance & networks – special issue: complexity, innovation and policy (2017) 7-21 j.w. stoelhorst: better than rational a naturalistic view of economic governance doi: http://dx.doi.org/10.20377/cgn-39 21 university of bamberg press endnotes i http://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/2009/press.html ii the quotes in this paragraph are either from the press release of the 2009 prize in economic sciences in memory of alfred nobel (see previous footnote) or from the advanced information: http://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/2009/advanced.html iii my focus on economic theory is in line with the topic of this special issue, but comes at the expense of ignoring related discourses in the social sciences on institutions (e.g. hall & taylor, 1996), evolutionary arguments (e.g. sanderson, 2007) and complexity (e.g. byrne & callaghan, 2014). iv the other attractions are that it can be modelled relatively easily and that it implies a world in which agents have maximum individual freedom. v toll goods are ostrom’s term for what economists had referred to as ‘club goods’. vi the problem of team production occurs when there are complementarities between agents’ inputs in the production process, so that the collective output is more than the sum of the individual inputs. note that this is the case in most production. this situation creates a metering problem in the sense that it becomes very costly, or even impossible, to assess the marginal contributions of individual agents to the collective output. this creates a social dilemma akin to the public good dilemma: as the market can only ‘reward’ the collective output, the individual incentive becomes to free ride on the contributions of others (i.e. minimizing one’s own effort while still sharing in the spoils of the collective output). alchian & demsetz (1972) explain the existence of firms as the result of the need to solve this problem. this is a different (and ultimately much more convincing) explanation than the one offered by coase and, by extension, williamson, ostrom’s co-recipient of the 2009 nobel (cf. demsetz, 1988). blair and stout (1999) offer a more fully developed team production theory of the modern corporation. vii in the words of sudgen (1986: 3, emphasis in original) “most modern economic theory describes a world presided over by a government (not, significantly, by governments), and sees this world through the government’s eyes. the government is supposed to have the responsibility, the will and the power to restructure society in whatever way maximizes social welfare; like the us cavalry in a good western, the government stands ready to rush to the rescue whenever the market “fails,” and the economist’s job is to advise it on when and how to do so.” viii van lange et al. (2014: 8) define social dilemmas as ‘situations in which a non-cooperative course of action is (at times) tempting for each individual in that it yields superior (often short-term) outcomes for self, and if all pursue this non-cooperative course of action, all are (often in the longer-term) worse off than if all had cooperated’. this definition includes prisoner’s dilemmas, chicken/hawk-dove dilemmas, and assurance/trust dilemmas. ix all five mechanisms explain how the (long-term) benefits of cooperation in terms of fitness can overcome the (shortterm) costs incurred by the individual displaying the cooperative behavior. in shorthand, kin selection explains cooperation in terms of genetic relatedness, direct reciprocity in terms of tit-for-tat, indirect reciprocity in terms of reputation mechanisms, network reciprocity in terms of assorting, and multi-level selection in terms the tension between within-group selection and between-group selection. x for some qualifications of the more extreme interpretations of the work on strong reciprocity, see hagen & hammerstein (2006) and guala (2012). xi see fehr & gintis (2007) and gintis, et al. (2003) for summaries from the perspective of economic theory, and van lange, et al. (2013) for summaries from the perspective of social psychology. xii these findings also begin to address some of the reservations of hagen and hammerstein (2006) about current interpretations of positive and negative reciprocity in experimental games. xiii wilson et al. (2014: 400) refer to networks of symbolic relations as ‘symbotypes’, which they describe as ‘tacit systems of higher-order relations’ that become independent of the physical objects originally triggering these relations, and that are ‘maintained by their utility, coherence, and role in a social community’. this is not unlike institutions, which regulate behavior by defining the rules of human interaction. xiv http://paulromer.net/skyhooks-versus-cranes-the-nobel-prize-for-elinor-ostrom/ http://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/2009/press.html http://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/2009/advanced.html http://paulromer.net/skyhooks-versus-cranes-the-nobel-prize-for-elinor-ostrom/ better than rational: a naturalistic view of economic governance 1. introduction 2. the problem of economic governance redefined 3. the ultimate explanation of human cooperation 4. proximate explanations of human cooperation 5. a naturalistic view of economic governance 7. conclusion a critical discussion of complexity theory: how does 'complexity thinking' improve our understanding of politics and policymaking? complexity, governance & networks – vol. 3, no 2 (2017), p. 1-11 paul cairney, robert geyer: a critical discussion of complexity theory how does complexity thinking improve our understanding of politics and policymaking? doi: http://dx.doi.org/10.20377/cgn-56 1 university of bamberg press a critical discussion of complexity theory: how does 'complexity thinking' improve our understanding of politics and policymaking? authors: paul cairney, robert geyer department of history and politics, university of stirling, great britain, e-mail: p.a.cairney@stir.ac.uk department of politics, philosophy and religion, lancaster university, great britain, e-mail: r.geyer@lancaster.ac.uk in this article, we present a critical discussion of complexity theory. we ask: what does it really offer policy studies? we suggest that its stated advantages-interdisciplinarity, theoretical novelty, and empirical advance--are generally exaggerated and based more on hope than experience. in that context, we identify a cautiously positive role for complexity theory, primarily as a way to bridge academic and policymaker discussions by identifying the role of pragmatic responses to complexity in policymaking. keywords: complexity theory; interdisciplinarity; policy studies; democratic accountability; pragmatism introduction it is customary to describe complexity theory as new, exciting, and interdisciplinary (mitchell, 2009, p. x). its advocates suggest that it offers a new way of seeing the world, a scientific paradigm to replace “reductionism,” a way for many academic disciplines to use the same language to explain key processes, and the potential for an impressively broad and rich empirical base (geyer & cairney, 2015; cairney, 2012a, 2012b; geyer & rihani, 2010, p. 12; mittleton-kelly, 2003, p. 26; sanderson, 2006, p. 117; room, 2011, pp. 6-7; klijn, 2008, p. 314; little, 2008, pp. 29-30; lewis & steinmo, 2008, pp. 15-20, 2010, p. 237). in this article, we present a critical discussion of these high expectations, examining how they translate into something new for political and policy science. we ask: what does complexity theory really offer policy studies? we argue that its focus on greater interdisciplinarity is potentially misleading, its theoretical appeal may be more about conceptual consolidation than novelty, and it may take some time to demonstrate that it is empirically more valuable than more established theories. we identify a cautiously positive role for complexity theory as a way to bridge academic and policymaker discussions on politically important concepts such as democratic accountability and responsibility. in particular, we compare a democratic imperative, to hold elected policymakers to account for policy, with a pragmatic imperative, identified in complexity theory to recognise the limits to central government control. http://dx.doi.org/10.20377/cgn-56 mailto:p.a.cairney@stir.ac.uk mailto:r.geyer@lancaster.ac.uk complexity, governance & networks – vol. 3, no 2 (2017), p. 1-11 paul cairney, robert geyer: a critical discussion of complexity theory how does complexity thinking improve our understanding of politics and policymaking? doi: http://dx.doi.org/10.20377/cgn-56 2 university of bamberg press complexity theory: a language that spreads across disciplines? complexity theory can be sold as a way to encourage interdisciplinarity: if we can share a theoretical outlook, language, and set of research methods, we can combine disciplinary approaches to tackle major social and environmental problems. there are three obstacles to the realization of such a vision. first, it is possible that we can maintain a similar language among disciplines only if it is highly abstract or superficially similar. the danger is that the same words mean different things in each discipline. certainly, one would often get this impression about superficial similarity when speaking with natural scientists about complexity ideas in physics and biology. for example, in our conversations with colleagues in other disciplines, we have found some interesting differences in language that are easily resolved – such as that ‘first’ order change means almost no change in policy studies (with reference to hall, 1993) but major change in physics (akin to hall’s 3rd order, kuhnian change) – but others that are more difficult to overcome, such as the meaning of ‘chaos’ and the role of deterministic arguments (which seem more useful to explain natural rather than social systems). in complexity, a key difference in meaning may be in the discussion of “emergence” in the absence of a central brain or central control. in cell biology we can witness completely local interaction without a centre. in politics, there is a centre – central government – and our focus is on emergence despite – rather than the absence of its role. these are different processes which require different explanation (cairney and geyer, 2015). second, for us, complexity theory makes most sense when grounded in a well-established literature. as we discuss below, its value to policy studies may be to consolidate existing concepts rather than provide a completely new way of thinking. yet, the necessity to make specific sense of complexity in each discipline undermines a general understanding: it takes time and training to become sufficiently aware of the relevant literature in each discipline; and, the translation involves using well-established concepts in one discipline that mean little in another. third, scholars express major differences of approach and understanding even within single disciplines. for example, when cairney (2012b) wrote a policy-focused article on complexity in 2012, each reviewer noted the profound absence of discussion of particular individuals (such as prigogine) or schools (such as santa fe or brussels) without agreeing on whose absence mattered. in other words, we identify a tendency for scholars to become excited at the prospect of interdisciplinarity when they begin to talk and recognise each other’s language and interest in complexity, only to find that they may not be talking about the same thing (or the same complexity scholars). complexity does not represent new ways of looking at the policymaking world some of complexity’s claims to novelty come from a sense that complexity theory is “antireductionist” (see geyer & rihani, 2010, pp. 74-75) or marking a major challenge to the old ways of doing science, in which, for example, we establish general laws or study individual parts of larger systems (mitchell, 2009, p. x). yet, in the social sciences, there have been decades-long debates about the use of systems to explain social and political behaviour (cairney, 2012a, pp. 113-114) and a greater sense of ‘antireductionism’ – or ‘post-positivism’ – for some time (fischer & forester, 1993). one can point to http://dx.doi.org/10.20377/cgn-56 complexity, governance & networks – vol. 3, no 2 (2017), p. 1-11 paul cairney, robert geyer: a critical discussion of complexity theory how does complexity thinking improve our understanding of politics and policymaking? doi: http://dx.doi.org/10.20377/cgn-56 3 university of bamberg press isolated examples in political science of scholars identifying simple regularities or trends in behaviour, but also a tradition of case study analysis to generate rich descriptions of specific decisions and events. it is therefore difficult to establish the sense that complexity is a new way of thinking, rather than simply the right way to think. a new-ish way of looking at the world? instead, complexity theory may help us bring together many strands of the political science literature into one framework. for example, cairney (2013) uses a discussion of complexity to supplement existing discussions of “evolutionary theory,” in policy studies, which also includes multiple streams analysis and punctuated equilibrium theory (kingdon, 1984, 1995; baumgartner & jones, 1993, 2009;). or, as we demonstrate below, elements of complexity theory link to a key elements of policy studies. we make reference to four main properties of complex systems and their relevance to policy studies. 1. negative and positive feedback ‘negative and positive feedback’ describes a tendency in complex systems for some inputs of energy to be dampened while others are amplified. small actions can have large effects and large actions can have small effects. this process is described extensively in jones and baumgartner’s work on information processing (jones & baumgartner, 2005; baumgartner & jones, 1993, 2009; baumgartner et al., 2014). they suggest that the cognitive ability of policymakers, and their ability to gather information, is limited. they can only pay attention to a small fraction of the issues for which they are responsible. they have to ignore most and promote a few to the top of their agenda. they receive the same amount of information over time, ignoring most for long periods (negative feedback) and paying disproportionate attention to some (positive feedback). consequently, the controlling capacity of the centre is limited to the small number of issues to which policymakers pay particular attention or seek to solve energetically. 2. strange attractors: regularities of behaviour which may be interrupted by short bursts of change positive and negative feedback extends to local parts of complex systems. for example, the instructions of central governments are dampened or amplified by actors responsible for policy delivery. much depends on the patterns of attention paid by policymakers at the “centre.” in theory, they could pay attention to, and influence, any part of the system. however, to do so, they have to ignore most other parts. consequently, all rules that develop in institutions or policy networks could be challenged at any time, but most tend not to be. consequently, ‘strange attractors’ can describe the tendency for regular patterns of policymaking behaviour to persist in most cases despite the ever-present potential for policy instability (again, this is a feature of baumgartner and jones’ work; see also bovaird, 2008: 320; geyer and rihani, 2010: 39). for example, they may be present in ‘institutions’ which represent sets of rules to which people adhere, causing regular patterns of behaviour. http://dx.doi.org/10.20377/cgn-56 complexity, governance & networks – vol. 3, no 2 (2017), p. 1-11 paul cairney, robert geyer: a critical discussion of complexity theory how does complexity thinking improve our understanding of politics and policymaking? doi: http://dx.doi.org/10.20377/cgn-56 4 university of bamberg press 3. sensitivity to initial conditions and “path dependence” the phrase ”sensitivity to initial conditions” describes the contribution of events and decisions made in the distant past to the formation of institutions that influence current practices. for example, when an initial commitment to a policy becomes established it produces ‘increasing returns’ over time: as people adapt to, and build ‘institutions’ around, the initial decision, it becomes increasingly costly to choose a different path (pierson, 2000; room, 2011, 7-18). initial choices are reinforced when the rules governing systemic behaviour become established and difficult to change. as a result the bulk of policy is repetitive: most policy decisions are based on legislation that already exists, most public expenditure is devoted to activities that continue by routine, and policy implementation continues long after policymakers have lost interest (rose, 1990; hogwood and peters, 1983; lindblom, 1959; 1979). ‘critical juncture’ begins to describe the infrequent and often dramatic series of events or decisions required to challenge such routines. 4. “emergence” from the interaction between elements at a local level the idea of ‘emergence’, in policy studies, can relate to attempts by central governments to control the system (or, in some cases, recognise the limits to centralisation and encourage ‘localism’). emergence refers to behaviour which results from local interaction, based on locally defined rules, with an emphasis on the extent to which local behaviour takes place despite central government policies or rules. this concept resonates with the well-established literature on policy implementation and governance. for example, lipsky (1980) frames local behaviour in terms of the limits to which ‘street level’ actors can meet central demands, and the extent to which they draw on their own judgement and professional training when interacting with service users. local actors face so many targets, rules and laws that no public agency or official can be expected to fulfil them all. in fact, many may be too vague or even contradictory, requiring ‘street level bureaucrats’ to choose some over others. or, central governments may introduce performance measures which limit the discretion of delivery organizations but relate to a small part of government business. the broad theme of emergence has also been a key feature of modern accounts of ‘governance’ (rhodes, 1997; bevir and rhodes, 2003: 6; kooiman, 1993). it examines how governments have sought to respond to limited central control, particularly during the peak of new public management (npm) which describes the application of private business methods to government (including attempts to secure order through hierarchical management structures and targets for public bodies to meet). governance scholars often note that central governments struggle to maintain order and, in many cases, have exacerbated their limited control by introducing a wide range of new public service delivery functions which rely on public bodies and organisations in the third and private sectors for their success. newish normative advice for policymakers the same point about newishness can be said for complexity theory’s practical or normative advice: it is justified in a new (and often convincing) way, but the advice itself is not new. complexity theory’s focus on the lack of central government control on local behaviour and policy outcomes can be linked closely to the need to be pragmatic in government, to act with a sense of http://dx.doi.org/10.20377/cgn-56 complexity, governance & networks – vol. 3, no 2 (2017), p. 1-11 paul cairney, robert geyer: a critical discussion of complexity theory how does complexity thinking improve our understanding of politics and policymaking? doi: http://dx.doi.org/10.20377/cgn-56 5 university of bamberg press realism regarding what policymakers can achieve (cairney, 2015a; teisman & klijn, 2008, p. 288; blackman, 2001; cairney, 2012b, p. 349; kernick, 2006; sanderson, 2006, 2009). in other words, if policymakers deny their reliance on other actors to help them understand and adapt to their policy-making environment, they are doomed to make the same mistakes as their predecessors. instead, central government policymakers should embrace interdependence, to pursue more pragmatic solutions based on increasing the freedom of local actors to learn and adapt to environmental signals, such as the responses they get from service users. to address the ever-presence of uncertainty, they should make greater use of trial and error policy making. to address the inevitable gap between policymakers’ aims and policy outcomes, they should change their expectations and the way they think about policy success and evaluation (little, 2012, p. 16; geyer, 2012, p. 32). such advice may be used by policymakers in parts of the uk and scottish governments, although without the need to make reference to the intricacies of complexity theory (cairney & st denny, 2015). indeed, these are the kinds of recommendations provided by lindlom in 1959 (when policymakers were men): making policy is at best a very rough process. neither social scientists, nor politicians, nor public administrators yet know enough about the social world to avoid repeated error in predicting the consequences of policy moves. a wise policy-maker consequently expects that his policies will achieve only part of what he hopes and at the same time will produce unanticipated consequences he would have preferred to avoid. if he proceeds through a succession of incremental changes, he avoids serious lasting mistakes. what is its original empirical base? we expand on some of these issues with reference to our new edited collection on complexity theory (geyer & cairney, 2015). a large part of it is devoted to chapters which outline new ways of thinking in a range of disciplines, or discuss the benefits of particular methods such as agent based modelling. some chapters discuss empirical case studies, but without giving us the ability to combine their insights to help produce accumulated knowledge in a straightforward way. for people seeking a payoff in terms of a bank of empirical case studies, complexity theory may seem like a source of as-yet-untapped potential. it is reasonable for scholars to wonder if we will ever get beyond a focus on new thinking and methods, towards a series of studies which can be linked to each other in a meaningful way. the rise of systematic reviews of the wider policy literature magnifies this problem in two main ways. first, there is now some evidence of accumulated knowledge in more established policy theories (examples include multiple streams analysis – jones et al, 2015 and cairney and jones, 2016 social construction theory (pierce et al, 2014), and the advocacy coalition framework (weible et al, 2009). second, these reviews show how difficult it is to accumulate knowledge when empirical studies make only vague reference to a common theory and struggle to ‘operationalise’ key concepts in a way that can be compared meaningfully to other studies. a broad common reference point – such as complexity theory – is not enough to prompt the accumulation of empirical knowledge. http://dx.doi.org/10.20377/cgn-56 complexity, governance & networks – vol. 3, no 2 (2017), p. 1-11 paul cairney, robert geyer: a critical discussion of complexity theory how does complexity thinking improve our understanding of politics and policymaking? doi: http://dx.doi.org/10.20377/cgn-56 6 university of bamberg press what is complexity theory useful for? yet, we should persist with complexity theory because, if we rejected theories on the basis of the concerns we have raised so far, we would have no theories left. complexity theory still offers a way of thinking about, organising, and explaining future empirical studies. for example, cairney and st denny (2015) use the language of complexity to explore the interaction between new policy solutions/ ideas and the old ways of doing things in policymaking systems. although it is not easy to compare their case study results with others, it is valuable to use a complexity framework to provide detailed understanding of cases, and to use multiple viewpoints – zooming in to examine the perception of actors, or zooming out to observe systems and their environments – to understand the same processes. consequently, the concerns we raise should prompt scholars to explore common topics with people in other disciplines carefully and perhaps to reduce the claims we make about theoretical and empirical novelty while we do so. for example, the way cairney and geyer (2015) describe webb’s (2015) chapter on complexity theory and legal studies ties it closely to the themes in politics and policymaking that we have discussed in this article, but in a way that requires closer examination: he draws on complexity terms, such as emergence and contingency, to argue that the legal process cannot be boiled down to a set of simple laws and rules to be implemented by government bodies such as the courts. rather, people interpret rules and interact with each other to produce outcomes that are difficult to predict with reference to the statute book. as we discuss below, complexity also helps us think about practical problems, such as our ability to be pragmatic (and humble) or somehow manage or influence emergent behaviour. complexity and democratic accountability in politics, a key practical question relates to traditional notions of democratic accountability: how can governments follow the advice of complexity theorists, to let go of ‘order’ and claim to be in control enough to take responsibility for their actions? for example, a common argument in british politics is that the uk government has exacerbated its own “governance problem”, or the gap between general expectations for central government control and its actual powers (see for example richards & smith, 2002, p. 3; marsh et al., 2001; rhodes, 1997; bevir & rhodes, 2003; gains & stoker, 2009). for example, a collection of post-war reforms, many of which were perhaps designed to reinforce central control, has produced a fragmented public landscape and a periodic sense that no one is in control. this outcome presents major problems for the “westminster” narrative of central government and ministerial accountability to the public via parliament. if ministers are not in control of their departments, how can we hold them to account in a meaningful way? yet, in many cases, it is misleading to link these outcomes to specific decisions or points in time, since many aspects of the “governance problem” are universal: policymakers can only pay attention to a small fraction of the issues for which they are responsible; they do not have enough information to make decisions without major uncertainty; policy problems are too multi-faceted and ‘cross-cutting’ to allow policymaking without ambiguity; there is an inescapable logic to delegating decisions to ‘policy communities’ which may not talk to each other or account meaningfully to government; and, delivery bodies will always have discretion in the way they http://dx.doi.org/10.20377/cgn-56 complexity, governance & networks – vol. 3, no 2 (2017), p. 1-11 paul cairney, robert geyer: a critical discussion of complexity theory how does complexity thinking improve our understanding of politics and policymaking? doi: http://dx.doi.org/10.20377/cgn-56 7 university of bamberg press manage competing government demands. in this context, policymaking systems can be described usefully as complex systems, in which behaviour is always difficult to predict, and outcomes often seem to emerge in the absence of central control. the literature on complexity provides some advice about how governments should operate within such complex systems. unfortunately, much of this literature invites policymakers to give up on the idea that they can control policy processes and outcomes. while this may be a pragmatic response, it does not deal well with the need for elected policymakers to account for their actions in a way which stresses central control. what seems sensible to one audience, engaged in developing insights from complexity theory, may be indefensible to another, engaged in the articulation of simple lines of democratic accountability. the language of complexity does not mix well with the language of westminster-style accountability (cairney, 2015a). so, while complexity thinking takes us beyond simple notions of central government accountability, we need to combine a governmental acknowledgement of the limits to its powers with the sense that we can still hold elected policymakers to account in a meaningful way. ideally, this response should be systematic enough to allow us to predict when ministers will take responsibility for their actions, redirect attention to other accountable public bodies, and/ or identify the limited way in which they can be held responsible for certain outcomes. beyond this ideal, we may settle for a government strategy based on explicit trade-offs between pragmatism, in which governments acknowledge the effect of administrative devolution (or, in the case of local authorities, political devolution), and meaningful representation, in which they maintain some degree of responsibility for decisions made in their name. for example, there is a literature describing the extent to which the scottish and uk governments have wrestled with the need to present two images of politics and policymaking: competing for elections on the basis of traditional westminster-style accountability, in which central governments are in control and responsible; but also developing new pragmatic forms of accountability with reference to direct institutional accountability, partnerships between national and local public bodies, and direct forms of accountability driven by the involvement of service users in the design of public services (see gains & stoker, 2009; flinders & skelcher, 2012; richards & smith, 2004; osborne & strokosch, 2013; smith & smyth, 2010, pp. 277-278; cairney, 2015b). as the existence of this literature suggests, these problems are not new, but they remain unresolved. complexity theory gives us a new language to describe important processes and consider how to respond, but as yet without an answer to the dilemma that governments face. complexity, norms, responsibility and pragmatism in other words, a key challenge for complexity is its lack of a clearly developed normative framework. from a complexity perspective, so long as a system maintains core boundaries, is relatively stable and open, and encourages a wide variety of interactions, a plethora of particular structures and outcomes are possible. detailed practical interventions and/or normative choices are uncertain. yet, society (amplified by the mass media and willing political actors) continues to try to reassert a more causal/modernist policy position: someone must be in control and in democracies this should be our democratic representatives (and the bureaucratic machinery of national/regional/local policy that it is supposed to control). this implies a limited view of democracy where: the trend in political life is to insist on controlling responsibility by tightening the external standards of accountability. yet the real problem of our public life is the failure of responsibility, http://dx.doi.org/10.20377/cgn-56 complexity, governance & networks – vol. 3, no 2 (2017), p. 1-11 paul cairney, robert geyer: a critical discussion of complexity theory how does complexity thinking improve our understanding of politics and policymaking? doi: http://dx.doi.org/10.20377/cgn-56 8 university of bamberg press not accountability. increasing demands for accountability often obscure actual responsibility and enhance the gap between responsibility and accountability. (ansell, 2011, p. 134) this dilemma of normative choice and democratic responsibility is not new and something that a variety of philosophic traditions have been wrestling with throughout the past two centuries – for example pragmatist philosophy (see the work of dewey in particular). more recently, there has been a growing exploration of the overlaps between pragmatism and complexity. these overlaps have been touched upon by academics in different fields (doll et al. 2005; rescher 2003), but most importantly by the policy informed work of sanderson (2002, 2006, 2009). for sanderson, dewey provided three relevant contributions. first, dewey’s commitment to a continued “scientific” search for the “truth” through an engaged, democratic and open approach leads to policy making that is based on the best evidence “while recognising that such evidence does not constitute the final word” (sanderson, 2009, p. 710). second, mirroring the earlier labour government’s focus on “what works,” deweyian pragmatism encouraged policy makers to view their actions as “policy hypotheses” that must be continually tested with the best evidence available and through democratic societal debate. third, sanderson pointed out that pragmatism: leads us to recognise that policy making is not a ‘technical’ exercise in instrumental rationality but rather a domain of ‘practical reason’… we are not just concerned with the ‘instrumental’ notion of ‘what works’ but rather with a broader ‘practical’ notion of what is ‘appropriate’ in the circumstances. (p. 711) similar to complexity, pragmatism is based on an engaged view of democracy and society that does not know or propose a final societal outcome but knows that the best way forward is through an open, educated, democratic society engaged in continual learning and dialogue with itself (free expression, debate, speech and interaction) and its governmental structures. pragmatism is the philosophy of “evolutionary learning” and “democratic experimentalism” (ansell, 2011, p. 5). in essence, complexity provides a meta-theoretical position, linking the natural and human societies, while pragmatism provides a justification and framework for societal/public action in a complex and uncertain world. it is based on a reasonably optimistic vision of human rationality and a belief in that ability of well-intentioned individuals and societies to progress, in a generally positive direction, through discussion, learning, experimentation, debate and interaction. being pragmatic in politics is not easy. it may be sensible to produce a range of measures based on a more realistic policymaking philosophy, and potential strategies include: relying less on centrally driven targets, and punitive performance management, in favour of giving local bodies more freedom to adapt to their environment; trial-and-error projects, that can provide lessons and be adopted or rejected quickly; and, to teach policymakers about complexity so that they are less surprised when things go wrong. yet, as tenbensel (2015) makes clear, these strategies should not be selected simply because we reject a caricature of top-down policymaking. one of the best practical applications of complexity and systems inspired policy thinking is found in the excellent 2011 uk munro review of child protection. following a string of highprofile child abuse cases, professor eileen munro was asked to carry out a wide-ranging and indepth review of uk child protection policy. inspired by systems and complexity thinking, munro produced an impressive document that highlighted the failings of the former well-intentioned but misguided approach that resulted in a tick-box culture and a loss of focus on the needs of the child. these weaknesses were further amplified by a media and public culture which demanded that ‘lessons must be learned’ and some individual or process must take responsibility/blame. the core problem, which the review made clear, was that in highly complex situations there are no simple solutions, lines of responsibility or easy targets to blame. what made this situation http://dx.doi.org/10.20377/cgn-56 complexity, governance & networks – vol. 3, no 2 (2017), p. 1-11 paul cairney, robert geyer: a critical discussion of complexity theory how does complexity thinking improve our understanding of politics and policymaking? doi: http://dx.doi.org/10.20377/cgn-56 9 university of bamberg press even worse was a knee-jerk governmental response that demanded ever-growing targeting and audit regimes to show that “lessons” had really been learned. the difficulty, as the munro review aptly demonstrated, was that this did little for the actual protection of children, while greatly complicating the policy process of child protection. hence, one of the key conclusions of her report was that there needed to be a radical reduction in central prescription in order to help social workers move from a compliance to a learning culture, and that we had to recognize that the larger societal pressures to find “someone to blame” (amplified by the mass media) misshaped the policy response to child protection. what the review demonstrates is that complexity theory is not a panacea for all policy problems. however, it is a way of recognising the limits of the dominant evidence-based and target/accountability approach. it is only when these limits are recognised that policy makers can truly understand and take reasonable actions to improve policy making. furthermore, the existence of complexity presents a dilemma for the public: it must continually learn to accept that it cannot simply blame a small number of elected ministers for the ills of government. on the other hand, it should not absolve government entirely; complexity should not be an excuse used by policymakers to take no blame for their actions. helping society to find this zone of reasonable balanced policy making is probably the most significant impact that complexity theory can make. conclusion in many ways, the challenges faced by complexity scholars and policy makers are not new. complexity theory comes from a range of fields and represents a wide range of concepts and implications and is certainly not a unified theory or body of work. it is difficult to get a sense of the “state of the art” in complexity theory, to establish what we know and still need to know. moreover, it is not clear just how far we have come in generating a language of complexity that everyone understands and shares. this problem is magnified when we seek to combine insights from the natural and social sciences: we use the same language of complexity and emergence, but to refer to very different processes. these difficulties are amplified when academics engage with the policy world. political, normative and ethical factors come into play, and significantly increase the level of policy complexity. for example, imagine the challenges of mental health and policy. it is composed of a range of complex organisational systems surrounding individuals with ‘complex needs’ that place a wide variety of demands on social and public services. these public and private organizations and institutions operate within a complex policymaking system, in which mental health only appears very infrequently on the high-level political agenda, and in which policy is often made locally in the relative absence of central direction. the continually emerging and evolving cultural position of mental illness within a society adds another layer of complexity. hence, telling these policy actors that there world is ‘complex’ is not helpful. nevertheless, this realisation of the limits of complexity is actually one of its hidden strengths for this reflects the reality of the policy world. at its heart complexity challenges the more hubristic claims of other approaches (for example funding bodies enthralled to the promises of ‘big data’). this challenge to the ‘quest for certainty’ in policy making (and other public arenas) as john dewey put it in the early 20th century is an essential aspect of modern democracy. complexity doesn’t answer our questions, but it does teach us to be humble in our answers and responsive to the ever-changing natural, social and political world that surrounds us. http://dx.doi.org/10.20377/cgn-56 complexity, governance & networks – vol. 3, no 2 (2017), p. 1-11 paul cairney, robert geyer: a critical discussion of complexity theory how does complexity thinking improve our understanding of politics and policymaking? doi: http://dx.doi.org/10.20377/cgn-56 10 university of bamberg press this may be a simple – and, for some, obvious – point, but it is one that needs to be constantly restated to avoid a continuous return to policymaking built on a simplistic search for central government control. references ansell, c. (2011). pragmatist governance: re-imagining institutions and democracy. oxford university press. baumgartner, f. and jones, b. (1993; 2009). agendas and instability in american politics (1st and 2nd editions). university of chicago press. baumgartner, f., jones, b. and mortensen, p. (2014). ‘punctuated-equilibrium theory: explaining stability and change in public policymaking’ in sabatier, p. and weible, c. (eds.) theories of the policy process 3rd edition.westview press, 59103. bevir, m. and rhodes, r.a.w. (2003). interpreting british governance. routledge blackman, t. (2001). ‘complexity theory and the new public management’, social issues 1(2) bovaird, t. (2008). emergent strategic management and planning mechanisms in complex adaptive systems. public management review, 10 (3), 319-40. cairney, p. (2012a). understanding public policy. palgrave. cairney, p. (2012b). complexity theory in political science and public policy. political studies review, 10 (3), 346–58 cairney, p. (2013). what is evolutionary theory and how does it inform policy studies? policy and politics. 41 (2), 279-98. cairney, p. (2015a). how can policy theory have an impact on policymaking? teaching public administration. 33 (1), 22-39. cairney, p. (2015b). scotland’s future political system. political quarterly. 86 (2), 217-25. cairney, p. and geyer, r. (2015). introduction. in (eds) geyer, r. and cairney, p. handbook of complexity and public policy. edward elgar. cairney, p., & jones, m. d. (2016). kingdon's multiple streams approach: what is the empirical impact of this universal theory?. policy studies journal, 44(1), 37-58. doll, w., fleener, m. j., trueit, d. and st. julien, j. (eds.) (2005) chaos, complexity, curriculum and culture. lang. flinders, m. and skelcher, c. (2012). shrinking the quango state. public money & management, 32 (5), 327-334. gains, f. and stoker, g. (2009). delivering ‘public value’: implications for accountability and legitimacy. parliamentary affairs, 62 (3), 438– 455 geyer, r. (2012). can complexity move uk policy beyond “evidence-based policy making” and the “audit culture”? political studies, 60 (1), 20–43. geyer, p. and cairney, p. (2015). handbook of complexity and public policy. edward elgar. geyer, r. and rihani, s. (2010). complexity and public policy. routledge. hall, p. (1993). policy paradigms, social learning, and the state. comparative politics, 25 (3), 275–96 hogwood, b. and peters, b.g. (1983). policy dynamics. st martin’s press. jones, b. and baumgartner, f. (2005) the politics of attention. university of chicago press. jones, m. d., peterson, h. l., pierce, j. j., herweg, n., bernal, a., lamberta raney, h., & zahariadis, n. (2015). a river runs through it: a multiple streams meta‐review. policy studies journal. kernick, d. (2006). wanted – new methodologies for health service research. is complexity theory the answer? family practice, 23 (3), 385–90. kingdon, j. (1984, 1995) agendas, alternatives and public policies. (1st and 2nd editions), harpercollins. klijn, e. (2008). complexity theory and public administration: what's new? public management review, 10 (3), 299-317. kooiman, j. (1993). socio-political governance: introduction, in (ed.) j. kooiman, modern governance. sage. lewis, o. and steinmo, s. (2008). taking evolution seriously. european university institute, http://spot.colorado.edu/~steinmo/takingevolution.pdf. lewis, o. and steinmo, s. (2010). taking evolution seriously in political science. theory in biosciences, 129 (2–3), 235–45. lindblom, c. (1959). the science of muddling through. public administration review. 19, 79–88. lindblom, c. (1979). still muddling, not yet through. public administration review, 39 (6), 517–25. lipsky, m. (1980). street-level bureaucracy. russell sage foundation. little, a. (2008). democratic piety: complexity, conflict and violence. edinburgh university press. little, a. (2012). political action, error and failure: the epistemological limits of complexity. political studies, 60 (1), 319. marsh, d., richards, d. and smith, m.j. (2001). changing patterns of governance in the united kingdom. palgrave. mitchell, m. (2009). complexity. oxford university press. http://dx.doi.org/10.20377/cgn-56 http://spot.colorado.edu/%7esteinmo/takingevolution.pdf complexity, governance & networks – vol. 3, no 2 (2017), p. 1-11 paul cairney, robert geyer: a critical discussion of complexity theory how does complexity thinking improve our understanding of politics and policymaking? doi: http://dx.doi.org/10.20377/cgn-56 11 university of bamberg press mitleton-kelly, e. (2003). ten principles of complexity and enabling infrastructures, in e. mitleton-kelly (ed.) complex systems and evolutionary perspectives of organisations. elsevier. osborne, s. and strokosch, k. (2013). ‘it takes two to tango? understanding the co‐production of public services by integrating the services management and public administration perspectives’, british journal of management, 24 (s1), 3147 pierce, j., s. siddiki, m. jones, k. schumacher, a. pattison, and h. peterson (2014). social construction and policy design: a review of past applications. policy studies journal, 42 (1), 1–29. pierson, p. (2000). increasing returns, path dependence, and the study of politics, the american political science review, 94 (2) 251–67. rescher, n. (2003). epistemology: on the scope and limits of knowledge. suny press. rhodes, r.a.w. (1997). understanding governance. open university press. richards, d. and smith, m. (2002). governance and public policy in the uk. oxford university press. richards, d. and smith, m. (2004). the “hybrid state’, in s. ludlam and m. smith (eds) governing as new labour. palgrave macmillan. room, g. (2011). complexity, institutions and public policy. edward elgar. rose, r. (1990). inheritance before choice in public policy. journal of theoretical politics. 2 (3), 263–91. sanderson, i. (2002). evaluation, policy learning and evidence-based policy making. public administration, 80, (1), 1-22. sanderson, i. (2006). complexity, ‘practical rationality’ and evidence-based policy making. policy and politics, 34 (1) 11532. sanderson, i. (2009). intelligent policy making for a complex world: pragmatism, evidence and learning. political studies, 57 (4), 699–719. smith, s. and smyth, j. (2010). the governance of contracting relationships’ in (ed) osborne, s. the new public governance. routledge. teisman, g. and klijn, e. (2008). complexity theory and public management. public management review, 10 (3) 287–97. tenbensel, t. (2015). complexity and health policy, in geyer, p. and cairney, p. (eds) handbook of complexity and public policy. edward elgar. webb, t. (2015). critical legal studies and a complexity approach: some initial observations for law and policy, in geyer, p. and cairney, p. (eds) handbook of complexity and public policy. edward elgar. weible, c., sabatier, p. and mcqueen, k. (2009). themes and variations: taking stock of the advocacy coalition framework. policy studies journal. 37 (1), 121–41 http://dx.doi.org/10.20377/cgn-56 international institutions in hard times: how institutional complexity increases resilience complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 46-54 doi: http://dx.doi.org/10.20377/cgn-99 46 university of bamberg press international institutions in hard times: how institutional complexity increases resilience author: benjamin faude london school of economics and political science, united kingdom email: b.faude@lse.ac.uk this paper asks how institutional complexity affects the resilience of global governance. by drawing on sociological differentiation theory, it interprets growing levels of institutional complexity as a process of institutional differentiation which allows the “political system of world society” to mirror the increasing complexity of its social environment. more precisely, the paper suggests that growing levels of institutional complexity enhance the resilience of global governance by providing states with a more diverse set of governance tools and by making backup governance tools available. against this backdrop, it makes two interrelated contributions to the literature on global governance. first, by applying the concept of resilience to global governance, the paper provides the conceptual basis for a novel research agenda on the ability of contemporary global governance to operate under stress. so far, the analytical toolbox of global governance researchers does not contain a concept that enables a theory-driven analysis of international institutions’ ability to facilitate cooperation when confronted with high levels of stress. second, it offers a sense of how the central structural feature of contemporary global governance—institutional complexity—affects its resilience. with these two interrelated contributions, the paper seeks to start a scholarly conversation on the resilience of contemporary global governance. keywords: international institutions; institutional complexity; resilience; sociological differentiation theory introduction institutional complexity is the central structural feature of contemporary global governance. it implies that the latter is not carried out within stand-alone international institutions that operate in isolation from each other, but within international institutions with overlapping competences which interact on a permanent basis in the absence of a central coordinating instance (alter & raustiala 2018; faude & gehring 2017; orsini et al. 2013). as a result, states gain opportunities to pursue interests and maximize payoffs not only within, but also across separately established international institutions (jupille et al. 2013). international relations (ir) scholarship has addressed various implications of institutional complexity for global governance. most importantly, it has studied how institutional complexity affects the problem-solving capacity and the normative legitimacy of global governance (drezner, 2013; faude & große-kreul 2020; faude & parizek 2020; hale et al., 2013; kuyper 2014). at the same time, ir scholars have so far not researched how institutional complexity affects the ability of global governance to operate under stress. put differently, they have not studied how institutional complexity affects the resilience of global governance. indeed, resilience is not even part of the conceptual vocabulary used by ir scholars. at the same time, contemporary global governance has been under tremendous stress in the last two decades. it has been confronted with various instances of rapid and abrupt change, such as pandemics, financial crises, unprecedented numbers of refugees, and terrorist attacks. such exogenous shocks make inter-state cooperation particularly vulnerable. ir scholars define cooperation as the mutual adjustment of patterns of state behavior against the backdrop of discord. in other words, cooperation implies that states (partially) adjust their behavior toward the preferences of other states. international institutions are mailto:b.faude@lse.ac.uk complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 46-54 doi: http://dx.doi.org/10.20377/cgn-99 47 university of bamberg press the vehicles that facilitate such behavioral adjustment. in response to institutional provisions, so the theoretical argument goes, states adjust their behavior in a way that deviates from their preferred behavior in a world devoid of international institutions (keohane 1984: 51). however, history shows that exogenous shocks, such as the ones listed above, incentivize states to follow their preferred behavior instead of adjusting it to institutional provisions and, thus, to the preferences of other states. during pandemics and financial crises and when confronted with large numbers of incoming refugees or terrorist attacks, states tend to put a stronger emphasis on the shortterm implications of their crisis responses for their own constituency than on the long-term consequences of their behavior for the realization of cooperation gains through international institutions. during a recession, for instance, state representatives are inclined to increase trade barriers in order to stabilize their own economies. in general, exogenous shocks tend to lead states away from the cooperative solutions facilitated by international institutions and towards competitive unilateral behavior. thus, exogenous shocks create conditions under which it becomes particularly challenging for international institutions to influence the behavior of states in ways that facilitate cooperation (drezner 2014; kahler & lake 2013; kindleberger 1986). for precisely these reasons, exogenous shocks provide an excellent opportunity to examine whether and in how far contemporary international institutions are able to operate under stress and, thus, whether and in how far they are ultimately fit for this purpose (see gallopín 2006; hasenclever, mayer and rittberger 1996; young 1986). however, as mentioned above, global governance researchers lack a concept which facilitates such an analysis. against this backdrop, the present paper introduces the concept of resilience to the analytical toolbox of global governance researchers. resilience is defined as the ability of global governance to facilitate inter-state cooperation despite heightened pressure for states to deviate from cooperative solutions and, by implication, to follow their preferred behavior in a world devoid of international institutions. in other words, resilience is defined as the ability of international institutions to preserve the (partial) behavioral adjustment of states toward the preferences of other states despite heightened incentives to withdraw from such behavioral adjustment. on this conceptual basis, the paper interprets the growing levels of institutional complexity that global governance has developed over the last decades as a process of institutional differentiation which allows the “political system of world society” to mirror the increasing complexity of its social environment. more precisely, it suggests that growing levels of institutional complexity, first, provide states with a more diverse set of governance tools and, second, make backup governance tools available. by implication, growing levels of institutional complexity are conducive to the resilience of global governance. importantly, however, growing levels of institutional complexity are not caused by the intention of states to increase the resilience of global governance. rather, they are caused by their desire to respond to growing levels of transnational interdependence (see below). thus, the increased resilience of global governance constitutes an unintended result of purposive state action (see merton 1936). in other words, states did not deliberately invest in resilience by increasing institutional complexity. the paper proceeds as follows. the second section clarifies further what resilience is, how it matters, and what it promises to contribute to our understanding of contemporary global governance. the third section introduces the central tenets of sociological differentiation theory and develops how institutional differentiation enhances the resilience of global governance by identifying two causally relevant factors: a diverse set of governance tools and backup governance tools. the fourth section concludes. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 46-54 doi: http://dx.doi.org/10.20377/cgn-99 48 university of bamberg press the concept of resilience applied to global governance the concept of resilience has traditionally been used, first, by biologists to study the ability of populations or species to survive in the face of adversity and, second, by psychologists to study the ability of people to cope with stressful situations.1 it generally describes the ‘‘capacity of a system ... to experience disturbance and still maintain its ongoing functions and controls’’ (holling and gunderson 2002, 50) or, in other words, the capacity to deal with rapid environmental changes (aligica & tarko 2014; herrfahrdtpähle & pahl-wostl 2012; steinberg 2009). in short, the concept of resilience grasps persistence in unfavorable circumstances. by implication, international institutions are resilient, if they are able to preserve the adjustment of states’ behavior towards the preferences of others during exogenous shocks which incentivize states to determine their behavior (more) in accordance with their own preferences (see above). the resilience of international institutions therefore depends on their ability to prepare for and to respond productively to external shocks. more precisely, resilience includes readiness and preparedness on the one hand, and response and adaptation on the other (ponomarov & holcomb 2009). at the level of adaptation, we may differentiate between static adaptation (i.e., featuring redundancy, supplements, and resources in order to deal with shocks without changing the inner structure of the system) and dynamic adaptation (i.e., reorganization, restructuring, abandoning of old routines and adopting of new routines). static adaptation allows international institutions to keep operating in the same way but requires considerable resources. dynamic adaptation requires a (sometimes painful) fundamental reorganization, but keeps the international institution better prepared for future exogenous shocks. however, even if one accepts that states rationally design individual international institutions in a way that those institutions further their own goals (koremenos et al. 2001), one has to concede that it is virtually impossible for them to design individual international institutions in ways that enable them to prove resilient during exogenous shocks. the reason is that exogenous shocks are, first, difficult to anticipate and, second, difficult to navigate because they are characterized by high levels of uncertainty. as a result, their considerable cognitive limitations prevent states from deliberately designing international institutions in ways that enable them to prove resilient during exogenous shocks. importantly, however, due to the proliferation of international institutions over the last decades, contemporary global governance does not consist of (rationally designed) stand-alone international institutions. rather, contemporary global governance consists of sets of international institutions with overlapping competences which interact with one another on a permanent basis without being coordinated by a hierarchical instance (alter & raustiala 2019; faude & gehring 2017). this institutional configuration is described as “institutional complexity” by global governance researchers. it suggests that, when seeking to analyze the resilience of international institutions, we should not conceive of them as stand-alone entities and analyze the ability of, for example, the international monetary fund (imf) to facilitate cooperation during a financial crisis. rather, we should conceive of international institutions as being embedded in and affected by institutional complexity in their ability to facilitate cooperation during exogenous shocks. randall henning’s analysis of the euro crises from the perspective of institutional complexity adopts such an approach, without invoking the concept of resilience, however (henning 2017). 1 it was first used by the canadian ecologist c.s. holling (1973) to describe a characteristic of a complex system. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 46-54 doi: http://dx.doi.org/10.20377/cgn-99 49 university of bamberg press studying resilience from the perspective of sociological differentiation theory the preceding section established that, given the proliferation of international institutions over the past decades, any analysis of the resilience of contemporary international institutions should take their embeddedness into institutional complexity (as defined above) into account. importantly, such an analysis needs to be informed by a general assessment of how the proliferation of international institutions over the past decades has affected the resilience of global governance. in other words, we need to get a sense of how growing levels of institutional complexity have affected the resilience of global governance on a general level before we can turn to more specific research questions. sociological differentiation theory offers a theoretical lens that enables us to develop a sense of how the proliferation of international institutions (that is, growing levels of institutional complexity) affect the resilience of global governance as defined above. at its core, sociological differentiation theory suggests that the growth of institutional complexity constitutes a process of institutional differentiation which is not a destructive, but a productive process (see zürn & faude 2013). it therefore contrasts starkly with prominent arguments put forward by ir researchers which suggest that institutional proliferation gradually weakens international commitments and undermines the general problem-solving capacity of global governance by letting states pursue their parochial interests through forum-shopping (benvenisti & downs, 2007; drezner, 2013; hale et al., 2013a).2 thomas hale and his co-authors, to take a prominent example, forcefully describe institutional proliferation as giving rise to “dysfunctional fragmentation, reducing the ability of multilateral institutions to provide public goods” (hale et al. 2013b: 228). according to them, growing levels of institutional complexity impede “effective cooperation instead of facilitating it” (hale et al. 2013a: 16). by drawing on sociological differentiation theory, the remainder of this section develops a theoretical alternative to this position which refers to the problem-solving capacity of international institutions during exogenous shocks and, thus, to their resilience. the concept of differentiation is inspired by cell division, that is, the division of a unit into several equal, similar, or dissimilar parts, whose interplay gives rise to higher-order-complexity (nassehi 2004: 100). sociological differentiation theory asks how and on the basis of which structuring principle different societal systems or subsystems are defined and distinguished from one another (see zürn et al. 2013). its core claim is that modernity is characterized by functional differentiation replacing earlier forms of differentiation. according to durkheim (1968: 257-264), it is the dynamic density of societal interactions that drives societies towards more complex forms of (functional) differentiation, that is, to a more complex specialization of tasks and division of labor (buzan & albert 2010: 5). thus, an “idea of evolution” in which simple forms of societal organization are replaced by more complex forms is key (buzan & albert 2010: 5). processes such as an increasingly developed “division of labor,” “specialization,” and “role differentiation,” all of which characterize modern society, can be seen as instances of functional differentiation. the present paper restricts the application of sociological differentiation theory to the political realm of the international system which is described as “global governance.” since the latter is characterized by high levels of institutional complexity, it is conceived of as carried out through functionally differentiated international institutions (albert et al. 2013). the central claim of sociological differentiation theory is that societal differentiation is a sensible response to increasing levels of societal complexity. applied to global governance, this claim implies that the proliferation of international institutions (that is, rising levels of institutional complexity) constitutes 2 please note that these contributions do not discuss how institutional complexity affects the resilience of global governance. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 46-54 doi: http://dx.doi.org/10.20377/cgn-99 50 university of bamberg press a functional response to the swelling tide of social problems (environment protection, financial stability etc.) that can be handled best at the international level (held et al. 1999). more precisely, establishing international institutions becomes a functional answer to managing rising levels of societal denationalization and transnational interdependence across a multitude of policy areas (zürn 1998). against this backdrop, growing levels of institutional complexity in global governance can be interpreted as a process of institutional differentiation through which the political system of world society “internally reflects the differentiation of its environment” (albert et al. 2013: 21). in short, increasing institutional differentiation is the corollary of a growing density of transboundary interactions. by differentiating its institutional structure, global governance conforms to the “law of requisite variety”, which posits that dealing successfully with a task requires a system that is at least as complex as the task at hand (ashby 1958). put differently, as the societal environment of a given governance system becomes more complex, the governance system itself must respond by developing a more complex structure. by implication, the “law of requisite variety” suggests that the institutional differentiation of global governance is an effective means to cope with an increasingly complex social environment. against this backdrop, this paper claims that the process of institutional proliferation described in this section has enhanced the resilience of global governance (as defined above) even though such an effect was not intended by the actors. rather, their intention was to manage rising levels of societal denationalization and transboundary interactions on a general level. in that sense, enhanced resilience constitutes an unintended consequence of purposive state action (see merton 1936). a global governance system that features high levels of institutional differentiation (like the contemporary one) compares to a global governance system that is characterized by low levels of institutional differentiation (like the one in 1970) like a well-quipped toolbox with various sizes and types of hammers, screwdrivers, and pliers to the starter set you can buy from ikea; the starter set will help you through most standard situations, but it might not have the tool you need in the case a water pipe bursts. and in such a do-it-yourself emergency situation, you would probably be glad if you knew that in case your screwdriver breaks, there is a spare one in the toolbox. the same two causal mechanisms –a diversity of tools and the availability of backup tools—explain how greater institutional differentiation creates a larger set of available governance actions and thereby increases the resilience of global governance. first, institutional differentiation increases resilience because it accords states a choice between institutions offering different tools for crisis prevention and response. this choice allows states to tailor their responses to unpredictable situations and facilitates cooperative state action in the face of interest heterogeneity (jupille et al. 2013). since crises can arise from a variety of different risk factors, small institutions specialized on specific areas are more likely to be able to identify and address those risks than a large central institution. the same is true for responses to external shocks. they are inherently difficult to predict, and response strategies just as difficult to plan in advance (see above). a greater choice is then conducive to an effective crisis response. for instance, in the case of financial crises, states have recourse to an array of overlapping regional and global institutions (henning 2017; pauly 2008). as a result, there are international institutions that states can use to pool the financial risks of their banks, others help affected states to avoid bankruptcy, yet others provide states with financial regulation mechanisms to reduce market volatility (mayntz 2015). when engaged in conflict prevention and/or management, states can similarly delegate individual tasks to different international institutions. for example, they may assign the military parts of a “post-conflict reconstruction operation” to nato, which has a strong expertise on military interventions, and carry the civilian parts out through the common security and defense policy (csdp) of the european union (eu), which is particularly suited to execute civilian operations (hofmann 2011). in sum, institutional differentiation increases the set of governance actions that are available to complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 46-54 doi: http://dx.doi.org/10.20377/cgn-99 51 university of bamberg press prevent and manage crises. as a result, the challenges created by exogenous shocks can be mastered collectively by sets of international institutions with overlapping competences. a greater choice of institutions does not only enable tailored responses by specialized institutions; it also makes it easier for states to coordinate international action despite interest heterogeneity (see henning 2017). institutional differentiation ensures that states with diverging interests do not have to agree to the crisis response mechanisms offered by a single institution. instead, different groups of states can organize collective responses in separate forums (jupille et al. 2013). when these responses complement one another, the result is likely to be more effective than if states with widely diverging interests had attempted to agree on a common response within a single institution. second, institutional differentiation increases the resilience of global governance because it makes different institutions performing the same or similar tasks available to state actors. a duplication of critical functions creates redundancy, which works as a fail-safe in case one institution fails to perform (kelley 2009). redundancy denotes that two or more international institutions are performing the same specific governance task, for instance regulating the level of co2-emissions or distributing loans to countries (keohane & victor 2011). redundancy and backup options are of crucial importance for global governance because individual institutions tend to become deeply rigid over time. as oran young notes, with the passage of time, (international) institutions “become increasingly entrenched. often, they fall prey to rigidification and suffer from what observers have called institutional arthritis (olson 1982). the effect of this is to sap their resilience and to increase their vulnerability to various types of stress” (young 2009, 378). in other words, “institutional arthritis” reduces the adaptive capacity of institutions, which includes their capacity to respond productively to exogenous shocks (see hanrieder 2015). rigid voting procedures and veto rights lead to the postponement of decisions that are needed to decrease the risk of shocks because they give small groups and even individual states the opportunity to “block” decision-making in situations of interest heterogeneity (see, for example, collier 2006 on deadlock in the wto). according to young, institutions that “are rigid in the sense that they are difficult to amend formally or to adjust informally run the risk of falling prey to rapid changes” (young 2009, 380). even in the absence of heterogeneous interests, however, bureaucratic hurdles and the need to align a large number of states are likely to slow down coordination processes in large, centralized institutions. against this backdrop, institutional differentiation gives “coalitions of the willing” the opportunity to circumvent dysfunctional institutions by shifting the coordination of their crisis prevention or response actions to institutions with overlapping competences (jupille et al. 2013). these coalitions do not necessarily have to be composed of materially powerful countries. the theoretical logic applies as long as a “coalition of the willing” is able to act collectively in the absence of consensus within an established international institution. taken together, these two causal factors lead us to expect that, other things being equal, its institutional differentiation has enhanced the resilience of global governance. a new institution performing a new task increases the diversity of actions available to avoid or respond to a crisis; a new institution performing a task already covered by an existing institution functions as backup in case the existing institution is blocked or becomes dysfunctional. importantly, the process of institutional differentiation was not driven by states’ intention to enhance the resilience of global governance. thus, growing levels of institutional differentiation should not be conceived of as an investment in resilience. rather, the “law of requisite variety” suggests that growing levels of institutional differentiation primarily respond to growing levels of societal denationalization and transnational interdependence (ashby 1958; held et al. 1999). the resilience of global governance has therefore been enhanced as an unintended consequence of purposive state action (see merton 1936). it is well-known that some institutional configurations, such as markets, have selfgenerating mechanisms that serve to alleviate the impact of exogenous shocks in the absence of deliberate complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 46-54 doi: http://dx.doi.org/10.20377/cgn-99 52 university of bamberg press initiatives on the part of actors to invest in resilience (young 2009, 382). the present paper suggests that contemporary global governance is an institutional configuration which is in possession of these selfgenerating mechanisms. yet there is a valid intuition that the positive effect of increasing levels of institutional differentiation on the resilience of global governance must have some limits. increasing the number of institutions also leads to greater transaction costs of communicating and reaching a common response. it increases the risk of different groups of states engaging in multiple, uncoordinated, redundant, and potentially incompatible actions in different fora. for example, it will be difficult to respond effectively to a financial crisis if one international institution puts in place spending packages to stimulate the economy, while another institution pushes austerity programs. while it is good to have several options available, it is important that those options are used in a coordinated fashion. thus, there is a trade-off between the greater diversity of tools and the availability of backup options, on the one hand, and, the greater need for coordination resulting from institutional differentiation, on the other hand. while acknowledging a trade-off between differentiation and coherence, i conjecture that, as long as the number of institutions remains within certain reasonable limits, greater differentiation is likely to have positive net effects on resilience in the mediumand long-term. it is by now well-known that, over time, institutional complexity tends to induce inter-institutional coherence among international institutions with overlapping competences (abbott et al. 2015; gehring & faude 2014; pratt 2018; oberthür & stokke 2011). thus, a trade-off between a greater availability of actions and a greater risk from reduced coherence should exist only in the short term. in the medium and long term, i expect greater differentiation to have a positive net effect on resilience. conclusion this paper argues that the concept of resilience is a valuable addition to the toolbox of global governance scholars. at a time when global governance is characterized by multiple crises and turbulences, it facilitates an analysis of the ability of international institutions to handle exogenous shocks. such an analysis has so far been hampered by the lack of a suitable theoretical concept. to facilitate the development of a research agenda on the resilience of global governance, this paper applies central tenets of sociological differentiation theory to global governance. this theoretical approach offers a sense of how institutional complexity affects the resilience of global governance. it suggests that the process of institutional proliferation that has unfolded over the past decades has been enhancing the resilience of global governance, because it has allowed the “political system of world society” to mirror the increasing complexity of its social environment. contemporary global governance therefore provides states with a more diverse set of governance tools and it makes backup governance tools available. this argument contrasts starkly with prominent arguments in the literature on the general impact of institutional complexity on the problem-solving capacity of international institutions (cf. benvenisti & downs 2007; drezner 2013; hale et al. 2013a&b). importantly, however, states did not deliberately invest in resilience by increasing institutional complexity. rather, the increased resilience of global governance is an unintended result of purposive state action (see merton 1936). the general insight that institutional differentiation has enhanced the resilience of global governance paves the way for a research agenda that includes more specific questions, for instance, which design features make individual institutions more prone to resilience-enhancing inter-institutional coordination and whether the additional choice options available as a result of institutional complexity led to a more balanced institutional reflection of member states’ interests. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 46-54 doi: http://dx.doi.org/10.20377/cgn-99 53 university of bamberg press references abbott, kenneth w.; genschel, philipp; snidal, duncan; zangl, bernhard (eds.) (2015) international organizations as orchestrators. cambridge: cambridge university press. albert, mathias; buzan, barry; zürn, michael (2013): introduction. differentiation theory in international relations. in: mathias albert, barry buzan & michael zürn (eds.): bringing sociology to ir. world politics as differentiation theory. cambridge: cambridge university press. aligica, paul dragos; tarko, vlad (2014): institutional resilience and economic systems. lessons from elinor ostrom’s work. in: comp econ stud 56 (1), pp. 52–76. doi: 10.1057/ces.2013.29. alter, karen; raustiala, kal (2018): the rise of international regime complexity. in: annual review of law and social sciences 14, pp. 329-349. alter, karen; meunier, sophie (2009): the politics of international regime complexity. in: perspectives on politics 7 (1), pp. 1324. ashby, w. ross (1958): requisite variety and its implications for the control of complex systems. in: cybernetica 1. pp. 83-99. benvenisti, eyal & downs, george (2007): the empire’s new clothes: political economy and the fragmentation of international law. in: stanford law review 60 (2), pp. 595-632. buzan, barry; albert, mathias (2010): differentiation theory: a sociological approach to international relations theory. in: european journal of international relations 16 (3), pp. 315-337. collier, paul (2006): why the wto is deadlocked: and what can be done about it. in: the world economy 29 (10), pp. 1423– 1449. doi: 10.1111/j.1467-9701.2006.00844.x. drezner, daniel w. (2014): the system worked. global economic governance during the great recession. in: world politics 66 (01), pp. 123–164. doi: 10.1017/s0043887113000348. drezner, daniel w. (2013): the tragedy of the global institutional commons. in: martha finnemore & judith goldstein (eds.): back to basics: state power in a contemporary world. oxford: oxford university press, pp. 280–310. durkheim, emile (1968): über soziale arbeitsteilung. studie über die organisation höherer gesellschaften. frankfurt am main: suhrkamp. faude, benjamin; große-kreul, felix (2020): let’s justify! how regime complexes enhance the normative legitimacy of global governance. in: international studies quarterly 64 (2), pp. 431-439. faude, benjamin & parizek, michal (2020): contested multilateralism as credible signaling: how strategic inconsistency can induce cooperation among states. in: review of international organizations. online first. faude, benjamin; gehring, thomas (2017): regime complexes as governance systems. in: wayne sandholtz & christopher a. whytock (eds.): research handbook on the politics of international law. cheltenham: edwad elgar, pp. 176-203. gallopín, gilberto c. (2006): linkages between vulnerability, resilience, and adaptive capacity. in: global environmental change 16 (3), pp. 293–303. doi: 10.1016/j.gloenvcha.2006.02.004. gehring, thomas; faude, benjamin (2014): a theory of emerging order within institutional complexes: how competition among international regulatory institutions leads to institutional adaptation and division of labor. in: review of international organizations 9 (4), pp. 471-498. hale, thomas; held, david; young, kevin (2013a): gridlock: from self-reinforcing interdependence to second-order cooperation problems. in: global policy 4 (3), pp. 223–235. hale, thomas; held, david; young, kevin (2013b): gridlock: why global cooperation is failing when we need it most, cambridge: polity. hanrieder, tine (2015): international organization in time: fragmentation and reform. oxford: oxford university press. hasenclever, andreas; mayer, peter; rittberger, volker (1996): interests, power, knowledge: the study of international regimes. in: mershon international studies review 40, pp. 177-228. held, david; mcgrew, anthony; goldblatt, david; perraton, jonathan (1999): global transformations. cambridge: polity. henning, c. randall (2017): tangled governance. international regime complexity, the troika, and the euro crisis. oxford: oxford university press. herrfahrdt-pähle, elke; pahl-wostl, claudia (2012): continuity and change in social-ecological systems. the role of institutional resilience. in: e&s 17 (2). doi: 10.5751/es-04565-170208. hofmann, stéphanie c. (2011): why institutional overlap matters: csdp in the european security architecture. in: journal of common market studies 49 (1), pp. 101–120. holling, c. s. (1973): resilience and stability of ecological systems. in: annu. rev. ecol. syst. 4 (1), pp. 1–23. doi:10.1146/annurev.es.04.110173.000245. holling, c.s., gunderson, l.h. (2002): resilience and adaptive cycles. in: gunderson, l.h., holling, c.s. (eds.), panarchy: understanding transformation in human and natural systems. island press, washington, d.c., pp. 25–62. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 46-54 doi: http://dx.doi.org/10.20377/cgn-99 54 university of bamberg press jupille, joseph; mattli, walter; snidal, duncan (2013): institutional choice and global commerce. cambridge: cambridge university press. kahler, milis & david lake (eds.) (2013): politics in the new hard times. the great recession in comparative perspective. ithaca: cornell university press. kelley, judith (2009): the more the merrier? the effects of having multiple international election monitoring organizations. in: perspectives on politics 7 (1), pp. 59–64. keohane, robert o. (1984): after hegemony. cooperation and discord in the world political economy. princeton: princeton university press. keohane, robert o.; victor, david g. (2011): the regime complex for climate change. in: perspectives on politics 9(1), pp. 7– 23. doi: 10.1017/s1537592710004068. kindleberger, charles p. (1986): the world in depression, 1929-1939. rev. and enl. ed. berkeley: university of california press. koremenos, barbara; lipson, charles; snidal, duncan (2001): the rational design of international institutions. in: international organization 55 (4), pp. 761–799. doi: 10.1162/002081801317193592. kuyper, jonathan (2014): global democratization and international regime complexity. in: european journal of international relations 20 (3), pp. 620-646. mayntz, renate (1988): funktionelle teilsysteme in der theorie sozialer differenzierung, in: renate mayntz/bernd rosewitz/uwe schimank u.a. (eds.), differenzierung und verselbständigung zur entwicklung gesellschaftlicher teilsysteme, frankfurt a.m: campus, pp. 11–44. merton, robert k. (1936): the unanticipated consequences of purposive action. in: american sociological review 1(6), pp. 894904. nassehi, armin (2004): die theorie funktionaler differenzierung im horizont ihrer kritik. in: zeitschrift für soziologie 33 (2), pp. 98–118. oberthuer, sebastian & olav schram stokke (eds.) (2011): institutional interaction and global environmental change: interplay management and institutional complexes. cambridge: mit press. olson, mancur (1982): the rise and decline of nations. new haven, yale university press. orsini, amandine; morin, jean-frédéric; young, oran r. (2013): regime complexes: a buzz, a boom, or a boost for global governance? in: global governance 19 (1), pp. 27–39. pauly, louis w. (2008): the political economy of global financial crises. in: john ravenhill (ed.): global political economy. oxford: oxford university press, pp. 241–272. ponomarov, serhiy y.; holcomb, mary c. (2009): understanding the concept of supply chain resilience. in: internal journal of logistics management 20 (1), pp. 124–143. doi: 10.1108/09574090910954873. pratt, tyler (2018): deference and hierarchy in international regime complexes. in: international organization 72 (3), pp. 561590. steinberg, paul f. (2009): institutional resilience amid political change. the case of biodiversity conservation. in: global environmental politics 9 (3), pp. 61–81. doi: 10.1162/glep.2009.9.3.61. young, oran r. (1986): international regimes: toward a new theory of institutions. in: world politics 39 (1), pp. 104-122. young, oran r. (2009): institutional dynamics: resilience, vulnerability and adaptation in environmental and resource regimes. in: global environmental change 20, pp. 378-385. zürn, michael (1998): regieren jenseits des nationalstaates. globalisierung und denationalisierung als chance. frankfurt am main:suhrkamp. zürn, michael; buzan, barry; albert, mathias (2013): conclusion: differentiation theory and world politics. in: mathias albert, barry buzan; michael zürn (eds.): bringing sociology to international relations: world politics as differentiation theory. cambridge: cambridge university press, pp. 228–245. zürn, michael; faude, benjamin (2013): on fragmentation, differentiation, and coordination. in: global environmental politics 13 (3), pp. 119-130. building flood resilience in trans-boundary social-ecological systems: adaptive governance in the lake champlain richelieu river basin complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6580 doi: http://dx.doi.org/10.20377/cgn-81 65 university of bamberg press building flood resilience in trans-boundary social-ecological systems: adaptive governance in the lake champlain richelieu river basin author: emma spett college of agriculture and life sciences, university of vermont, u.s.a. e-mail: emma.spett@uvm.edu the management of water bodies that cross political, cultural, and ecological boundaries entails engaging with a level of complexity that requires creative, adaptive management strategies that build resilience throughout the system and contribute to increased capacity in the face of disturbance. to characterize the extent to which such complexity can be managed, this paper explores the application of the social-ecological systems framework, developed by elinor ostrom and operationalized by brian walker and david salt, for assessing and managing resilience. elements of this framework will be applied to the lake champlain richelieu river basin, which is a freshwater basin that exists between the united states and canada, in vermont, new york, and quebec. the paper will end with considerations regarding how adaptive management and governance can be employed as tools to build resilience in this region, particularly with respect to flooding and flood resilience. keywords: social-ecological systems; resilience; environmental planning introduction the lake champlain richelieu river basin spans physical, social, ecological, economic, and political boundaries. lake champlain’s southern edge is located in central vermont and new york, and beyond its upper edge in quebec, the lake drains north into the richelieu river. surrounding tributaries, and the towns built around them, make up the greater basin. the lake is fed from tributaries that flow from the adirondack mountains and green mountains in new york and vermont, respectively. governance of this region is varied in scale and capacity; each political region has distinctly organized agencies, municipalities, and structures, as well as organizations, universities, and individual actors who have a stake in the management of ecological, as well as economic and social resources. the challenge of reckoning the complex governance structure within the basin is a primary, initial hurdle of the region’s capacity to build resilience. http://dx.doi.org/10.20377/cgn-81 mailto:emma.spett@uvm.edu complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6580 doi: http://dx.doi.org/10.20377/cgn-81 66 university of bamberg press figure 1. map of the lake champlain richelieu river basin, retrieved from: international joint commission. (2018b). international lake champlain -richelieu river study board. retrieved from international joint commission website: https://www.ijc.org/en/lcrr the social ecological-systems framework was developed by elinor ostrom in order to consider the interactions among resources units, resource systems, governance systems, and users within a related ecosystem. the framework, ostrom posits, “is useful in providing a common set of potentially relevant variables and their subcomponents to use in the design of data collection instruments, the conduct of fieldwork, and the analysis of findings about the sustainability of complex social-ecological systems” (ostrom, 2009, p. 420). utilizing a resilience perspective to understand the dynamics of social-ecological systems is useful when considering how systems “persist through continuous development in the face of change and how to innovate and transform into new more desirable configurations” (folke, 2006, p. 260). cornerstones of social-ecological resilience include social learning and social memory, mental models and knowledge-system integration, visioning and scenario building (folke, 2006). brian walker and david salt lay out the operationalization of assessing and building resilience in their books “resilience thinking” and “resilience practice.” http://dx.doi.org/10.20377/cgn-81 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6580 doi: http://dx.doi.org/10.20377/cgn-81 67 university of bamberg press their framework seeks to categorize thresholds of the social-ecological system, defined as, “levels of controlling variables where feedbacks to the rest of the system change crossing points that have the potential to alter the future of many of the systems that we depend on” (walker & salt, 2012b, p. 53) as well as the thresholds of potential concern, which are yet to be established levels. the resilience thinking described by these authors provides a series of assumptions essential to the analysis of social-ecological systems. they include the fact that the systems are self-organizing, and that there are limits to a system’s self-organizing capacity. the systems have linked social, economic, and biophysical domains, move through adaptive cycles, and function across multiple scales (walker & salt, 2012). adaptive cycles are heuristic models that pass through four phases, two of which, growth and exploitation, are considered to be fast moving, a slow-moving conservation phase, and a chaotic collapse and release phase. panarchies are adaptive cycles operating at multiple scales, which is indicative of the complex nature of social-ecological systems (walker et al., 2004 walker and salt developed a framework that assesses a social-ecological system through three lenses; specified resilience, general resilience, and transformability. the interaction of these different system capacities indicates the resilience of the system. the framework to best understand these interacting components, considered within the context of the social-ecological system of the lake champlain richelieu river basin, can be found here: figure 2. specified resilience, general resilience, and transformability of the lake champlain richelieu river basin. http://dx.doi.org/10.20377/cgn-81 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6580 doi: http://dx.doi.org/10.20377/cgn-81 68 university of bamberg press this framework will provide the basis for an analysis that explores the specified resilience, general resilience, and transformability of the lake champlain richelieu river basin as a social-ecological system, and will consider how the trans-boundary nature of the embedded political and social systems fits into this framework. ultimately, the goal of this research is to assess the resilience of this system, propose tools for management, and consider how the framework can be adjusted to better accommodate systems with added complexity. all of these components will be considered through the lens of flood management, as flooding and fluvial erosion are the most destructive and prominent natural hazard to the region (vermont emergency management, 2018). 1. assessing resilience this section will categorize key components of the social and ecological thresholds that highlight the social, economic, and biophysical variables of interest when considering the resilience of the lake champlain richelieu river basin. 1.1 social thresholds the social system within the lake champlain richelieu river basin consists of communities of varying sizes. the largest municipality within the basin is st. jean-surrichelieu, which is a small city 40 km southeast of montreal, and, as its name suggests, is along the richelieu river. the population of the city, according to the 2011 canadian census, was 92, 394, with a median household income of $55,412 and 40,411 private dwellings (statistics canada, 2018). the second largest city, burlington, vermont, has a population of approximately 42, 417 people according to the 2010 census, with 16,851 households and a median income of $37,078 (united states census bureau, 2019). other large municipalities include whitehall, new york, plattsburgh, new york, and venise-enquebec, quebec in canada. a breakdown of demographic information can be found here: http://dx.doi.org/10.20377/cgn-81 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6580 doi: http://dx.doi.org/10.20377/cgn-81 69 university of bamberg press table 1. population of the lcrr basin state/province county/regional county/municipality total population density (people/square mile) new york clinton county, ny essex county, ny franklin county, ny warren county, ny washington county, ny 81.505 37,587 15,884 57,695 38,145 78.5 22.6 14.5 142.3 69.4 vermont addison county, vt chittenden county, vt franklin county, vt grand isle county, vt lamoille county, vt rutland county, vt washington county, vt 36,926 160,510 48,625 6,945 25,136 60,133 58,963 28.2 299.1 76.7 84.9 54.8 64.7 85.8 quebec brome-missisquoi, qc le haut-richelieu, qc pierre-de saurel, qc rouville, qc la vallee-durichelieu, qc 58,315 117,445 51,025 36,535 124,420 91.4 325.0 221.2 195.9 547.5 note. retrieved from: united states census bureau. "population." united states census bureau, 2019, www.census.gov/topics/population.html. http://dx.doi.org/10.20377/cgn-81 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6580 doi: http://dx.doi.org/10.20377/cgn-81 70 university of bamberg press table 2. income breakdown of us portion of lcrr basin state county $9,999 & less $10,000 -14,999 $15,000 -24,999 25,00034,999 $35,000 -49,999 $50,000 -64,999 $65,000 -74,999 $75,000 & above ny clinton essex franklin warren washington 19.13 17.21 19.32 15.98 17.54 8.94 8.72 9.05 9.88 8.63 13.23 15.64 14.21 15.35 17.09 11.69 13.57 12.18 12.06 12.29 11.97 13.57 14.38 12.67 13.24 8.98 8.91 10.47 11.38 9.12 3.40 3.87 3.04 4.28 2.98 7.47 7.86 7.62 10.91 5.76 vt addison chittenden franklin grand isle lamoille rutland washington 20.82 19.78 16.29 17.09 15.55 12.50 13.70 8.49 7.86 10.13 8.31 9.78 13.20 4.10 13.97 12.67 13.24 13.04 16.23 18.80 17.90 13.56 11.12 12.58 12.72 14.43 9.50 12.20 14.68 13.64 15.97 16.08 15.36 14.80 12.20 9.40 10.13 9.58 9.98 8.07 10.10 15.90 3.48 4.01 4.29 4.54 3.17 0.90 2.90 9.66 14.18 9.90 11.91 10.31 28.50 16.90 note. retrieved from: united states census bureau. "population." united states census bureau, 2019, www.census.gov/topics/population.html. http://dx.doi.org/10.20377/cgn-81 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6582 doi: http://dx.doi.org/10.20377/cgn-81 71 university of bamberg press table 3. income breakdown of quebec counties in lcrr basin regional county municipality under $10,000 $10,00019.9999 $20,00029,000 $30,00039,999 $40,00049,000 $50,00059,000 $60,00069,000 $70,00079,000 $80,00089,000 $90,00099,000 $100,000 and over brome missisquoi 11.09 18.54 16.48 14.09 11.42 0.82 5.41 3.61 2.29 1.55 4.88 le hautrichelieu 10.51 17.51 15.38 13.55 11.84 0.88 6.18 4.67 2.76 1.73 3.95 pierre-de saurel 11.85 20.37 16.05 13.46 10.82 0.72 4.96 4.49 2.82 1.62 3.53 rouville 9.61 16.08 15.04 14.46 13.63 0.97 6.82 4.56 2.51 1.64 3.45 la vallee-durichelieu 9.35 13.18 11.46 11.25 11.44 0.95 7.63 6.55 4.68 3.15 8.52 note. retrieved from: statistics canada. "census program." statistics canada, 2018, www12.statcan.gc.ca/census-recensement/index-eng.cfm. http://dx.doi.org/10.20377/cgn-81 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6882 doi: http://dx.doi.org/10.20377/cgn-81 72 university of bamberg press population and income distribution are important social thresholds, since both indicate vulnerability, particularly to flooding. regions with high population density, coupled with proximity to floodplains, can increase the potential for regime shift following disaster. income is indicative of a vulnerability that reduces the resilience of individuals, households, and communities, and makes them more susceptible to the impacts of flooding. low-lying homes within the floodplains of the richelieu river and lake champlain were inundated with water following spring flooding in april and may of 2011. many homes, primarily in quebec, were evacuated, and some were destroyed. in may, when the flood stage reached its maximal water level, the military of canada was called in to assist in the evacuation of civilians in the town of st. jean-sur-richelieu. in vermont, 500 individuals were evacuated and 500 homes were impacted by lake champlain and subsequent tributary floods. 124 individuals were evacuated in new york, with 929 homes affected. quebec bore the brunt of the damage, though, with 1,651 individuals evacuated during the length of the flooding event, and 2,375 homes affected. known thresholds within the social realm, then, are the damages brought on by the flood stage moving beyond 100 feet and affecting the low-lying homes (federal emergency management agency 2018a; federal emergency management agency 2018b). in order to establish a list of prominent flooding events in the lake champlain richelieu river basin, an analysis of disaster declarations from the federal emergency management agency (fema) was conducted. disaster declarations were considered if they occurred in the new york counties: clinton; essex; franklin; warren; and washington. in vermont, declarations of disaster were considered if they occurred in the following counties: addison; chittenden; franklin; grand isle; lamoille; rutland; and washington. in vermont, fema disaster declaration data indicates the occurrences of 25 distinct flooding events on the vermont side of the lake champlain richelieu river basin, beginning in 1973. each flooding event is also associated with a cause. four events are associated with ice jams and occurred between december and january, ten events are associated with snowmelt and occur between march and june, and eleven flooding events occurred between july and september and were associated with heavy rain (federal emergency management agency, 2018b). across the lake in new york, fema disaster declaration data indicates the occurrences of 12 distinct flooding events, beginning in 1976. two events occurred in january, six occurred between march and june (caused primarily by snowmelt), three events occurred between july and september (caused by heavy rainfall), and one event occurred in november (federal emergency management agency, 2018a). in both vermont and new york, there are no more than nine disasters declared in a given decade, with vermont declaring nine disasters in the 1990s, and new york declaring five. new york and vermont also possessed very little overlap with respect to disaster declarations, indicating disproportionate effects of snowmelt, ice jams, and heavy rainfall on opposite sides of lake champlain. shared disasters include the spring and fall floods of 2011, as well as june 2013, september 1999, june 1998, january 1998, january 1996, and july/august of 1976 (federal emergency management agency, 2018). to the north, in quebec, public safety canada’s canadian disaster database (cdd) has recorded 40 flood events in quebec, occurring between 1928 and 2014. floods of consequence that encompassed the entire lake champlain richelieu river basin include both the spring and summer floods of 2011, and august of 1976. flooding most likely also occurred during events in vermont and new york, but was not categorized as a disaster. the canadian disaster database considers an event a disaster if 10 or more people are killed, 100 or more people are affected, injured, infected, evacuated, or homeless, if there is an appeal made for national or international assistance, if it proves to be historically significant, or if there is significant damage or interruption of normal processes such that the community affected cannot recover on its own (public safety canada, 2018). http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6882 doi: http://dx.doi.org/10.20377/cgn-81 73 university of bamberg press additional thresholds of interest within the realm of social and community life in the lake champlain richelieu river basin include the implications of flooding on agriculture and other economic ventures. there is little data to express the financial and social burdens of flooding, which would make these categories of potential concern. media sources did note that the overload of nutrients from lake flooding affected the soil quality and therefore the capacity of farms to produce at a normal rate. the tourism and recreation industries in burlington, vt and venise-en-quebec suffered during the floods, although the economic cost of this degradation is not well documented. the interconnected nature of the social lives of inhabitants along the lake champlain richelieu river basin highlight the need for a resilience management approach that considers the widespread effects of flooding on every aspect of life in the region. 1.2 ecological thresholds the lake champlain richelieu river flooding event of 2011 was deemed a hundred-year flood, meaning the region had not experienced an event of a similar magnitude in recent or recorded memory. prior to 2011, the highest recorded level that the lake had reached was 99.64 feet. on may 23rd, 2011, the maximal water level reached 103.5 feet. typically, projections point to any floodwater beyond 100 feet as a serious risk to the low-lying communities and ecosystems (lake champlain basin program, 2019). while the flood levels had serious implications for the social system along the basin, it also had an adverse effect on a variety of ecological factors. phosphorous and other nutrient loading accounts for the over-fertilization and subsequent water quality issues associated with lake champlain. sources of phosphorous in the lake champlain richelieu river basin include point and nonpoint sources, which indicate whether or not the runoff has come from a pipe. non-point phosphorous sources in the region, which account for over 90% of the lake’s phosphorous load, include agricultural land, urban and developed land, and forested land. the implications of phosphorous loading include harmful algal blooms, which can adversely affect animals and plants, and can affect the use of the lake by the fishing and recreation industries (lake champlain basin program, 2019). the established threshold for phosphorous in lake champlain varies based on a variety of factors including depth and land use, but by and large the standard that phosphorous should not exceed is about 10 micrograms/liter. during the flooding events of 2011, washout caused a spike in non-point source phosphorous loading, exacerbating the negative water quality of the lake (lake champlain basin program, 2019). additional thresholds of interest that can indicate vulnerability include the fish stock of the lake, the effects that flooding have on biodiversity and habitat, the introduction and prevalence of invasive species, and the overall state of ecosystem health with respect to how it handles development and disturbance. 1.3 adaptive governance the role that governance structures play in the establishment and retention of resilience of social-ecological systems is an imperative component of a resilience assessment. to determine the state of a governance structure’s resilience first requires the acknowledgement of how “even though some similarities can be identified, societies and ecosystems are also fundamentally different in many ways” and an awareness that “the notions of resilience…may have wholly different meanings depending on how the social system was configured to begin with and from whose viewpoint the resilience is assessed” (duit et al., 2010, p.365). http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6882 doi: http://dx.doi.org/10.20377/cgn-81 74 university of bamberg press the response to this variation is a greater consideration of complexity. ultimately, the need for new governance models to improve society’s ability to handle processes of complex change are necessary, though difficult to develop and even more difficult to implement (duit et al., 2010). one such answer to the problem of integrating adaptive governance into social-ecological systems resilience comes from karpouzoglou et al. (2016), who conducted a literature review on the subject of adaptive governance and capacity building. the major findings from their study indicate that there must be a “focus on governance capacity to adapt to change and self-organize,” and a “collective search for solutions to societal problems.” they also emphasized the need for “indigenous and scientific knowledge integration, learning by doing, and policy and management as experiments.” finally, they call for “mixed hierarchies and matching ecosystem management with the appropriate governance scale” (karpouzoglou et al., 2016, p.5). in an applied sense, these findings require mainstreaming a social-ecological systems perspective. fischer et al. (2014) advise prioritizing attention paid to long-term drivers that gradually shape social-ecological systems, a commitment by governments and society at large to support the development of a stronger science-society interface, and the interactions among power relations, equity, justice, and ecosystem stewardship to be better understood (fischer et al., 2014). 2. managing resilience the ultimate objective of undertaking a resilience assessment is to yield a resilient management approach, which, bound to the development of adaptive management policy, requires an understanding of where many potential management scenarios fit within the adaptive cycle, as well as when management structures would be most useful. this section of the resilience assessment will provide a series of tools and options for management, as well as a conception of how such tools can be integrated into an adaptive management plan for the lake champlain richelieu river basin. one specific consideration for addressing flooding in particular in the lake champlain richelieu river basin is to make sure that any tools utilized can be shared as a resource across the state and national borders that comprise the basin, which will require increased levels of collaboration and communication on the part of practitioners, and will need to be cognizant of the diverse, multi-faceted governance that exist throughout the lake champlain richelieu river basin. literature on trans-boundary water management will be analyzed, and then contextualized within the scope of the watershed of interest. 2.1 tools and options for management at present, some of the management options being utilized to deal with flooding on the transboundary lake champlain richelieu river basin include localized efforts to mitigate the impact of flooding, including sandbagging around neighborhoods and roads that are particularly vulnerable to damage. a current flood mitigation program is currently being developed by the international joint commission, a trans-boundary, bi-national water management agency, which will provide insight into management proposals, as well as a ground for criticism of where the proposal could further develop protocols for building resilience in the social-ecological system. when considering tools and options for management that acknowledge the prevalent, known thresholds previously described, there are a variety of measures that can be taken in order to build up resilience programs: management, financial interventions, governance, and education. management of resources, specifically in the trans-boundary lake champlain richelieu river basin, develops complications with respect to the conflicting nature of having multiple states and countries vested in a certain systematic mode of operating. in quebec, there is a community-wide http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6882 doi: http://dx.doi.org/10.20377/cgn-81 75 university of bamberg press desire to mitigate flooding in order to protect riverside homes and businesses. across the border, where flooding causes less of an immediate problem to the lakeside communities in vermont and new york, there is a greater vested interest in maintaining the water quality, since there is a substantial tourism and recreation industry. were any type of structural program implemented in order to control water flow into quebec, there would likely be a water quality implication upstream in lake champlain. so how is such a conflict mediated in order to develop a resilience management strategy that is satisfactory to the continued economic, social, and environmental success of all facets of the region? mark zeitoun and naho mirumachi describe the challenges associated with treaties meant to establish cooperation between entities, such as the boundary water treaty that created the international joint commission. they write that, “international agreements are generally seen as the pinnacle of cooperation…and emphasize ‘once cooperative water regimes are established through treaties, they turn out to be impressively resilient over time” (zeitoun and mirumachi, 2008, p.303) and acknowledge that there must be work done to “move beyond the notion of cooperation as treaties to a more dynamic view of trans-boundary water cooperation as an on-going and non-linear process in which state and non-state actors establish, challenge, modify, and legitimize multilayered governance structures” (zeitoun and mirumachi, 2008, p.303). best management practices for the lake champlain richelieu river basin, while not explicitly categorized as an area of conflict like many of the trans-boundary water systems described in the previously described article, can benefit from this shifting management paradigm. jens newig and oliver fritsch note that their analysis, “suggests that a highly polycentric governance system comprising many agencies and levels of governance yields higher environmental outputs” (newig and fritsch, 2009, p.197). in the lake champlain richelieu river basin, it is important to draw in the various ngos who direct their attention towards flood mitigation, water quality, and other issues at the nexus of socialecological issues. in vermont, the lake champlain basin program, lake champlain international, and 350vt have vested interest in the maintenance of lake champlain’s ecosystem health. in new york, the organizations that direct attention towards adirondack state park play a substantial role in the conservation of the area’s natural resources, and in canada, ngos tend to direct their attention towards mitigation of natural hazards. for example, following the 2011 flooding event in quebec, a car salesman named michel fecteau developed the organization sos richelieu, which grew out of a desire to active local social networks to aid in disaster relief. fecteau was later elected mayor of st. jean-sur-richelieu, which is one of the most effected flooding locations in the lake champlain richelieu river basin (rukavina, 2017). along with ngos, there must be an integration of local municipalities and governments, including planning boards, government agencies, and community groups. additionally, there should be an integration of universities, and space for public input. in the current international joint commission iteration, there is substantial input from professionals from local universities and government agencies, as well as representation from larger national organizations like the united states geological survey and environment canada, but little input from groups with localized information and investment. primarily, interaction with local stakeholders is facilitated through engagement with study membership. there is also a challenge that has to be mediated in order to effectively manage the region in a cooperative way, which is the social disconnect between the united states and canada. there is a lack of understanding amongst americans in the lake champlain richelieu river basin, particularly in the more heavily populated region in vermont, about the implications that flooding has on the canadians along the richelieu river. following a media review of the flooding event in the region, results showed that there were substantial news stories on the canadian news services regarding the health, economic impacts, and environmental degradation associated with http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6882 doi: http://dx.doi.org/10.20377/cgn-81 76 university of bamberg press the floods, while the american news services only briefly covered the fairly limited damage that the flooding caused in burlington, vermont. a useful strategy for building the resilience of the basin is the creation of a diversified educational program that seeks to build a sense of empathy amongst the basin residents that do not experience the worst effects of flooding. there are a variety of educational strategies that could be facilitated by the governance structures in vermont and new york that engage with building the resilience of the region to flooding, including the government departments involved, including the department of environmental conservation, along with university actors involved with the project. for the immediate present, the development of story-telling educational programs for engaged citizens would provide the opportunity for the community to develop a stake in the larger wellbeing of the entire lake champlain richelieu river basin. over a longer period of time, the integration of basin-wide, place-based education programming would be an effective tool for perpetuating the continued conversations around management of the region, and ultimately contribute to a substantially increased resilience due to raised awareness. place-based education has the capacity to, “help students develop stronger ties to their community, enhances students’ appreciation for the natural world, and creates heightened commitment to serving as active, contributing citizens” (sobel, 2004, p.6). funding for these programs should be diversified and represent the widespread investment of varied stakeholders, including local and national government, local and national organizations, municipalities, and universities. similar to how management programming is developed via a collective of scientists and professionals from the international joint commission, funding could be controlled through a similar avenue. 2.2 adaptive management when considering a framework for implementing the strategies and tools outlined above, with the shared goal of management and resilience, adaptive management and adaptive governance, respectively, are useful tools for integrating programs. a framework for adaptive management is laid out by byron k. williams, in which he addresses the necessary state a system must be in to benefit from adaptive management, as well as components of an adaptive management program. to begin, the system must fulfill a set of criteria. the system must be “dynamic, changing through time in response to environmental conditions and management,” there must be “environmental variation [that is] only partially predictable,” and the system, “is subjected to periodic and potential management interventions that potentially vary over time.” finally, “effective management is limited by uncertainty about the nature of resource processes and the influence of management on them” (williams, 2011, p.1348). this accurately describes the lake champlain richelieu river system. from there, the framework calls for two phases, both of which involve the critical process of learning in order to further develop the system’s capacity to adapt to changes, which will be significant in the coming years, when climate change increased unpredictable patterns of ecosystem function, weather, development, and more. to effectively implement an adaptive management program, williams notes that there must be a commitment to learning from the past and moving towards change. he explains how, in a management program that seeks to be adaptive, “little attention is given to the institutional barriers to its implementation, and little effort is expended on redesigning organizational structures and processes to accommodate an adaptive style of management” (williams, 2011, p.1352). there would need to be, at the very least, an acknowledgement by project leadership that there needs to be a commitment to often reconsidering the actors involved in the program, and any potential new programs that could be integrated that would represent new perspectives. this is exemplified by the resilience-focused management of the trans-boundary columbia river basin. the columbia http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6882 doi: http://dx.doi.org/10.20377/cgn-81 77 university of bamberg press river treaty was created in 1964 to allow for the co-management of the columbia river’s dams. the treaty’s stakeholders have maintained dedication to integrated changing values and uses, as well as new concerns like climate change into the management of the basin, which was made evident by the columbia river symposium of 2009, when all such concerns were addressed (international joint commission, 2018). another form of this type of programming is adaptive governance, which acknowledges that, “a society’s ability to manage resilience resides in actors, social networks, and institutions” (lebel et al., 2006, p.6). some of the recommendations laid out in lebel et al. (2006) include participation and deliberation, polycentric and multi-layered institutions, and accountable and just authorities. in essence, the management programming is secondary to the ability to the individuals and groups in charge of managing the system in question to be adaptable, progressive, and understanding of what is required to maintain a cohesive understanding of the social-ecological system (lebel et al., 2006). going forward, there are a variety of tools that contextualize the thresholds within the scope of adaptive management planning. a framework that would be best suited to implementing programming that is grounded in resilience thinking while also addressing thresholds of concern is proposed by ryan plummer and derek armitage, who developed a resilience-based framework for evaluating adaptive co-management. their framework notes that adaptive co-management requires consistent evaluation of performance and outcomes, as well as foundations for multi-site comparisons. this, paired with understanding of the processes, livelihoods, and ecological systems, and the way that those components are embedded, contribute to social-ecological resilience and sustainability. ultimately, the framework calls for the following means of evaluation of an adaptive comanagement schematic. first, they call for the identification of ultimate parameters of concern. in their case, enhanced livelihoods and ecological sustainability, and in the case of the lake champlain richelieu river basin, perhaps the added component of cross-border interactions and public engagement would also be included. next, first order, second order, and third order parameters are established. this allows for prioritization of programming, with considerations of what is tangible and intangible. the use of a tool like this would mediate issues of trans-boundary disconnect and challenges related to differing values, since the achievement of overcoming those issues could be integrated into the framework (plummer and armitage, 2007) 3. governance and resilience in the lcrr basin in lebel et al. (2006), their central question is, “how do certain attributes of governance function in society to enhance the capacity to manage resilience?” (lebel et al., 2006). they define governance as, “the structures and processes by which societies share power, shapes individual and collective action…it includes laws, regulations, discursive debates, negotiation, mediation, conflict, resolution, elections…and other decision-making processes” (lebel et al., 2006, p.4). alternatively, brondizio et al. (2009) writes that governance systems are, “construed as a form of social capital that communities establish and rely on to guide human-environment interactions in a variety of settings” (brondizio et al., 2009). when considering how governance systems can contribute to the development of resilience, there are certain attributes that are considered to be particularly impactful to the management of resilience. lebel et al. (2006) highlight public participation, polycentric institutions, and accountability as major components of managing resilience. public participation, specifically through the lens of deliberation, lends to the alignment of priorities between citizens and scientists, experts, and decision-makers (lebel et al., 2006). polycentric institutions also play an important role in the maintenance of resilience. they are described as, “arrangements that are nested, quasihttp://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6882 doi: http://dx.doi.org/10.20377/cgn-81 78 university of bamberg press autonomous decision-making unites operating at multiple scales…spanning from local to higher organizational levels, polycentric institutions provide a balance between decentralized and centralized control” (olsson et al., 2006, p.2). finally, accountability lends to the maintenance of resilience through mechanisms such as, “transparency, independent monitoring, polycentricity, separation of powers, legal resources” and have social justice as a goal (lebel et al., 2006). garmestani and benson explore some other means of achieving and monitoring resilience, with specific attention paid to panarchy as a tool for characterizing the “cross scale dynamics of social-ecological systems and a framework for how governance institutions should behave to be compatible with the ecosystems they manage” (garmestani and benson, 2013, p.1). through their framework for resilience-based governance, garmestani and benson explore panarchy’s role in adaptive governance, noting that the panarchy model can be used to, “reconceptualize social-ecological systems in a manner that has the capacity to better match governance to the environment” (garmestani and benson, 2013, p.6). while the tools laid out as frameworks for integrating governance and resilience are often made through theoretical frameworks and conceptualizations, there are issues with the translation of these ideas into practice. olsson et al. (2015) note that there are types of resilience: bounce back, bounce back and transform, and that each type has a descriptive and prescriptive capacity. they also critique how resilience theory “suggests that ‘critical changes in social-ecological systems are determined by a small set of three to five key variables’” and that a better means of creating strong foundations for resilience are to “search for integrative theories that combine disciplinary strengths while filling disciplinary gaps” (olsson et al., 2015, p.8). 4. discussion and conclusions beyond the traditional aspects of assessing resilience, there is further information that can be gathered from exploring the more intricately developed social networks and phenomenon that exist within the lake champlain richelieu river basin. for instance, there is value to quantifying the relationships that exist within communities, and whether or not those ties can be utilized to build resilience. additionally, the idea of activating social memory, or restoring it, can be useful in galvanizing support for issues related to hazard mitigation and increased resilience. for instance, in the case of the 2011 flooding of the lake champlain richelieu river basin, the experience varied across borders; quebecois experienced weeks of devastating floods, ruined homes, monetary loss, and implications for mental and physical health. to the south, while flooding affected communities along the lake, there was substantially less damage. tropical storm irene’s impact just four months after the spring floods is far more prominent in the memories of vermonters and new yorkers. so, how is such an event standardized to a non-traditional community, as with the lake champlain richelieu river basin? further research is required to explore perceptions of flooding and vulnerability across the basin, which has the potential to yield public education campaigns, programming, and ultimately a widened perspective. additionally, this region represents a potential frontier for research into integrated, adaptive management of trans-boundary social-ecological systems and their subsequent resilience. the exercise of attempting to understanding the nuances of complex social-ecological systems on a community scale comes to a head at the lake champlain richelieu river basin. in this space, there is a dispute that has the privilege of cooperation on multiple governance scales as well as support through the international joint commission to explore how to utilize governance and community structures to better understand and build on current knowledge and perceptions. in many ways, the work currently being done to enhance the resilience, particularly to flooding, of the lake champlain richelieu river basin can provide insight into how to utilize public input to develop the solutions, and how to active a sense of community around natural rather than social and political boundaries. this initial investigation into assessing and managing resilience is only http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6882 doi: http://dx.doi.org/10.20377/cgn-81 79 university of bamberg press the first look into how creative, dynamic frameworks can be applied to this region to ultimately develop its capacity to adapt and exist in a sustainable way. references brondizio, e. s., ostrom, e., & young, o. r. (2009). connectivity and the governance of multilevel social-ecological systems: the role of social capital. annual review of environment and resources, 34, 253-278. duit, a., galaz, v., eckerberg, k., & ebbesson, j. (2010). governance, complexity, and resilience. federal emergency management agency. (2018a). new york severe storms, flooding, tornadoes, and straight-line wind (dr-1993). retrieved october 30, 2018, from fema.gov website: https://www.fema.gov/disaster/1993 federal emergency management agency. (2018b). vermont severe storms and flooding (dr-1995). retrieved october 30, 2018, from fema.gov website: https://www.fema.gov/disaster/1995 fischer, j. et al., (2014). advancing sustainability through mainstreaming a social–ecological systems perspective. current opinion in environmental sustainability, 14. folke, c. (2006). resilience: the emergence of a perspective for social–ecological systems analyses. global environmental change, 16(3), 253-267. international joint commission. “guiding principles.” international joint commission, 2018, http://www.ijc.org/en_/guiding_principles international joint commission. (2018b). international lake champlain -richelieu river study board. retrieved from international joint commission website: https://www.ijc.org/en/lcrr karpouzoglou, t., dewulf, a., & clark, j. (2016). advancing adaptive governance of social-ecological systems through theoretical multiplicity. environmental science & policy, 57, 1-9. lake champlain basin program. "flooding." lake champlain basin program, 2019, www.lcbp.org/waterenvironment/water-quality/flooding/. lebel, l., anderies, j. m., campbell, b., folke, c., hatfield-dodds, s., hughes, t. p., & wilson, j. (2006). governance and the capacity to manage resilience in regional social-ecological systems. newig, j., & fritsch, o. (2009). environmental governance: participatory, multi‐level–and effective?. environmental policy and governance, 19(3), 197-214. olsson, l., jerneck, a., thoren, h., persson, j., & o'byrne, d. (2015). why resilience is unappealing to social science: theoretical and empirical investigations of the scientific use of resilience. science advances, 1(4). olsson, p., l. h. gunderson, s. r. carpenter, p. ryan, l. lebel, c. folke, and c. s. holling. (2006). shooting the rapids: navigating transitions to adaptive governance of social-ecological systems. ecology and society 11(1): 18. url: http://www.ecologyandsociety.org/vol11/iss1/art18/ ostrom, e. (2009). a general framework for analyzing sustainability of social-ecological systems. science, 325(5939), 419422. plummer, r., & armitage, d. (2007). a resilience-based framework for evaluating adaptive co-management: linking ecology, economics and society in a complex world. ecological economics, 61(1), 62-74. public safety canada. (2018). canada disaster database. retrieved from public safety canada website:http://cdd.publicsafety.gc.ca resilience alliance. "key concepts." resilience alliance, 2019, www.resalliance.org/key-concepts. http://dx.doi.org/ https://www.fema.gov/disaster/1995 http://www.ijc.org/en_/guiding_principles http://www.ecologyandsociety.org/vol11/iss1/art18/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 6882 doi: http://dx.doi.org/10.20377/cgn-81 80 university of bamberg press rukavina, s. (2017, may 10). mayor of st. jean-sur-richelieu, aka "the bulldozer", shares flood wisdom. retrieved from cbc website: https://www.cbc.ca/news/canada/montreal/mayor-of-saint-jean-sur-richelieu-aka-the-bulldozer-sharesflood-wisdom-1.4108375 sobel, d. (2004). place-based education: connecting classroom and community. nature and listening, 4(1), 1-7. statistics canada. "census program." statistics canada, 2018, www12.statcan.gc.ca/census-recensement/index-eng.cfm. united states census bureau. "population." united states census bureau, 2019, www.census.gov/topics/population.html. walker, b., holling, c. s., carpenter, s., & kinzig, a. (2004). resilience, adaptability and transformability in social– ecological systems. ecology and society, 9(2). walker, b., & salt, d. (2012a). resilience practice: building capacity to absorb disturbance and maintain function. island press. walker, b., & salt, d. (2012b). resilience thinking: sustaining ecosystems and people in a changing world. island press. williams, b. k. (2011). adaptive management of natural resources—framework and issues. journal of environmental management, 92(5), 1346-1353. vermont emergency management. state hazard mitigation plan. 17 nov. 2018. zeitoun, m., & mirumachi, n. (2008). transboundary water interaction i: reconsidering conflict and cooperation. international environmental agreements: politics, law and economics, 8(4), 297. http://dx.doi.org/ 1. assessing resilience 1.1 social thresholds 1.2 ecological thresholds 1.3 adaptive governance 2. managing resilience 2.1 tools and options for management 2.2 adaptive management 3. governance and resilience in the lcrr basin 4. discussion and conclusions complexity and its implications in advanced leadership development complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 59 university of bamberg press complexity and its implications in advanced leadership development james k. hazy abstract: this article describes an advanced leadership seminar for senior managers and executives. the three module seminar describes how thoughtful leaders can learn to succeed even under conditions of extreme complexity. the first module describes complexity in the scientific sense and highlights the need to allow events to develop so that emergent patterns can be identified. the second module shows how this new complexity perspective can be used to frame and communicate social and economic imperatives in ways that move the organization forward. the third module puts this knowledge to work and enables participants to be better stewards of collective leadership within their organizations. some reflections on the success of the program and potential improvements are offered in the concluding section. keywords: complexity leadership; collective leadership; leadership development introduction this article describes a complexity-inspired advanced leadership seminar that was given to senior managers and executives at a multi-billion dollar healthcare system in the us. the program was custom designed to offer participants an opportunity for a transformative learning experience (mezirow, 1991) by offering an alternative view of leadership that can succeed in today’s rapidly changing, complex and globally connected political economy. the program assumes that participants already have significant leadership experience and deep academic knowledge of leadership studies. the article begins with a discussion of what leaders should know. this is followed by an overview of the three modules of the program. it ends with some thoughts on how the program can be improved going forward. throughout, the term “complex” is taken in the technical sense wherein many interacting events and multiple levels of analysis are being confronted simultaneously with a changing mix of uncertainty and predictability. to succeed in a given situation, a leader must understand what is happening, accept uncertainty and still take appropriate action while leading others who are likewise experiencing uncertain events. this seminar presents complexity leadership as an emerging imperative for today’s leaders. james k. hazy, adelphi university, 1 south avenue, garden city, ny 11530, hazy@adelphi.edu http://dx.doi.org/10.20377/cgn-55 mailto:hazy@adelphi.edu complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 60 university of bamberg press what leaders should know about complexity the suggested approach involves a transformative journey. the facilitator guides the participants in a deepening discourse (mezirow, 2003) about the challenges that emerge from complexity, a conversation that is further enhanced if the facilitator is able to share stories about the challenges and insights that occurred along his or her own personal journey into and through an acceptance of a world that is chock full of complexity. the journey involves three modules that together build a foundation from which participants can successfully lead their organizations into the middle of the 21st century. the three modules are intended to clarify 1) how the term “complexity” in this science differs from common use of the term “complex,” 2) how these new ideas inform life in today’s complex organizations, and 3) how individuals can use these ideas to become more successful as leaders. during each of these modules, the instructor can go into more or less detail on specific topics depending on the nature of one’s own journey, the length of the session, and how deeply the program manager wants to explore the subject matter. (table 1 presents a list of possible topics.) in addition, from a pedagogical perspective, each session features limited lecture time that is intended to present material for exploration and individualized discovery and includes more or less class discussion and various project-oriented in situ exercises. in the healthcare context, specific work situations currently being experienced by the participants are explored using this new perspective to enable participants to internalize what is being learned and apply these ideas in real time to their work. programmatically, each module begins with a video intended to demonstrate some important and subtle aspect of complexity that will be explored metaphorically throughout the module. the first module begins the program by immediately addressing the issue of complexity as a discipline and how this usage differs from how the term is used in common language (anderson, 1999; marion & uhl-bien, 2001; uhl-bien & marion, 2009; goldstein, hazy & lichtenstein, 2010; lichtenstein, 2014). the second module explores what complexity means for organizations. the third takes a new look at “leadership” as a system or collective phenomenon (ospina and foldy, 2015; yammarino et al., 2012) in the context of complexity (hazy & uhl-bien, 2014). module 1: clarify what complexity really means the study of complexity science has made considerable progress over the last half a century. at the conceptual level one can think about the study of complexity as an attempt to understand and model phenomena and interaction effects that do not behave linearly. this means that increasing an input, like nursing labor hours, does not necessarily drive a proportional increase in output, such as patient satisfaction. in complex organizations, events do not always drive proportional effects. this module offers participants a fresh view of how the physical and social world works when complexity findings are incorporated into one’s worldview. its key lessons are: first, a complex system is one where considerable uncertainty still remains in the system. although a level of predictability is present in complex systems, there is also potential for surprise which can have broad implications. complexity science builds models, but these models are not the same as the system. the system can surprise even if the model being used does not provide a warning. second, patterns and observed stability can be modeled, and this can be used to predict outcomes, but only to a degree (arthur, 2015). states of localized dynamic stability operate in different path-dependent regimes, and the boundaries between regimes can be discontinuous http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 61 university of bamberg press (hirsch, smale & devaney, 2004; mandelbrot, 1982). this means that change is often sudden, even precipitous. it is important for practitioners to know and to expect that when the dynamic stability of a “system” approaches a boundary between stable regions, its stability may be fragile. third, when an organization is in a dynamically stable regime it might suddenly “flip” to a different dynamically stable regime due to pressure or stress coming from outside the boundaries of the organizing system. furthermore, when the organizing system assumes its new state, it doesn’t necessarily behave the way that an observer might hope or expect given the purpose of the organization. the trick is to learn to recognize what is driving these dynamic changes and determine how to weaken forces pushing in the wrong direction and reinforce others pushing toward a new state that is more efficacious. beauty and surprise although the study of complexity is ultimately a scientific discipline, one of the best ways for managers and professionals to understand and appreciate complexity is to experience its beauty. one way for them to see this beauty is to let them experience and discover it for themselves. this is why it is useful to show a short video before each module. to begin the exploration of the deep meaning of complexity in this first module, a video journey into a mandelbrot (1982) zoom is useful for setting the stage for a transformative experience (mezirow,1991). showing it serves three purposes as the participants’ begin their journey into complexity. first, the mandelbrot zoom highlights self-similarity (hirsch, smale & devaney, 2004) as well as the depth and sweeping breadth of complexity. here, the instructor would prompt the participants by suggesting that what they are experiencing emotionally and cognitively during this “zoom” is similar to what is happening to each individual in their workgroups, their department, their organization and their entire ecosystem. each individual can explore deeply into the system and find the same and different patterns appearing again and again. one job of a leader is to find resonances within this complexity, a leadership imperative that goldstein, hazy and lichtenstein (2010) call “interaction resonance.” second, this video and what it represents highlights for the viewer the fragile and improbable nature of dynamic stability in a complex world. it demonstrates for participants that in the pattern itself, each of its buds and bulbs, reflects a distinct territory of comfortable predictability, some signifying increased complexity followed by others which represent surprising simplicity. ultimately, when an individual, work unit, or an organization ventures outside of this beautiful complex shape, complete dissolution of order, a “strangeness” that is sometimes called “chaos,” is the result. another job of the leader therefore is to guide others to enable the collective to stay in a place where one can recognize and appreciate the beauty of order and organization even as events remain complex. third, this module offers a bridge to the modules that follow by offering a hint at the complexity of “leading” in such a dynamic and “strange” world. complexity helps the participants understand why leadership is so difficult and suggests the unlikely potential for success, unless of course, one is lucky enough to begin in a predictable place and has the wisdom to find a way to allow the organization to prosper by staying there. often, this involves finding the people who are skilled at engaging complexity and putting them in the right roles as hazy (2012) demonstrated using computer modeling. during the session, each of these themes is developed through reflection, discussion and group exercises rather than lecture. http://dx.doi.org/10.20377/cgn-55 62 university of bamberg press complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 models find simplicity on the other side of complexity after the opening video, this session begins by making a rather counterintuitive assertion: engaging complexity actually means reimagining simplicity. the challenge for the observer is to find and recognize a new simplicity that emerges on the other side of complexity. oliver wendell holmes is often quoted as saying: “i would not give a fig for simplicity on this side of complexity, but i would give my right arm for the simplicity on the other side of complexity.” this is relevant for managers because having a better model – where “better” means that it predicts what can be predicted and also accepts what cannot – to predict events in one’s organization helps a leader be more effective at his or her job. it also helps one better lead a large organization as it moves from one state of dynamic stability to another within the organization’s own “mandelbrot set” (mandelbrot, 1982). it is critical for a manager to realize that finding and understanding new simplicity in their organization takes time. “waiting for the smoke to clear” can be difficult, particularly when events are complex and uncertain and others are looking for direction. however, failure to understand the changing dynamics of stability can mean carelessly stepping across the line into areas of unbridled divergence: dissolution, decay or implosion. patience and thoughtful vigilance are as essential as timely action. sometimes a small incremental step that is intended to understand what is happening, what is called generative leadership (surie and hazy, 2006), is the right choice. one then iterates the process again. one of the key discoveries of complexity science is that in very complex environments, even ones that are apparently unpredictable in the details, often the observed outcomes that are most important continue to exhibit a level of predictability. one simply has to peer patiently through the complexity of the moment to see the emergent simplicity. the capacity to focus on what is relevant in the bigger picture, at a higher level of scale, is known as “course-graining” (gell-mann, 1995; west, 2017). this is in contrast to being lost in “fine-grained” details that may not be relevant to the big picture. fragile predictability and emergence with these ideas as background, the scene is set to explore selected complexity topics in detail. (see table 1 for suggested areas to cover.) the instruction begins with a discussion about complexity science itself and also about the ways that work life in organizations is complex in this technical sense. the goal is to try to have the participants relate science to experience. the map or model of complexity is not the inherent complexity of the object or the system. a manager’s mental model of what is happening in an emergency room during a crisis is not what is actually happening as events unfold. however, it is in the context of models, that one takes action in organizations. at the same time, as hazy and backström, (2013a) describe, every individual in the organization has a unique model and all of these unique perspectives interact with one another to produce outcomes. even though some aspects of the world appear to be dynamically stable, switching from one state to another state does not necessarily happen smoothly. often there is a tipping point, which is followed by a sudden change (thom, 1989; guastello, 2002). the mandelbrot zoom is, in fact, a boundary set that reflects transitions between different stability regimes. as an individual agent acting inside an organization, it can be difficult to move the organization from one region of dynamic stability to another. this uncertainty is why organizational life is so difficult to navigate. http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 63 university of bamberg press the last lesson in this module is critical for success in an organization and serves as a transition to module 2. when a bounded system, like an organization, is put under pressure from outside, the “system” may respond to events without the need for top down governance. new physical and social structures emerge, and the system spontaneously assumes a new dynamic state that alleviates this pressure more quickly. these emergent regularities (gell-mann, 1995) are called “dissipative structures” (prigogine, 1995; macintosh & maclean, 1999) because they dissipate the pressure more quickly than before. the relevance of these structures is explored in later modules. recap of module 1 the key lessons of module 1 are: first, a complex system is one where considerable uncertainty still remains in the system. therefore, although there is a level of predictability, there is also a potential for surprise, and this surprise can have broad implications. second, states of localized dynamic stability operate in different path-dependent regimes, and the boundaries between regimes can be discontinuous. it is important for practitioners to know and to expect that when the dynamical state of a “system” approaches the boundary of its stable region, its behavior can be wildly unpredictable even in response to minor event like, for example, the angry words of a frustrated manager or leader. third, when an organization is in a stable regime it might suddenly switch to a different stable regime due to pressures from outside the boundaries of the organizing system. however, when the organizing system assumes its new state, it doesn’t necessarily behave the way that an observer might hope or expect given the purpose of the organization. module 2: the complexity of organizations the study of complexity in the business and organizational context has become increasingly relevant over the last couple of decades (allen, maguire, mckelvey, 2011). at the conceptual level one can think of the human organizational ecosystem as a technology enabled complex system with spatial, temporal and social dimensions all interacting with one another to distribute information (hidalgo, 2015). because the events in an organization are complex, events as inputs do not necessarily result in proportional outcomes. as a result, organizational life can be thought of as attempts by individuals to understand, model and predict – in other words make sense of – complex phenomena, including one’s own actions, as a means to benefit proactively from an unfolding future. to make this point, the instructor might ask participants to consider some static “object” or phenomenon from their experience and to ask these questions: how complex is this? how does one quantify something’s complexity? consider a car, for example, a 2012 toyota camry le. with a health care audience, one might reference a magnetic resonance imaging (mri) machine. how complex is this object in the context of what is relevant to one’s life? this perspective reflects the complexity of an object in the context of what an observer sees as relevant at a moment in the ecosystem. paul cilliers (2005) would have said that the object itself, when it is removed from its social context, is complicated rather that complex because it can be broken down into component parts and reassembled. but this is not the case when the object is considered not in isolation but in the context of its use. a camry or an mri, even when frozen in http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 64 university of bamberg press time, is more than the sum of its parts when considered in the context of its use. one might think of this as the object’s “static complexity” as it is used as part of a dynamic process. now consider the moving image of that car in-use as it navigates the process of passing through a busy traffic circle, or imagine the mri machine in-use during the process of providing images for a medical team considering treatment options for a struggling patient. how complex is this process? this process of use over time is an example of “dynamic complexity” where one considers an object but this time in the context of its use in a process over time. how complex is the dynamic situation of the object as it interacts with other objects across space and over time? this is a more difficult question to answer. understanding the complexity or objects in use is important because the level of complexity helps one understand what can be predicted versus what one cannot know before the event occurs. a key learning from complexity is that static thinking is not enough. in complexity, one must always engage in dynamic thinking by assessing every situation in the context of emerging patterns in what scientists call phase space (ruelle, 1989). this second module looks at how managers and leaders might recognize and engage dynamic complexity in their organizations and help others navigate through complexity with them. it begins with a discussion of algorithmic or descriptive complexity and suggests how the articulation of descriptive models would enable the leader to exert informational influence over others to enact organizational outcomes over time (prokopenko, boschetti & ryan, 2009). that discussion is followed by a description of dissipative structures (macintosh & mclean, 1999) and structural attractors (allen & strathern, 2003) and how these resonate with social structures such as norms that organize human activity (giddens, 1984). these structures when present exert normative influence on individuals and serve to channel choice and action. as such these complex structures are important tools – as well as impediments – for managers and leaders alike. the third and final lesson of this module is the importance of periodically transferring energy and information into the structure of the system. this serves to both reinforce and reenergize positive structures and to dampen negative ones. periodic forcing involves regular and continual insistence that useful order be maintained. this is done by inserting requisite energy and resources that are needed to keep things on track. effective leaders judiciously apply this type of incentive and coercive influence as periodic forcing to establish and enforce sets of local interaction rules that keep various parts of the organization connected and moving in the right direction. descriptive complexity and informational influence to set the stage for a discussion of complexity and organization, it is useful to begin this module with a popular complexity video of flocking behaviors in “a murmuration of starlings”. by observing flocking behavior, participants see first-hand the beauty and “complexity” of organizing as forms morph over time. these organizing forms emerge from relatively simple local rules of interaction as these also interact with objects in their ecosystem. the leadership challenge is to figure out how one should think about influencing complex adaptive systems like these as they unfold over time. http://dx.doi.org/10.20377/cgn-55 65 university of bamberg press complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 descriptive complexity and coarse-graining the idea of coarse-grained “descriptive complexity” is that mathematical or even conceptual models can be used to study complex systems at different levels of scale, depending on the events that are relevant to the observer. the details are not always important or relevant. some events and some details do not need to be predicted. consider the problem of predicting the outcomes of a particular system, the starlings swarming in the video, for example, or the hustle and bustle of hospital emergency room during a natural disaster. for a model to be a useful enabler of action, one must first define the purpose of the model and what it is intended to predict. second, it must be able to process information available at time period t = 0 and produce predictions for time period t =1 before time actually reaches t = 1 (hazy & backström, 2013a). for the leader who seeks to act in a complex organization, the question is this: given what is relevant under the circumstances – patient outcomes for example – how large would a model have to be for it to take inputs, run and output accurate predictions (some subset of) the outputs of the system (within an acceptable margin of error) in time to take corrective actions before the outcomes actually occur? the size of this model, measured either by the time it takes to run (where the time is less than the time from t = 0 to t = 1) or the memory required to store it, is called descriptive complexity (hazy & backström, 2013a). one might note that the size of the model one uses to predict events depends upon how one defines relevant aspects of the physical and social system during what is referred to as the “coarse-graining process.” in the camry example described earlier, the atoms and molecules, and even the parts of the automobile probably don’t matter. if one cares only about the camry as part of a coarse-grained “system” that describes patterns of motion in a traffic circle with five possible exit paths, then the model would only need to predict a single output, the camry’s one exit path from the circle out of five. a model to predict the camry’s exit path might be designed to detect, sample, and process the current state of the camry’s turn signal. this coarse-graining could be a useful model if the potential for error that arises from this gross coarse-graining simplification process is acceptable. one should note that for this model, most of the state-data (for example, all of the other automotive systems) are discarded during the coarse-graining process in return for the relative simplicity of this predictive model. also note that the fine-grain physical and social system retains all of its considerable uncertainty; it is only this coarse-grained model that is “simple.” in contrast, if it is important for the observer to predict the precise route, timing, and velocity of the car, then the coarse-grained model must be much more complex. for a more complex model to also be more useful – to pilot a self-driving car for example – additional uncertainty and thus additional fine-grain detail must be accounted for by the model. this is the same as saying that more information about the fine-grained dynamics of the system must be incorporated in the observer’s model. this, in turn, means that the model must use more memory and take more time to process and would therefore have greater descriptive complexity. this distinction highlights the importance of selecting the level of coarse-grained scale from which to observe a physical and social system and choosing the relevant variables for modeling. at the same time it also highlights the difference between the complexity of a model of the phenomenon and the complexity of the phenomenon itself. the complexity of the model depends upon the information that is relevant to the observer and is therefore dependent upon the position of the observer in the system. as shown in table 1, nobel laureate murray gellmann (1995) calls sentient observers within a complex adaptive system “information gathering and using systems (igus)” because their behavior depends on the information that they are able http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 66 university of bamberg press to gather and use given their unique and limited position and capacities. to predict the behavior of an igus one must also predict the coarse-grained descriptive complexity of its model-in-use. descriptive complexity and management insights from complexity science about coarse-graining and descriptive complexity are particularly relevant to managers and leaders today because information and communications technologies (ict) together with machine learning and artificial intelligence (ai) are advancing so rapidly that the old ways of organizing no longer work (arthur, 2015; west, 2017). new, more thoughtful approaches are needed. the key lesson for managing through complexity is this: many aspects of coarse-grain models that worked yesterday still work today even though many, even most of the irrelevant finegrain details have changed. furthermore, many of these coarse-grained processes will still be working tomorrow even after all of the complexity is better understood. routines and processes are the descriptive models that reflect the algorithmic aspects of organizing that managers and leaders can control. they can control them because they designed them to get the job done, and they sustain them through their actions (nelsen & winter, 1982; beyer, 1999). managers should double down on routine activities that are known to drive outcomes that continue to be relevant. at the same time, however, there are also other aspects of our businesses that can no longer be run using the same routines or “algorithms” because the old way of working will no longer offer correct or relevant predictions about events and outcomes. these are the activities that the manager did not create and does not sustain by his or her actions. in health care, this might include natural disasters which overwhelm medical facilities or regulatory changes that impact economic flows. in these cases, it is the manager’s challenge to search for new and emerging coarse-gained simplicity that will eventually be relevant on the other side of complexity. one can then generate new algorithmic processes to drive desired outcomes. in these cases, the leader must enable a generative process (surie & hazy, 2016) where each member of the organization has the knowledge, skills, and capacity to build his or her particular piece of this new model that is customized for use in his or her particular situation. but at the same time, all of these models must fit together to be effective in the context of the organization’s mission and objectives within its ecosystem (surie & hazy, 2006; hazy & backström, 2013b). by communicating, supporting and evolving existing models that still work, while at the same time building new coarse-grained models of emerging complex structures, today’s leaders use their position and power to exert informational influence across the organization and enable others to be successful in their organizations. dissipative structures and normative influence models of events in the physical and social world are one thing, but complexity also helps us understand the ordered structure of the physical and social world and events within it. to lead others in a complex world, managers must accept that, although organizations are dynamically stable at the coarse-grained level in the short term, they are also constantly changing at the finegrained level. this means that every individual must create and maintain his or her own individual internal model to predict unfolding events. this can be difficult and time consuming. http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 67 university of bamberg press further, if each person had to do this independently, there would be considerable redundancy and conflict. fortunately, conceptual models can be shared (tomasello, 2014). however, to do so effectively and without loss of predictive power, they must all be built upon a common and robust conceptual and analytical platform which is likewise shared. dissipative structures one type of dynamic pattern that can be identified and modeled by managers as coarse-grain organizing structures are called “dissipative structures” by nobel laureate chemist ilya prigogine (1995). prigogine studied dynamic flow structures that emerge in closed systems in response to excess energy entering the system from outside the system’s boundary (lichtenstein, 2014). he found that these new flow patterns served to dissipate the excess energy accumulating inside the system by sending it back across the boundary and into the ecosystem more quickly. for example, consider the case where heat is applied to a “closed system” – a round glass container, flat on the bottom and top – containing a viscous liquid. before heat is applied, the viscous fluid can be represented using a coarse-grained model that reflects a relatively homogeneous highly symmetric fine-grained system in thermodynamic equilibrium. as low heat is applied to the bottom of the container, however, the molecules of fluid on the bottom become warmer until internal fine-grained differences in temperature appear as the system gradually transitions into a more heterogeneous non-equilibrium fine-grained system. in this low heat phase, excess heat is dissipated through the system by fine-grained conduction dynamics without the need for disruptive change in organizing structure. during conduction, molecules – or “agents” – in the liquid remain roughly in the same place, bouncing off one another in finegrained interactions. as this occurs, kinetic energy (and information) are gradually transferred to neighboring molecules. eventually, the heat is dissipated into the environment as molecules collide with other molecules on the boundary and beyond. as the amount of external heat applied at the bottom of the container is increased beyond a threshold point, a phase transition suddenly occurs as circular convection flows appear to dissipate excess energy from the system more quickly. molecules at the bottom of the container move en masse to transport heat more quickly from the hot bottom region to the cool top region. during the transition at the fine-grained scale, the observer must likewise transition from a conduction coarse-grained model to a new coarse-grained model where convection “cells” are modeled as coarse-grained “dissipative structures.” this discontinuous change is represented in phase space as “coarse-grained simplicity on the other side of fine-grained complexity.” it is easy to see why organizational theorists find metaphorical relevance in dissipative structures for the study of organizations (lichtenstein, 2014). however, one must be careful not go too far and confuse mere metaphor with hard science (hazy, 2014). coarse-grained structures drive normative influence what is important to realize about dissipative structures is that coarse-grained models of finegrained regularities are sometimes able to consistently predict behaviors at a higher coarsegrained level of scale than that of an individual agent. this implies that when dissipative structures form, individual agents are influenced to move en masse with others according to http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 68 university of bamberg press shared expectations even though they may not realize why they are doing this. from the participating agent’s perspective, it just makes sense to go along, to “go with the flow.” by conforming, these agents increase the pressure on others to likewise conform, and so on and so forth. this is how normative influence acts to organize individuals. it biases or “attracts” the choices and actions of individuals by enabling them to resonate with one another, reducing cognitive load and simplifying the individual choices that must be made. in this sense, one can see how social structures (giddens, 1984) can act as “attractors” (surie & hazy, 2006). there is a need for caution here. a dissipative structure (and therefore normative influence more broadly) forms inside a system to resolve local tensions or problems. the structure does not form to further the organization’s objective. in a recent us government scandal at the veterans health administration (vha), the health system had too many patients coming in and too few doctors to provide treatment for them (bronstein & griffin, 2014). as a result of tension from excessive patient demand flow that conflicts with available resources and institutional performance metrics, a dissipative structure formed as health care professionals began to refuse to add new patients to the waiting list. this allowed the system to maintain dynamic stability, albeit it did so by failing to serve its institutional purpose. this obvious misstep is not the fault of the individuals. it is a system failure. but even more, it is a leadership and management failure. if no one recognizes that a pathological structure is forming or no one has the foresight and skills to guide its development in a productive direction, the system as a whole can become counter-productive to the organization’s stated objectives. leaders and managers must be aware that this occurs and learn how to guide these complex dynamics toward an organizational purpose. this caution aside, leaders and managers can learn to use normative influence to further an organization’s objectives by constructing and reinforcing useful dissipative structures that form in organizations to solve local problems (macintosh & maclean, 1999). they can also learn to counter the influence of dissipative-like structures that are counter-productive and learn to do so before these structures harden into the more entrenched structural attractors, like a smoking area, for example, the very presence of which attracts certain activities toward them (allen & strathern, 2003). because human beings have cognitive abilities that allow them to overcome normative influence (tomasello, 2014), leaders can learn to overcome counter-productive normative influence in organizations by using informational influence. effective leaders must be skilled at each of these influence types, as well as a third type, incentive and coercive influence, that operates at the physiological and safety layers of maslow’s (1943) hierarchy of needs. periodic forcing through fine grained incentive and coercive influence once social structures have formed within a system, and once models are generated and articulated, individuals can use the information in them to predict events. models motivate individuals through payouts (of various types) while norms enable cooperation by coordinating collective behaviors (hazy & backström, 2013a). in this way, both informational and normative influence act upon individuals to move the organization forward. this occurs whether the ecosystem forces that are driving order and biasing collective behavior across levels of scale were formed spontaneously to relieve local stress or were guided by leadership to serve organizational purposes. however, due to “frictions” within the ecosystem, additional energy (and information) must be injected into the system periodically to reinforce and evolve processes, functions and http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 69 university of bamberg press orderly activity. like peddling a bicycle, periodic forcing applies incremental energy into the system to sustain dynamic stability. like a clock that signals a shift change in a hospital, periodic forcing can also inject “time ordering” into the organizing system. in organizations, temporal order can be reinforced through periodic forcing of reporting requirements or status meetings that sustain the organization’s “ordering” regime through space and time. this extrinsic force is called incentive and coercive influence and is discussed in more detail as part of module 3. recap of module 2 the key lessons of module 2 are: first, leaders must learn how to recognize and engage dynamic complexity in their organizations. a nuanced understanding of descriptive complexity helps a leader articulate models such as economic games which exert informational influence over others to enact organizational outcomes (prokopenko, boschetti & ryan, 2009). second, leaders must learn how to recognize normative influence forces exerted through structural attractors such as dissipative and social structures which organize human activity at the institutional level (giddens, 1984). the final lesson highlights the importance of periodically transferring energy and information into the system to replenish what is dissipated by frictions and doing so in ways that reinforce and reenergize positive structures and dampen negative ones. effective leaders apply incentive and coercive influence to guide the choices and actions of individuals to keep the organization moving in the right direction. module 3: the implications of complexity for leadership the third module applies what was learned in the prior modules to identify lessons for leaders in today’s complex organizations. it is useful to begin the session with a video of busy activity at a local establishment, preferable one that is familiar to the participants, like a local starbucks. this video will be used later to offer some examples about how leadership and management work in familiar complex contexts. the goal is for participants to search inside themselves and explore how they can have a greater impact, and how they can help others to be more effective. complexity leadership offers a new way of thinking (hazy & uhl-bien, 2015). it offers different insights because it focuses on what leadership does for an organization, and why leadership is needed to keep social systems moving in the right direction. this is in contrast to more common leadership approaches that help an individual learn to be perceived as a leader or to accumulate power that can be used to move others in a desired direction. the complexity leadership difference complexity leadership approaches are distinct from and complementary to traditional leadership study. this is because traditional approaches (which may be discussed in some detail during the session but are not discussed here) relate to how the leader activates the follower to move in a certain direction, but they are agnostic about that direction. the underlying assumption in http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 70 university of bamberg press traditional leadership theories is that the leader knows what to do and how to do it and that the only challenge is organizing others to follow that direction. in contrast, complexity leadership is interested in the process of identifying, framing and then moving the system in the right direction. the implicit assumption is that organizations need to continually acquire and process the resources that they need while adapting to change under uncertainty. as such, leadership is a collective phenomenon. as hazy (2012) shows through agent-based computer simulation, the hard part of leadership is figuring out what things needs to be done, who can do them, and how. once the way forward is identified and the team is organized around a champion, all of the traditional leadership approaches are useful in various situations according to the capabilities and needs of the organization (helfat, finkelstein, mitchell, peteraf, singh, & teece & winter, 2007). part of the process of moving the organization forward in a collective leadership effort includes shepherding individuals and groups on a journey through fine-grained complexity to find the coarse-grained simplicity that emerges on the other side. this module describes how a leader can act as a catalyst by framing and shaping the environment to channel influence interactions that enable the organization as a whole to identify and achieve its objectives. generative leadership and informational influence the first leadership challenge involves navigating and directing an organization through the information gathering and using process to find, create and achieve value for stakeholders on a competitive landscape (gell-mann, 1995; hazy & uhl-bien, 2015). we call this process “generative leadership” because it recombines an organization’s existing capabilities (helfat et al, 2007) in the context of emerging value potentials in the ecosystem. by doing so, it generates the next iteration of the organization, and then the next, and so on (surie & hazy, 2006). one can think of this metaphorically as identifying and facilitating the nascent formation of purposeful dissipative-like structures (prigogine, 1995) as the organization exchanges information and resources with other entities in the ecosystem (goldstein, hazy & lichtenstein, 2010). across the organization, recombination decisions are framed as choices to be made at various positions and in various contexts. this approach is distinct from command and control leadership since is places decision authority with those who are in the best position to identify and realize the value potentials that are available to move the organization in new and innovative directions (hazy & uhl-bien, 2015). the generative leadership approach activates intrinsic motivation by acting at the self-esteem and self-actualization levels of maslow’s hierarchy of needs (maslow, 1943). from a practical perspective, generative leaders combine behavioral economics (kahneman, 2011) and economic game theory (von neumann, morgenstern, 1944) to organize and coordinate individual choices in the context of organizational needs and goals. the idea is that the challenge of getting any organizing initiative going is a game with a perceived payoff matrix – or “value potential” – for players. the goal is to frame information about the situation in terms of the payout (monetary and otherwise) that each player will receive under various outcomes (guastello, 2002). by framing situations in this way, generative leadership uses information about potential benefits to participants to exert informational influence on each individual’s choices and actions as they interact in a specific “organizing game” that the leader designs. note that the payoffs and choices can vary among players and can be competitive or http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 71 university of bamberg press cooperative (axelrod, 1984, 1997). to complicate the matter, some individuals might be vying for promotions while others might be worried about putting their kids through college. community-building leadership and normative influence the approach described in the prior section assumes that actors are motivated by their individual payoffs. but often, the maximum payoff can only be achieved with a level of cooperation (axelrod, 1984, 1997). thus, another significant leadership challenge involves building a generalized sense of trust in the organization (zak, 2017). to do this, leadership must create, nurture and support a strong community with a shared identity that enables cooperation (hazy & silberstang, 2009). trust is closely related to the sense of physical and psychological safety that is engendered by being part of the group. evolutionary biologists call this “nesting safety” (nowak, tarnita, & wilson, 2010) because with trust, individuals can let their guard down (zak, 2017) along some dimensions. together, they become “us” through what maslow (1943) calls “belongingness.” the “attraction” to be one of “us,” enabled through generalized trust, reduces cognitive load by enabling normative influence that biases individuals toward cooperative choices. individuals are able to reduce their cognitive load by imitating those around them. from the perspective of the “system of individuals in an organization,” one can see that the formation of “us” is related to the formation of dissipative structures (macintosh & maclean, 1999). under the distorting presence of a huge payoff, like an initial public offering (ipo), or a loss, like in a bankruptcy, a dissipative structure might form to generate normative influence that helps organize individuals to act en masse to realize an organization level opportunity or avoid a threat. hazy and boyatzis (2014) argue that such changes would be signaled by an alignment of the emotional states of individuals in a common direction through a process of emotional contagion. this distorting potential can be strong enough to dominate the individual identity that, absent the activation of an overriding social identity, would cause an individual to “defect” in an effort to satisfy his or her own narrow interests. when one is feeling the “power of us,” one is being attracted by the gravity-like pull of normative influence. as many have experienced, once a group forms it can become more effective than the sum of its parts. the presence of structures, even an identity structure like “us,” attracts others who seek to join, a process called cumulative advantage (barabasi, 2002). these social and physical structural attractors (allen & strathern, 2003) “attract” others to participate in them because doing so provides benefits. how leaders create, evolve and use physical structural attractors – like warehouses and office space design – or social structures – like project teams and affinity groups – to guide, dampen and amplify informational and normative influences on members of the organization are a critical aspects of effective leadership in complex organizations. administrative leadership exerts incentive and coercive influence structural attractors are self-reinforcing artifacts in the physical, social, or the symbolic world that, like dissipative structures, by their dynamic presence exert normative influence on individuals, biasing their choices in their direction (allen & strathern, 2003). for many readers and students, this sounds like a roundabout way of saying “bureaucracy” which is indeed an important part of the structural attractor landscape in organizations. http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 72 university of bamberg press bureaucracies and structural attractors more generally are used by leaders and managers to organize recognizable signals that individuals can sense and unconsciously decode to do their jobs. because they carry information about process and function, they are also useful transmitters of informational influence. a written agenda, a meeting room table, an amphitheater, even roads, highways and bridges are structural attractors that can be used to organize routine work. warehouses and factories, hospitals and clinics, shopping centers and main streets, even social structure and status (giddens, 1984) carry signs and signals that enhance their usefulness as attractors to organize of human activity. because they are so powerful, it is important to remember that bureaucracies and structural attractors can influence individuals both for good and for ill. this is because they form initially as agents act to dissipate tensions that are locally stressing the system (macintosh & maclean, 1999). an overwhelmed manager may ask for weekly status meetings as an ad hoc or stop gap measure to figure out what to do, for example, but it is not uncommon for those meetings to continue to occur regularly, months or even years later. thus, an organizing structure that solved a local problem at a point in time may soon become counter-productive. corrupt practices, for example, might begin with harmless good intensions – to clear a warehouse of old inventory – but then over time could become institutionalized in a corrupt practice wherein remaining inventory “falls off the truck” at the end of each month. as these examples show, a structural attractor that forms doesn’t necessarily guide the “dissipation of tension” in a manner that serves the organization’s purpose. for these reasons, a complex organization requires administrative leadership to continually “administer” the extrinsic forces that are exerted by structural attractors, reinforcing positive forces and dampening negative ones. effective leaders learn how to manipulate the local environment to use these powerful sources of influence to guide the organization in the desired direction. to do this, leaders must realize that beyond their normative influence, physical assets such as financial resources, and social assets like reputation and status, can be manipulated to become vehicles of extrinsic motivation at the individual level. control of these resources by the leader provides a means to offer incentives or exercise coercion in order to promote compliance. this type of influence acts on individual motivation at the physiological and safety levels of maslow’s (1943) hierarchy. even as effective leaders apply the brakes to stale routines, they reinforce useful ones (hazy & uhl-bien, 2014). administrative leadership is responsible for establishing and reinforcing structural attractors including their informational and normative influences. to do so, leaders must periodically exert targeted incentive and coercive influence at the individual level to inject energy, information and order into the system to sustain its useful functioning. traces of leadership in daily activities – an illustration in a seminar setting, to make these points it is useful to observe an in situ structural attractor that is quite complex and perhaps one that is familiar to participants. a video of the morning rush at a starbucks or perhaps a trauma center can be used as an example because it readily shows the many information signals and normative influencers that enable efficient operations under changing ecosystem conditions. here we consider the starbucks case. one can observe in such a video that generative leadership implemented by starbucks management in the past has “set the agenda” for a “starbucks morning” game. one can see that over the years starbucks leadership has created an informational and normative influence http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 73 university of bamberg press environment that draws the consumer into the “game” without that individual even being aware that he or she is playing the starbucks game by starbucks rules. although originally created years ago, these structural attractors have been reinforced or dampened over time to optimize game play for the benefit of several sets of stakeholders. this is done through the skillful application of periodic forcing of daily routines using measured incentive and coercive influence that ensures, for example, that tables are regularly cleared and cleaned and inventories are restocked. this is an illustration of how the “structural attractors” that organize human activity are sustained and evolved. leaders must remember that structural attractors, like bureaucracies, are not machines. they are organic entities. as such they respond and grow according to local conditions without regard to the broader mission or purpose of the organization. as a result of the many mindless adjustments that have occurred locally over months and years to accommodate idiosyncratic local conditions, structural attractors meander like rivers. they morph opportunistically around obstacles, as happened when the us veterans administration employees simply stopped adding new patients to the waiting list to meet their numbers. sometimes meandering organizations can completely change direction as happens when an enterprise becomes corrupt. it is therefore important to realize that just because a structural attractor has momentum doesn’t mean it is furthering the larger goal. leaders must use their knowledge of complexity to stop what needs to be stopped, change what need to be changed, and reinforce what continues to make sense in the broader context of the organization’s purpose. recap of module 3 the key lessons of module 3 relate to the levers of influence that leaders can use to move their organizations forward in complex ecosystems. first, generative leadership actualizes models and descriptive complexity to frame as organization’s coarse-grained outcomes so as to exert informational influence across the organization. second, community-building leadership engenders and defines shared identities that support evolving structural attractor forces which exert efficacious normative influence on the organization’s members. finally, administrative leadership exerts incentive and coercive influence to reinforce various routines and capabilities that hold a complex and dynamic organization together. concluding thoughts: outcomes and next steps this seminar was offered twice in 2016 to executives from a not-for-profit health care system in the us that employs over 100,000 medical, technical and administrative personnel mostly in the eastern and mid-western us. the thirty-seven participants over the two sessions were midcareer and all were chosen because they were in line for promotion into the top management roles of the governing organization. each seminar occurred over three days, and each of the modules described herein took place in a three hour time-slot. other topics were covered in other time slots. participant responses to this complexity-informed advanced leadership development program have been very positive and supportive. participants consistently rated the program highly (scoring between 4 and 5 on a low = 1 to high = 5 satisfaction scale) with respect to both http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 74 university of bamberg press interest and relevance. comments tended to highlight the importance of relating the complexity material to the workplace. in terms of future improvements, the team is developing additional connections between complexity ideas and real-life conditions in today’s complex workplace. these connections will be used as prompts to enable fruitful discussions about immediate and relevant work situations that could benefit from the complexity perspective. likewise, the team is developing experiential exercises that relate directly to the executive work situation, learning-by-working to solve their current leadership problems. finally, the program would benefit from the addition of one-on-one counseling sessions that would help participants apply what is being learned directly to their organizational situation and according to their personal leadership style. http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 75 university of bamberg press references allen, p. m. maguire, s. & mckelvey, b. (eds.) (2011). sage handbook of complexity and management. thousand oaks: sage. allen, p. m., & strathern, m. (2003). evolution, emergence, and learning in complex systems. emergence, 5 (4), 8-33. anderson, p. (1999). complexity theory in organization science. organization science. 10, 216-232. arthur, w.b. (2015). complexity and the economy. oxford: oxford university press. axelrod, r. (1984). the evolution of cooperation. new york: basic books axelrod, r. (1997). the complexity of cooperation. princeton: princeton university press. bak, p. (1996). how nature works." new york: springer-verlag. bak, p. & paczuski, m. (1995). complexity, contingency, and criticality". proceedings of the national academy of sciences of the usa, 92 (15), 6689–6696. barabasi, a. l. (2002). linked: the new science of networks. cambridge, ma: perseus publishing. beyer, j. (1999). transforming an industry in crisis: charisma, routinization, and supportive cultural leadership. leadership quarterly, 10(3), 483. bronstein, scott; griffin, drew (april 23, 2014). "a fatal wait: veterans languish and die on a va hospital's secret list". cnn. retrieved february 16, 2015. burt, ronald s. (1992). structural holes: the social structure of competition. cambridge, ma: harvard university press. isbn 978-0-674-84371-4. cilliers, p. (2005). complexity, deconstruction and relativism. theory, culture &amp; society, 22(5), 255-267. cover, t, & thomas, j. a. (2004). elements of information theory 2nd edition. hoboken, nj: wileyinterscience crutchfield, j (1994). is anything ever new? considering emergence. in g. cowan, d. pines & d. meltzer 9eds.), complexity: metaphors, models, and realty 515-537, reading, massachusetts: addison wesley publishing. crutchfield j.p. & feldman, d.p. (1997). statistical complexity of simple one-dimensional spin systems, physics review e 55 (2), 1239-1242. epstein, j. m. (1997). nonlinear dynamics, mathematical biology, and social science (vol. iv). reading, massachusetts: addison-wesley publishing. feldman d. p., & j. p. crutchfield (1998). statistical measures of complexity: why? physics letter a, 238 (415), 244-252. fleming, l., & sorenson, o. j. (2001). technology as a complex adaptive system: evidence from patent data. research policy, 30, 1019–1039. forrester, j. w. (1987). nonlinearity of high-order models of social systems. european journal of operational research, 30, 211–221. http://dx.doi.org/10.20377/cgn-55 http://en.wikipedia.org/wiki/per_bak http://en.wikipedia.org/wiki/maya_paczuski http://pnas.org/cgi/content/abstract/92/15/6689 http://en.wikipedia.org/wiki/proceedings_of_the_national_academy_of_sciences http://en.wikipedia.org/wiki/proceedings_of_the_national_academy_of_sciences https://en.wikipedia.org/wiki/drew_griffin http://www.cnn.com/2014/04/23/health/veterans-dying-health-care-delays/ http://www.cnn.com/2014/04/23/health/veterans-dying-health-care-delays/ https://en.wikipedia.org/wiki/international_standard_book_number https://en.wikipedia.org/wiki/special:booksources/978-0-674-84371-4 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 76 university of bamberg press gell-mann, m. (1995). the quark and the jaguar: adventures in the simple and the complex. new york: henry holt & company. giddens, a. (1984). the constitution of society. berkeley, ca: university of california press. goldstein, j. hazy, j. k. & lichtenstein, b. (2010). complexity and the nexus of leadership. new york: palgrave macmillan. guastello, s. (2002). managing emergent phenomena: nonlinear dynamics in work organizations. mahwah. nj: lawrence erlbaum associates. granovetter. (1973). the strength of weak ties. american journal of sociology, 78(6), 1360–1380. haken, h. (2006). information and self-organization: a macroscopic approach to complex systems. berlin: springer. hazy, j. k. (2012). leading large organizations: adaptation by changing scale-free influence process structures in social networks. international journal of complexity in leadership and management, 2(1/2), 52-73. hazy, j.k. (2014). ‘is this “thermodynamic inquiry” metaphor or more?’, journal of consulting psychology, 66(4), 200-305. hazy, j.k.. and backström, t. (2013a), ‘human interaction dynamics (hid): foundations, definitions, and directions’, emergence: complexity and organization, 15(4), 91-111. hazy, j.k.. and backström, t. (2013b), ‘editorial: human interaction dynamics (hid)—an emerging paradigm for management research’, emergence: complexity and organization, (15)4, i-ix. hazy, j. k. & boyatzis, r. e. (2015). emotional contagion and proto-organizing in human interaction dynamics. front. psychol. 6:806. doi: 10.3389/fpsyg.2015.00806 hazy, j. k. & silberstang, j. (2009). the emergence of collective identity as a means for creating and sustaining social value. in j. a. goldstein, j. k. hazy and j. silberstang (eds.), complexity science and social entrepreneurship,( pp.447-470). litchfield park, az: isce publishing. hazy , j. k. & m. uhl-bien. ( 2014). “changing the rules: the implications of complexity science for leadership research and practice .” in the oxford handbook of leadership and organizations, edited by d. day, 711 – 32 . oxford : oxford university press . hazy , j. k. & m. uhl-bien. (2015). “towards operationalizing complexity leadership: how generative, administrative and community-building leadership enact organizational outcomes.” leadership 11 ( 1 ): 79 – 104 . helfat, c. e., finkelstein, s., mitchell, w., peteraf, m. a., singh, h., teece, d. j. & winter, s.g. (2007). dynamic capabilities: understanding strategic change in organizations. new york: blackwell publishing. hidalgo, c. (2015). why information grows: the evolution of order, from atoms to economics. new york: basic books. hirsch, m. w., smale, s., & devaney, r. l. (2004). differential equations, dynamical systems and an introduction to chaos (2nd ed. vol. 60). amsterdam: elsevier academic press. holland, j. h. (1975). adaptation in natural and artificial systems. cambridge, ma: the mit press. http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 77 university of bamberg press holland, j. h. (2001). exploring the evolution of complexity in signaling networks. retrieved august 22, 2002, 2002, from http://www.santafe.edu/sfi/publications/wplist/2001 kahneman, d. (2011). thinking, fast and slow. new york: farrar, straus & giroux. kauffman, s. a. (1995). at home in the universe: the search for the laws of self-organization and complexity. new york: oxford university press. lichtenstein, b. (2014). generative emergence. oxford: oxford university press lorenz, e. n. (1963). deterministic nonperiodic flow. journal of atmospheric science 20, 130-141. macintosh, r., & maclean, d. (1999). conditioned emergence: a dissipative structures approach to transformation. strategic management journal, 20, 297-316. mandelbrot , b. (1982).the fractal geometry of nature. new york: w h freeman & co, 1982; isbn 0-7167-1186-9 maslow, a.h. (1943). a theory of human motivation. psychological review, 50, 370-396. marion, r., & uhl-bien, m. (2001). leadership in complex organizations. the leadership quarterly, 12(4), 389-418. mezirow, j. (1991). transformative dimensions of adult learning. san franciso: jossey-bass. mezirow, j. (2003). transformative learning as discourse. journal of transformative education 1(1), 58-63. nelson, r. r., & winter, s. g. (1982). an evolutionary theory of economic change. cambridge, ma: the belknap press of harvard university press. nowak, m., tarnita c.e., & wilson, e.o. (2010). the evolution of eusociality. nature 466, august 26, 2010, p.1057. doi: 10.1038/nature0920. ospina sm and foldy eg (2015) enacting collective leadership in a shared-power world. in perry jl, christensen rk (eds) handbook of public administration, 3rd ed. san francisco: jossey-bass, 489–507. prigogine, i. (1995). the end of certainty: time, chaos, and the new laws of nature. new york: the free press. prigogine, i. & stengers, i. (1984). order out of chaos. new york: heinemann. prokopenko, m., boschetti, f. & ryan, a. j. (2009). an information-theoretic primer on complexity, self-organization and emergence. complexity, issn 1099-0526, 15(1), 11-28. ruelle, d. (1989). chaotic evolution and strange attractors. cambridge: cambridge university press. surie, g., & hazy, j. k. (2006). generative leadership: nurturing innovation in complex systems. emergence: complexity and organization (e:co), 8(4), 13-26. thom, r., (1989), structural stability and morphogenesis: an outline of a general theory of models, addison-wesley: reading, ma. tomasello, m. (2014). a natural history of human thinking. cambridge, ma: harvard university press. uhl-bien, m., & marion, r. (2009). complexity leadership in bureaucratic forms of organizing: a meso model. the leadership quarterly, 20, 631–650. http://dx.doi.org/10.20377/cgn-55 http://www.santafe.edu/sfi/publications/wplist/2001 https://books.google.com/books?id=xj4qiebnp4gc&printsec=frontcover https://en.wikipedia.org/wiki/special:booksources/0716711869 http://en.wikipedia.org/wiki/ren%c3%a9_thom complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 78 university of bamberg press von neumann, j. and morgenstern, o., (1944), theory of games and economic behavior, princeton: princeton university press. watts, d. j. (1999). "networks, dynamics, and the small‐world phenomenon". american journal of sociology. 105 (2): 493–527. doi:10.1086/210318. west, g. (2017). scale: the universal laws of growth, innovation, sustainability, and the pace of life in organizations, and companies. new york: penguin press. yammarino fj, salas e, serban a, shirreffs k, and shuffler ml (2012) collectivistic leadership approaches: putting the “we” in leadership, science and practice. industrial and organizational psychology 5(4): 382–402. zak, p. j. (2017) the neuroscience of trust. harvard business review, january-february, 84-90. http://dx.doi.org/10.20377/cgn-55 https://en.wikipedia.org/wiki/digital_object_identifier https://doi.org/10.1086%2f210318 79 university of bamberg press complexity, governance & networks – vol.4, no 1 (2018) special issue: teaching complexity, p. 59-79 doi: http://dx.doi.org/10.20377/cgn-50 table 1 – important complexity concepts that can be covered concept relevance researcher self-similarity & fractals stability or instability at a level of scale are also likely at other levels of scale. mandelbrot (1982) ruelle, (1989) critical points: tipping points, bifurcation &, selforganized criticality conditions that enable stability can change quickly and unpredictability. bak (1996); bak & paczuski (1995) thom (1989) algorithmic or descriptive complexity naïve-simple models, such as heuristics, can predict outcomes that one can control. complex models predict the behavior of a system outside one’s control. complex models are probabilistic. crutchfield (1994) crutchfield & feldman (1997) feldman & crutchfield (1998) information gathering & using systems (igus) complex systems select pathways that are modeled. the action of each “agent” is limited by the information it can access. gell-mann (1995) cover & thomas (2004) self-organization as information storage discontinuous change is signaled by information buildup. “learning” takes time. haken (2006) dynamical systems & phase space each spatial dimension has two degrees of freedom: position and rate of change in position. this implies thinking in 6dimensional phase space. hirsch, smale & devaney (2004) forrester (1987) epstein (1997) sensitivity to initial conditions & attractors significant changes in outcomes follow small changes in inputs. “butterfly effect” lorenz (1963) complex adaptive systems (cas) complexity involves individual “agents” interacting according to each one’s local rules of interaction. holland (1975, 2001) axelrod (1984, 1997) fitness landscape and social networks too many control signals among individuals agents leads to catastrophe. too few connections reduce adaptability. kauffman (1995), burt (1992) barabasi (2002) granovetter (1993) watts (1999) dissipative structures in a non-equilibrium system. recognizable structures form spontaneously as excess energy flows through an organization. these dissipate excess energy to prevent “overheating”. prigogine (1995) prigogine & stengers (1984). cumulative advantage positive feedback that is proportional to size allows “the rich get richer.” barabasi (2002) structural attractor physical and social artifacts that provide signals about “choices” to help individuals navigate complexity. allen & strathern (2003) http://dx.doi.org/10.20377/cgn-55 http://en.wikipedia.org/wiki/per_bak http://en.wikipedia.org/wiki/maya_paczuski adaptive governance and sub-national climate change policy: a comparative analysis of khyber pukhtunkhawa and punjab provinces in pakistan complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 81 university of bamberg press adaptive governance and sub-national climate change policy: a comparative analysis of khyber pukhtunkhawa and punjab provinces in pakistan authors: muhammad mumtaz1 and saleem h. ali2 1 phd research fellow, são paulo school of management, fundação getulio vargas, brazil 2 professor, energy and the environment, university of delaware, usa e-mail: mumtaz86@hotmail.com this study explores the adaptive governance and effective implementation of climate policies at the subnational level in a developing country context. we focused on pakistan as our central case as it is considered one of the most vulnerable countries to climate change and has also gone through a recent governance devolution process. this study is conducted to investigate climate governance at subnational level in pakistan by looking at the province of punjab and khyber pukhtunkhawah (kpk). we employ the ostrom’s institutional analysis and development (iad) framework for this study. the framework as methodology is important to uncover the complexity of adaptive governance at subnational level after devolution and transformation of environmental institutions in pakistan. different aspects of governance such as engagement of local actors, activism of political leadership, awareness campaigns, and capacity building are the notable initiatives in the provinces. the study identifies the differences of initiatives in these provinces are manifest in subnational climate change policy differentiation, research capacity and institutional maturity. the study finds that the provincial government of the kpk follows more participatory and decentralized approach while punjab is more centralized. the iad framework provided an effective means of understanding these complex differences in outcome and scale. keywords: framework, climate change, institutions, governance, pakistan introduction climate change poses serious existential threats to pakistan’s wellbeing and is considered one of the most serious national crises in pakistan (salik et al. 2015). pakistan’s vulnerability to climate change stems from its diverse geographical and climatic conditions (rasul et al. 2012) . climate change has caused various disasters in the form of floods, droughts, and other natural calamities in the country over the past two decades (banuri et al. 2012) climate change is particularly a real challenge for the agrarian economy of pakistan which comprises the most significant economic and livelihood sector (rasul et al. 2012). food security of millions of pakistanis is dependent on climate, which is highly sensitive to changes, keeping in view the important role of the agriculture sector which provides 45% of total employment in the country (muhammad abid et al. 2011; pervin et al. 2013) to manage the potentially fatal consequences of climate change, pakistani national government responded with various national policy initiatives. these major initiatives are in the form of national climate change policy (nccp) of pakistan and framework for implementation of climate change policy. both these documents mainly focused on adaptation measures. the nccp clearly states that “adaptation effort is the focus of this document” (nccp 2012: p6) pakistan is ranked in the list of the countries that have the least adaptive capacity due to extreme poverty and lack of physical and financial resources (muhammad abid et al. 2011; adger et al. 2007; http://dx.doi.org/10.20377/cgn-68 mailto:mumtaz86@hotmail.com complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 82 university of bamberg press weber and khademian 2008). the subnational governments in pakistan are principally responsible for implementation of climate change adaptation strategies. subnational governments are key and more effective in curbing climate change due to their proximity to the consequences of climate change ( puppim de oliveira 2019) pakistan recognizes the important role of subnational governments/provinces for effective response to climate change. after the 18th constitutional amendment in 2010, the responsibility of implementing climate change policies rests with respective provinces/subnational governments in the country. the objective of this study is to investigate adaptive governance by studying the case of punjab and khyber pakhtunkhawa in the agricultural sector. this study focuses to identify the key adaptive governance initiatives in agriculture sector in both provinces. the following sections present, background of climate adaptive governance, institutional analysis and development (iad) framework, research methodology, analysis of our study cases, and conclusion. background of climate adaptation governance historical evidence from a wide variety of sources informs us that there is a strong linkage between the social and ecological systems (sess) in the context of climate change; sustainable outcomes require an integrative approach to understanding ecosystem governance (folke et al. 2005). global warming and increased rates of biodiversity loss tell us that the sess do not and cannot exist in isolation; they are highly interconnected (berkes 2002). this complexity of the system is further complicated by human decisions for not taking proactive scientific decisions. these complications emerged as results of not properly managing the sess while promoting free market capitalism and dealing with the global economic, ecological and social capitals (zia 2013). the field of environmental governance (eg) emerged as a means of understanding sustainable natural resource use patterns and managing the complexity of the sess. the eg literature is considered as a link between the social and the ecological, and a mechanism to influence the trajectory of the sess (folke et al. 2005). the eg is a “set of regulatory processes, mechanisms and organizations through which political actors influence environmental actions and outcomes” (folke et al. 2005). other scholars define eg differently. for instance, kay et al. (2001) define it as the process of resolving trade-offs and to provide a vision and direction toward sustainability (chaffin et al. 2014) consider that “…eg is the system of institutions, including rules, laws, regulations, policies, and social norms, and organizations involved in governing environmental resource use and/or protection, and there are a variety of different approaches”. adaptive governance (ag) emerged an important approach to deal with uncertainty and complexity of sess. the ag literature emerged from the eg research. the ag deals with different form of resource management and confronts the complexity and uncertainty associated with rapid environmental change (chaffin et al. 2014). the ag is a form of governance which emerged from the theoretical search for modes of managing uncertainty and complexity in sess (folke et al. 2005). the concept of ag gained prominent attention in the scientific community as an alternative form of governance during the last decade (rijke et al. 2012). dietz (2007) first formally coined the term “adaptive governance.” folke et al. (2005).described the ag as a strategy to resolve the social conflicts of complex ecosystems while (chaffin et al. 2014) defines “adaptive governance as a range of interactions between actors, networks, organizations, and institutions emerging in pursuit of a desired state for the sess”. governance systems, especially the top-down or centralized systems, rarely match ecological complexity, especially in the face of rapid environmental change (chaffin et al. 2014). centralized http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 83 university of bamberg press governance via top-down directives and centralized policies often fails to generate concrete solutions for highly contextualized situations (lemos and agrawal 2006). more succinctly, (lemos and agrawal 2006) argued that the centralized decisions are often falls short in efforts due to largescale ecosystems that cross multiple jurisdictional boundaries. due to the governance challenges of a top-down approach, a number of bottom-up governance approaches emerged. these bottom-up approaches are framed by different groups of local actors, social networks, and as a result of collaboration among community leaders (lemos and agrawal 2006). the bottom-up approaches seem quite effective but these approaches also suffer in coordination due to complex geographies (cosens et al. 2014). there are other challenges for bottom-up approaches of governance. local governance is not always evidently representative of all stakeholders’ especially marginalized communities, like indigenous communities, who are deprived of rightful access to resources (cosens et al. 2014). to fix these challenges, there is a need to bring new approaches of governance which are capable to handle the governance hurdles and the complexity between the sess components. the ag is increasingly recognized as a right form of governance to address these challenges and uncertainties (t. dietz et al. 2003; folke et al. 2002; lebel et al. 2006). the ag addresses uncertainty through continuous learning, involving multiple stakeholders in decision making process and self-organization of governance systems (rijke et al. 2012). the learning process is an important component to understand and deal with the complex dynamics and uncertainty associated with such systems (folke et al. 2005). this learning process happens due to the interaction among individuals, organizations, institutions and agencies at multiple levels (olsson et al. 2006). such learning facilitates the replication of successful practices from each other and coordination of effective resource utilization. various international efforts are taking place to utilize different knowledge system and learning environments for enhancing the capacity building which is dealing with the complex adaptive system (armitage 2005). the ag is essentially involved in devolution of management rights and power sharing to ensure the participation of relevant stakeholders in decision making (folke et al. 2005). this devolution of rights, delegation of power and responsibilities, and access to heterogeneous local bodies can essentially contribute to the ag (e ostrom 2005). the presence of multiple communities contributes to accumulate diverse knowledge by interaction which ultimately enhances the adaptive capacity of the community (e ostrom 2005). decentralized networks have been proven to be effective to confront local problems at local levels (bodin et al. 2006). leadership plays an important role in organizational effectiveness and for the effective governance at local scales. leadership is instrumental for collaboration and interaction among different actors at different levels of governance structure. such leadership is important to frame change and reorganize to incorporate innovation and keep flexibility to deal with the complex dynamics of the system (folke et al. 2005). further, leaders are important for performing the key functions for the ag like trust building, linking actors, compiling and generating knowledge, establishing links among the networks. the lack of leadership can lead to inertia in the sess, while visionary leadership gives directions for positive change and transforming governance. along with proactive leadership, transformability is a key aspect of the ag. transformability is the capacity to produce a new system when the existing system of social, political, and economic impact remains untenable. transformability produces novel components and ways of life from various existing sub-systems to define a new system (walker et al. 2004). this transformation has four phases: (1) to prepare the system for change (2) to open an opportunity, (3) http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 84 university of bamberg press to navigate and transition and (4) to chart/establish new direction for management to build resilient of the new governance regime. to evaluate the innovative governance or ag, multiple frameworks and theories can be utilized. for instance, zia et al. (2014) suggest that the generalized autocatalytic set theory can be applied to understand the sess. in the developing world where the ag is in nascent stages and not fully evolved, to apply generalized autocatalytic set theory appears too early. therefore, in this study we employ the institutional analysis and development (iad) framework (e ostrom 2005) to understand and analyze the ag in two provinces of pakistan. institutional analysis and development (iad) framework the iad framework was originally established in the 1980s by elinor ostrom (cole 2014). the framework was refined by ostrom and other scholars in the following decades. the iad framework is one of the most distinguished and tested frameworks in the field of policy sciences (gibson et al. 2005). the iad framework is widely used as research methodology to study local management practices (benson et al. 2013) the iad framework is presented in figure 1. it consists of exogenous variables, an action arena, composed of actors located within action situations and affected by a set of external variables. actors’ interaction within action situations leads to outcomes, which feedback into the external variables and the action arena. figure 1: the iad framework source (e ostrom 2005) some researchers selected parts of the framework to focus on the action situations leading to interactions and outcomes (mcginnis 2011). it is a tested and most distinguished framework in the field of policy sciences (gibson et al. 2005). the iad framework is applied in many research disciplines. for instance, (abel et al. 2014) applied the iad framework to study polycentric governance and climate change. figure 2 shows how abel and his colleagues operatized the framework. http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 85 university of bamberg press figure 2: state climate policy action situations. source (abel et al. 2014) the core aspect of this framework is the “action arena,” which is composed of action situations and actors. “action situation” means social space where actors act, resolve the problems, and exchange the good and services (elinor ostrom 2007). actors are those who interact in the action situation to address a common problem. in our study, the action situation is to promote adaptive governance at subnational level to address the issue of climate change. in the action arena the interests of various stakeholders are discussed and produced or (re)shaped initiatives are presented to tackle the common challenge. therefore, in our case, we are examining the actions of two provinces to address adaptive challenges to climate change; the drivers behind these initiatives; and how these initiatives have produced differential outcomes between the provinces. methodology a comparative case study approach is employed for this study which is effective in providing a nuanced analysis of differentiated outcomes in organizational studies (de vaus 2002). in the initial phase of this study, we collected information from reports, policy documents, internet news archives, physical newspapers, and scientific & academic articles. a field study is conducted from november 2016 to april 2017. during this phase, 30 in-depth semi-structured interviews including six initial exploratory interviews were conducted with relevant stakeholders in both provinces. the respondents included policy experts, government officials, think tanks working in the area, nongovernmental organizations, academics, ministry of climate change, and environmental protection agencies in the respective provinces, farmer community, civil society, and climate change activists. these respondents were chosen because they are directly linked with climate governance in pakistan. the list of our respondents is attached in appendix a. the provinces of khyber pakhtunkhwa (kpk) and punjab were chosen as the sites of the case studies due to three major reasons. firstly, both the provinces are highly dependent on the agriculture sector. secondly, the provinces shared some common agro-ecological zone. therefore, it is expected that both the provincial governments might have taken same kind of initiatives and it http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 86 university of bamberg press is easy to compare these two provinces. thirdly, it was easy to collect data for from these provinces and they have comparatively better climate governance structure in pakistan. the case study areas kpk is the 3rd largest province in the country. kpk accounts for 11.9% of pakistan's total population and it contributes to 10% of pakistan's gross domestic product (gdp). it has approximately 27.5 million inhabitants (70% rural) and a gdp per capita of usd 1,037. agriculture is the major livelihood of the people in the province. the agriculture sector contributes to 48 percent of the total labor force and 40 percent to the gdp of the province (khan 2012; nomman and schmitz 2011). climate change threatens to have adverse impacts on agricultural productivity throughout kpk. the province is one of the most affected regions due to mega floods back in 2010 in pakistan. to deal with the issue of climate change, kpk is set to establish multiple initiatives in compliance with federal policies and plans. punjab is the most populous and second largest province of pakistan. punjab is a fertile agricultural region which holds an extensive irrigation network and plays a leading role in the development of the economy (muhammad abid et al. 2015). the province accounts for 56.2% of the total cultivated area, 53% of the total agricultural gross domestic product and 74% of the total cereal production in the country (badar et al. 2007; pbs n.d.). punjab mainly contributes for agricultural sector in the percentage of land (57.2%) in agricultural sector and the percentage share (53% percent) of pakistan’s agricultural gross domestic product (hanif et. al, 2010). agriculture sector in punjab is facing the impacts of climate change (ahmed et al. 2018). adaptation strategies are an important response to climate change. it has been pointed out that agriculture adaptation measures can reduce losses (fleischhauer and bornefeld 2006; hansen and jones 2000; nawaz khan 2010). the subnational government of punjab is taking adaptation steps to tackle climate change. findings of the study in this section the findings of the study will be discussed in the context of the iad framework in each province. action arena in the “action arena” of the iad, the involvement of relevant stakeholders is vital in our study. the major stakeholders are provincial governments, farmers’ communities, civil society organizations working in area of climate change, academics, and the federal government. provincial governments both provinces took steps to tackle the impacts of climate change, especially towards agriculture adaptation. climate change policies play an important role in handling the impacts of climate change. kpk province has established and operationalized its provincial climate change policy whereas punjab is still in the process of framing its policy. the main driver behind why kpk has established a provincial climate change policy, but not in punjab, is identified in this study as the political-will prioritization and sequencing of development tasks. the political leadership of http://dx.doi.org/10.20377/cgn-68 https://en.wikipedia.org/wiki/list_of_pakistani_provinces_by_gross_domestic_product complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 87 university of bamberg press kpk is actively involved to addresses climate change in the province because it sees this as a means of differentiating itself through self-reliance on its agrarian economy, whereas punjab sees external economic investment as its key differentiator. the sharing of knowledge is equally important for climate adaptation. it is noted that in both the provinces, autonomous adaptation is transferred from one place to another place or from one farmer’s community of one area to another farmer’s community of another area. there is a need to channel this transformation and the subnational governments have to play their role in order to fully incorporate such autonomous measures in actually government action plans. capacity building is important to adapt to climate change. the united nations framework convention on climate change (unfccc) recognizes and highlights the significance of capacity building for effective adaptation and mitigation to climate change. in the climate change adaptation process, capacity building plays a key role. this study identified that the importance of capacity building is well recognized in both the provinces, although punjab is ahead of the kpk in this arena due to its economic development ascendancy. both provinces especially the kpk need to enhance capacity building for effective adaptation policies. some international governmental organizations like the united nations development programme, the international union for conservation of nature and others are involved in both provinces for climate change adaptation training, policy advocacy, and capacity building. in parallel with governmental initiatives, the farmers’ community is also taking steps for climate change adaptation in agriculture sector in both the subnational governments. farmers’ community it is important to point out that local farmers are actively involved in autonomous adaptation in both provinces. the subnational governments encourage the engagement of farmers in climate adaptation policies. this research study noted that in both the provinces the farmers have changed their planting dates because their crops were badly affected due to heat/rise in temperature or unpredictable rainfall patterns. some others have changed their variety of crops as the farmers opt for the crops which are heat tolerant and have fewer damages in case of floods, droughts or heat waves. changing fertilizers and planting shade trees are other strategies that have been adopted by the farmers’ community. the overall textures of the autonomous initiatives in both provinces are fairly similar. four important autonomous adaptation initiatives have been identified in this regard: changing planting dates, changing crops types, changing fertilizers, and planting shade trees. all of these initiatives complemented organizational efforts from the provincial authorities. agriculture extension departments agriculture extension departments which work closely with the farmers at local levels collect data from the fields and gives training to the farmers at local levels. based on these initiatives, the farmers are able to execute what they learned in their farming practices. for instance, they are advised that they should plant seeds which have been tested in scientific labs and shown to have the capabilities to survive severe weather conditions. it is noted that many trained farmers approached the agriculture extension departments to obtain suitable seeds with respect to weather conditions. we were told by the agriculture extension department in faisalabad that after attending the training, the farmers’ community is encouraged to approach the department for more information about climate change, suitable seeds, and solutions for damages due to pests etc. http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 88 university of bamberg press coordination among relevant line departments is important for the implementation of any policy. the subnational government of punjab has established a link among the 26 agriculture institutes throughout the province in order to set up comprehension strategies for climate change and the agriculture sector. they regularly arrange meetings among these institutes to discuss the new challenges and the existing strategies to manage the negative impacts of climate change. for instance, ayub agricultural research institute (aari), which manages climate change related activities, is well familiar with what is happening in the agriculture extension departments at various levels and vice versa. by being aware of the activities of agriculture extension departments and others, aari can disseminate the positive activities among other institutions and set new targets accordingly. this study identified that agriculture extension departments in punjab province are active and more established as compared to the province of kpk. the kpk is behind due to lack of human resources, weak intuitional arrangements, and scarcity of financial resources. civil society organizations civil society organizations are playing an important role in both the provinces. some international and local non-government organizations are also working in the area of climate change adaptation. for instance, the climate change research center at agriculture university peshawar is a notable research institution in the kpk province. it is coordinating research activities for climate change, creating linkages with national and international research institutions and to actually train local farmers. one of the important works of this center is to develop district-wise climate scenarios which are very important for establishment of local adaptation action plans. some other international organizations are also providing assistance for policy research and advocacy for climate change adaptations for agriculture sector in the provinces. these organizations are one of the major pressure groups who push the provincial governments to take concreate action for managing climate change. academics academics are also major stakeholders in both provinces. people from academia were involved in establishing provincial climate change policies in punjab and kpk. academics play a vital role in handling climate change by producing scientific and academic studies. these scientific and empirical studies help the respective governments to establish action plans that are adaptive to empirical findings. furthermore, these academic institutions are a mean of promoting awareness about climate change through teaching courses on climate change. for instance, a professor at department of environmental sciences at the university of peshawar told us that many students have started working on climate change and agriculture sector in recent years due to the course offerings provided. the professor further highlighted that we are devising curriculum with provincial government so that new courses related to climate change can be offered in academic institutions throughout the province in alignment with skill sets needed by policy makers. likewise, in punjab, the agricultural university faisalabad is a leading academic institution that produced many studies on climate change, agriculture and related areas. these studies contribute to understand the impact of climate change and tell us the future projection of climate change and its likely impacts. these findings are quite helpful to proactively act against the negative consequences of climate change. http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 89 university of bamberg press federal government-ministry of climate change the ministry of climate change (mocc) at the federal level is also a stakeholder with provinces to deal with climate change. the ministry is responsible for dealing with all the climate related material at the international level. the ministry arranged quarterly meetings with all the provinces and federal units so as to assess the overall progress of pakistan to address climate change. the ministry is not directly involved in implementation of climate change policies and action plans. however, it can give guidance and support to the provinces for effective governance. after 18th constitutional amendment, provinces are responsible for implementation of climate change and other related policies. bureaucratic interaction a detailed organizational analysis of the bureaucracy for addressing climate change in both provinces allows us to understand the interactions, patterns, and outcomes in the form of governance. the ag in both provinces is still in a new phase of absorbing and understanding the responsibilities and functions of decentralization. the provinces got more autonomy in 2010 when the 18th constitutional amendment was promulgated in pakistan. to some extent participation from different actors are seen in decision making, establishing a policy, or framing an action plan as a result of the constitutional amendment. we have seen in this study that local institutions are emerging and are in line to perform more effectively and better interaction is observed among different departments. figure 3, presents climate change related institutions in kpk. http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 90 university of bamberg press figure 3: organizational chart of kpk climate change bureaucracy note: the lists of members of advisory committee and implementation committee are given in appendix b and appendix c environmental protection agency climate change cell on provincial level secretary fe&wd provincial climate change policy climate change action plan advisory committee on pccp pccpic implementation committee 17 members from different line departments and academia, technical experts 17 members from different line departments administration secretaries forestry wildlife environment http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 91 university of bamberg press the figure shows the interactions among various organizational units in kpk while devising provincial climate change policy and dealing with other climate change related matters. an official at the climate change cell responded that the concerns of all the related departments and stakeholders were fully addressed before sending the policy to cabinet – a point which has been validated by stakeholders as well. the involvement of related stakeholders clearly shows that the local institutions are getting space in decision making, policy formulation and policy action that allows for more effective adaptive governance in case of climate change shocks another salient aspect of the kpk’s approach is to involve and incorporate local knowledge and practices in action plans to deal with climate change. the government is exploring the adaptation strategies of farmers at local level so that their traditional knowledge and effective local practices can be well utilized for upcoming action plans. likewise, in punjab, an official at aari informed us that they arrange regular meetings of 26 agriculture related research institutions and other relevant stakeholders to discuss and establish common agenda. to him, each and every department has opportunity to explain its point of view and every view point of every stakeholder is discussed before taking any decision. for instance, agriculture extension departments bring the challenges and best practices of local farmers, these things are discussed and evaluated within the meeting so that implementable and sustainable strategies can be established. figure 4 presents climate change related institutions in punjab province. http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 92 university of bamberg press figure 4: structure of climate governance in punjab province. source: punjab climate change policy draft. the mocc at the federal level is also partially involved in the interaction process. although, it is not the responsibility of the federal government to formulate and implement aspects of climate change policies and action plans but it oversees the international obligation and legally binding elements of provisions in national laws. a committee composed of all provinces and federal units’ representation arranges a quarterly meeting at the ministry to discuss the overall achievements to curb climate change in the country. so far 15 meetings have been arranged at the ministry since its establishment. evaluation the iad framework does not give us a detailed criterion for the evaluation of policies; instead it provides some metrics such as efficiency, equity, finance, accountability, confirmation to the value and sustainability (nigussie et al. 2018). to establish the evaluation criteria for this study, the following key dependent variables associated with ag are focused: learning and knowledge sharing mechanism; stakeholder involvement; role of leadership; capacity building, inter-governmental relations; and monitoring evaluation efficacy. learning and knowledge sharing in our case study, both provincial governments are responsible for formulation and implementation of climate change strategies. therefore, both the governments have to promote and share information about climate change in their respective provinces. public awareness about climate change is an important factor for climate adaptation. public awareness aims to ensure that all relevant regional and sub-regional bodies understand the impacts of, and take action to respond to, certain climate impacts (awareness campaigns for behavioral change, 2015). http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 93 university of bamberg press it has been observed that both the subnational governments have launched public awareness campaigns about climate change in their respective spheres. both provinces have achieved success of public awareness up to a certain level. however, there is still a need to extend the public awareness campaign even further, perhaps even to the union council levels. the major drivers for promotion of awareness among the masses are to promote knowledge of climate change, to educate farmers and to tackle climate change more effectively. the sharing of knowledge is important for effective governance. it is noted that in both the provinces, autonomous adaptation is transferred from one place to another place or from one farmer’s community of one area to another farmer’s community of another area. however, there is a need to channel this transformation and the subnational governments have to play their role in order to fully incorporate such autonomous measures in actually government action plans. stakeholder involvement the involvement of relevant stakeholders is key for the implementation of effective adaptation strategies. in this study, the involvement of farmers in establishing actions and plans is highly important. adaptation to climate change requires the joint efforts of individuals, businesses, industries, governments and other actors who are confronted by the impacts of climate change (corfee-morlot et al. 2009). both subnational governments have shown that they are keen to involve the farmers’ community for the establishment of adaptation action plans for the agriculture sector. the governments are monitoring autonomous adaptation initiatives so that effective initiatives can be incorporated in upcoming action plans. role of leadership leadership plays a key role in success of climate governance, particularly in hierarchical societies such as pakistan. one of the most important factors which influenced the success of climate change governance is the involvement and active interest of the top political leadership (meadowcroft 2009). when the political leadership takes a keen interest then things can move forward smoothly. imran khancurrent prime minister of pakistan, with a following among the youth due to his erstwhile role as a famous sportsman, is very concerned about the impacts of climate change and has galvanized major interest to address climate change. the kpk province is governed by his party and it is considered the most vulnerable to climate change province in pakistan. in 2010, kpk was the major victim of the flooding in the country. it was historical the worst flood in pakistan. he has discussed climate change and its impacts on many occasions in pakistani society. based on his personal interest in this matter the kpk established a climate change policy and launched a reforestation project (billion tree tsunami project) in 2015. they successfully achieved the targets in 2017 by planting 1 billion saplings in the province. capacity building capacity building is important to adapt to climate change and the unfccc recognizes and highlights the significance of capacity building for effective adaptation and mitigation to climate change. http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 94 university of bamberg press in our study, the importance of capacity building is well-recognized in both the provinces, although punjab is ahead of kpk in this respect as shown in the above section. both provinces, especially kpk, need to enhance capacity building for effective adaptation policies. some international governmental organizations like the united nations development programme, the international union for conservation of nature and others are involved in both provinces for climate change adaptation training, policy advocacy, and capacity building. inter-governmental relations in the case of pakistan which has nascent democratic institutions, intergovernmental relations between the federal and provincial levels are very important to deal with climate change. before the 18th constitutional amendment in 2010, the federal government was responsible for all climate related matters including establishing climate policies and action plans. presently, it is the responsibility of provinces but it is important to get due help and information about climate change. moreover, the federal is responsible to present pakistan at international level. the study finds that there is weak coordination among the provinces and at the federal level. although, there is an evaluation committee at federal level but it’s role is very limited and it could not bring any substantial results. likewise, the coordination among the provinces is very weak. there is a need to establish a strong links among the provinces so that successful actions can be learnt and transferred to another place in other provinces. the coordination with ministry of climate change is equally important so the actual success/failure can be shared and presented at international level. monitoring and evaluation monitoring and evaluation are by definition vital for the ag, as only through such mechanisms adaptation can be operationalized. monitoring and evaluation helps to improve management decision-making, increase transparency and accountability, and effective implementation plans (bellamy et al., 2001; stem et al., 2005). in both provinces there is a nascent setup for monitoring and evaluation of climate change policy. for instance, in punjab at the arc, they arrange meetings of all 26 agriculture institutions to know the progress and outcome so that further plans can be established. however, in kpk, there is no such proper system. both the governments need to establish a comprehensive and dedicated evaluation and monitoring mechanism so that weaknesses and strengths of the actions can be monitored properly. conclusions to uncover the ag in pakistan remained complex in nature due to confusion in powers and scope of the institutions between the provincial and federal governments. the establishment of novel institutions at provincial levels after 18th constitutional amendment in pakistan and political structure in the country are also creating confusion while uncovering the ag. provincial governments are responsible for dealing climate change but there are certain institutional complexities to fully understand the ag in pakistan. the iad framework is important; it can help us to explore those complexities and explain the ag at the subnational level in pakistan. the concept of ag is in early stages of implementation in pakistan. effective ag emerges smoothly in politically stable and institutionally balanced societies. however, there is a case for ag as an enabling mechanism is also fostering such stability and thus proactive investment to foster such mechanisms is important in developing countries like pakistan as well. to shape the ag in http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 95 university of bamberg press countries like pakistan requires a highly deliberative process, keeping in view pakistan’s heterogeneous society and the unstable political atmosphere in the country. however, the evolution/emergence of the ag in the context of climate change has come about with relative success in both kpk and punjab provinces – which are ruled by different political parties and have divergent economic development priorities. on the heels of 18th constitutional amendment, at multiple fronts the political and institutional powers have delegated to the provinces at subnational level. this devolution of power and decentralization provided a foundation for emergence of the ag in the pakistan. by utilizing the iad framework, the study finds local actors are involved in establishing policies and action plans for addressing climate change challenges. multiple stakeholders are being involved in climate change governance at subnational levels. political leadership is active especially in kpk to promote sound and sustainable mechanisms to address climate change. the institutions are newly established and lack the specialty but they are considering their responsibilities in both the provinces for effective governance. the role of academics and voices of civil society are being properly considered to address the challenge of climate change in both provinces. the differences of initiatives in these provinces are manifest in subnational climate change policy differentiation, research capacity and institutional maturity. kpk government has developed and officially launched provincial climate change policy, whereas punjab government is in the process of formulating its policy. punjab, however, is ahead in terms of carrying out research work and developing institutional capacity. the most important initiative of the punjab government, inter alia, is launching an awareness campaign about climate change by publishing related literature in local languages, establishing a radio station, arranging farmer day, and writing articles for newspapers. moreover, the study finds that the provincial government of kpk follows more participatory and decentralized approach while punjab is more centralized. there is a room for improvements in order to overcome the weak aspects of the governance in the provinces. the reforms especially in the legal system, further delegation of power at community level and ensure proper evaluation and monitoring can enhance the ag and transformation of the ag in pakistan. references abel, t. d., stephan, m., & daley, d. (2014). climate risk polycentricity and the iad framework. abid, m. et al, scheffran, j., schneider, u. a., & ashfaq, m. (2015). farmers’ perceptions of and adaptation strategies to climate change and their determinants: the case of punjab province, pakistan. earth system dynamics, 6(1), 225–243. abid, m., ashfaq, m., hassan, s., & fatima, n. (2011). a resource use efficiency analysis of small bt cotton farmers in punjab, pakistan. pak. j. agric. sci, 48(1), 65–71. adger, w. n., agrawala, s., mirza, m. m. q., conde, c., o’brien, k., pulhin, j., et al. (2007). assessment of adaptation practices, options, constraints and capacity. climate change, 717–743. ahmed, i., ur rahman, m. h., ahmed, s., hussain, j., ullah, a., & judge, j. (2018). assessing the impact of climate variability on maize using simulation modeling under semi-arid environment of punjab, pakistan. environmental science and pollution research, 25(28), 28413–28430. armitage, d. (2005). adaptive capacity and community-based natural resource management. environmental management, 35(6), 703–715. badar, h., ghafoor, a., & adil, s. a. (2007). factors affecting agricultural production of punjab (pakistan). pak. j. agri. sci, 44(3). banuri, h., alam, m., khan, s., manzoor, j., ali, b., khan, y., et al. (2012). an android runtime security policy enforcement framework. personal and ubiquitous computing, 16(6), 631–641. benson, d., jordan, a., & smith, l. (2013). is environmental management really more collaborative? a comparative analysis of putative ‘paradigm shifts’ in europe, australia, and the united states. environment and planning a, 45(7), 1695–1712. berkes, f. (2002). back to the future: ecosystem dynamics and local knowledge. understanding transformation in human and natural systems. http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 96 university of bamberg press bodin, ö., crona, b., & ernstson, h. (2006). social networks in natural resource management: what is there to learn from a structural perspective? ecology and society, 11(2). chaffin, b. c., gosnell, h., & cosens, b. a. (2014). a decade of adaptive governance scholarship: synthesis and future directions. ecology and society, 19(3). cole, d. h. (2014). formal institutions and the iad framework: bringing the law back in. available at ssrn 2471040. corfee-morlot, j., kamal-chaoui, l., donovan, m. g., cochran, i., robert, a., & teasdale, p.-j. (2009). cities, climate change and multilevel governance. cosens, b., gunderson, l., allen, c., & benson, m. (2014). identifying legal, ecological and governance obstacles, and opportunities for adapting to climate change. sustainability, 6(4), 2338–2356. de vaus. (2002). surveys in social research (5th ed.). routledge. dietz, k. (2007). vulnerabilität und anpassung gegenüber klimawandel ansatzpunkte für eine multi-level-governanceanalyse aus der perspektive der problemkonstitution. in multi-level-governance (pp. 161–188). nomos verlagsgesellschaft mbh & co. kg. dietz, t., ostrom, e., & stern, p. c. (2003). the struggle to govern the commons. science, 302(5652), 1907–1912. fleischhauer, m., & bornefeld, b. (2006). klimawandel und raumplanung. raumforschung und raumordnung, 64(3), 161–171. folke, c., carpenter, s., elmqvist, t., gunderson, l., holling, c. s., & walker, b. (2002). resilience and sustainable development: building adaptive capacity in a world of transformations. ambio: a journal of the human environment, 31(5), 437–440. folke, c., hahn, t., olsson, p., & norberg, j. (2005). adaptive governance of social-ecological systems. annu. rev. environ. resour., 30, 441–473. gibson, c. c., andersson, k., ostrom, e., & shivakumar, s. (2005). the samaritan’s dilemma: the political economy of development aid. oxford university press on demand. hansen, j. w., & jones, j. w. (2000). scaling-up crop models for climate variability applications. agricultural systems, 65(1), 43–72. doi:doi 10.1016/s0308-521x(00)00025-1 khan, s. a. (2012). binding constraints: the role of the army in pakistan’s economic under-development. lebel, l., anderies, j. m., campbell, b., folke, c., hatfield-dodds, s., hughes, t. p., & wilson, j. (2006). governance and the capacity to manage resilience in regional social-ecological systems. ecology and society, 11(1). lemos, m. c., & agrawal, a. (2006). environmental governance, governance annual reviews environment resources. annual reviews, school of natural resources and environment, university of michigan, 298–303. mcginnis, m. d. (2011). an introduction to iad and the language of the ostrom workshop: a simple guide to a complex framework. policy studies journal, 39(1), 169–183. meadowcroft, j. (2009). climate change governance. the world bank. mustafa, z. (2011). climate change and its impact with special focus in pakistan. in pakistan engineering congress, symposium (vol. 33, p. 290). lahore. national climate change policy (2012). national climate change policy. policy document, ministry of climate change, government of pakistan. nawaz khan, a. (2010). chapter 10 climate change adaptation and disaster risk reduction in pakistan. in climate change adaptation and disaster risk reduction: an asian perspective (pp. 197–215). emerald group publishing limited. nigussie, z., tsunekawa, a., haregeweyn, n., adgo, e., cochrane, l., floquet, a., & abele, s. (2018). applying ostrom’s institutional analysis and development framework to soil and water conservation activities in north-western ethiopia. land use policy, 71, 1–10. nomman, a. m., & schmitz, m. (2011). economic assessment of the impact of climate change on the agriculture of pakistan. business and economic horizons, 4(1), 1–12. olsson, p., gunderson, l. h., carpenter, s. r., ryan, p., lebel, l., folke, c., & holling, c. s. (2006). shooting the rapids: navigating transitions to adaptive governance of social-ecological systems. ecology and society, 11(1). ostrom, e. (2005). understanding institutional diversity: princeton univ pr. ostrom, e. (2007). a diagnostic approach for going beyond panaceas. proceedings of the national academy of sciences, 104(39), 15181–15187. pbs. (n.d.). pakistan bureau of statistics: agricultural statistics of pakistan. government of pakistan, statistics division, pakistan bureau of statistics, available at: http://www.pbs.gov.pk/content/ agricultural-statistics-pakistan. last access: 1 august 2014), 2011. pervin, m., sultana, s., phirum, a., camara, i. f., nzau, v. m., phonnasane, v., et al. (2013). a framework for mainstreaming climate resilience into development planning. international institute for environment and development: london, uk. puppim de oliveira, jose a. (2019). intergovernmental relations for environmental governance: cases of solid waste management and climate change in two malaysian states. journal of environmental management, 233, 481–488. rasul, g., mahmood, a., sadiq, a., & khan, s. i. (2012). vulnerability of the indus delta to climate change in pakistan. pakistan journal of meteorology, 8(16). rijke, j., brown, r., zevenbergen, c., ashley, r., farrelly, m., morison, p., & van herk, s. (2012). fit-for-purpose governance: a framework to make adaptive governance operational. environmental science & policy, 22, 73–84. http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 97 university of bamberg press salik, k. m., ishfaq, s., saeed, f., noel, e., & syed, q. (2015). pakistan: country situation assessment. sustainable development policy institute. walker, b., holling, c. s., carpenter, s. r., & kinzig, a. (2004). resilience, adaptability and transformability in social– ecological systems. ecology and society, 9(2). weber, e. p., & khademian, a. m. (2008). wicked problems, knowledge challenges, and collaborative capacity builders in network settings. public administration review, 68(2), 334–349. zia, a. (2013). post-kyoto climate governance: confronting the politics of scale, ideology and knowledge. routledge. zia, a., kauffman, s., koliba, c., beckage, b., vattay, g., bomblies, a. (2014) from the habit of control to institutional enablement: re-envisioning the governance of social-ecological systems from the perspective of complexity sciences. complexity, governance and networks. 1(1): 79-88. doi: 10.7564/14-cgn4 http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 98 university of bamberg press appendix a respondents’ profiles responden t id respondents' responsibilities/ roles respondents’ organization 1 inspector general forests ministry of climate change 2 director general environment ministry of climate change 3 charmain and member of board of governance sustainable development policy institute 4 professor of policy study and sustainable development center for policy study, comsats university, islamabad 5 head of department center for climate research and development 6 senior researcher for agriculture and climate change global chang impact studies center 7 director agronomic and an active member for climate negotiation and policies in punjab ayub agriculture research institute, faisalabad 8 phd research student working on climate adaptation and agriculture sector university of agriculture, faisalabad 9 regional director for agriculture extension department agriculture extension department in faisalabad 10 farmer’s community(8interviews) farmers in punjab 11 director working on climate adaptation strategies helvetas swiss intercooperation organization, peshawar, pakistan 12 professor of environmental sciences university of peshawar 13 deputy director and involve in framing climate policy in the kpk center for climate change, peshawar 14 professor of agriculture and climate change agriculture university peshawar 15 climate change policy researcher pakistan forest institute, peshawar 16 director dealing with floods and rescues provincial disaster management authority, kpk 17 farmer’s community(7 interviews) farmers in the kpk http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 99 university of bamberg press appendix b members for advisory committee source: climate change cell, kpk http://dx.doi.org/10.20377/cgn-68 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 81-100 doi: http://dx.doi.org/10.20377/cgn-68 100 university of bamberg press appendix c members for implementation committee source: climate change cell, kpk http://dx.doi.org/10.20377/cgn-68 organizational choice and key questions for forming partnerships complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 109-120 doi: http://dx.doi.org/10.20377/cgn-91 university of bamberg press organizational choice and key questions for forming partnerships author: luke fowler boise state university, school of public service, usa email: lukefowler@boisestate.edu partnerships receive significant attention in public administration scholarship, with the mass of this literature focusing on whether partnerships work, how to make them work, or how they fit into existing institutions (provan and milward, 2001; vigoda, 2002; mcguire, 2006; thomson and perry, 2006; andrews and entwistle, 2010; mcquaid, 2010; o’toole, 2015). although partnership has been used variously by different scholars, in general, partnerships refer to formal arrangements between two or more organizations that are characterized by defined responsibilities, obligations, and/or governance structure, as compared to other forms of cooperative behaviors which may be more informal, unorganized, or involve few obligations. in general, existing scholarship on partnership formation argues that partnerships are a function of resource-exchanges, available partners, or fragmented authorities, and assumes a pragmatic managerial approach to these arrangements (grady and chen, 2006; feiock and scholz, 2009). however, scholarship is limited in linking these mechanisms together and explaining how organizations go from isolated and autonomous to integrated and interdependent. as such, it is difficult to determine how initial decisions in the partnership process eventually lead to success or failure in collaboration. furthermore, much of this scholarship is written with a solely academic audience in mind, making difficult for practitioners, non-academics, or non-subject area experts to consume. to remedy this, we use cohen, march, and olsen’s (1972) garbage can theory (gct) of organizational choice as a guiding framework to identify key issues that affect partnerships formation and tie this disjointed set of literature together. we then synthesize these issues into three key questions that can be operationalized by practitioners: 1) is there a problem that cannot be managed unilaterally?; 2) what new capacities are needed?; and 3) what partnership opportunities are there? from this perspective, forming partnerships unfolds in organized anarchies, where decision-makers must sort through ambiguous problems, solutions, and participants in order to determine if partnership is the right choice for their organization. in general, the purpose of this discussion is to identify and examine key issues that likely affect partnership choices made by practitioners and that can provide guidance to those who are considering engaging in collaboration or partnership. finally, we discuss links between partnership formation and broader understandings of collaborations and networks. keywords: partnership formation; collaboration; organizational choice partnerships as organizational choice in recent decades, understanding the why’s and how’s of managing public services through multiple linked organizations has been at the center of the public administration research agenda. contemporary literature in this area mainly focuses on inter-organizational ties as an independent variable and implications for organizational performance, management strategies, and normative questions for the administrative state. on the other hand, scholarship focusing on partnership as a dependent variable argue for pragmatic bureaucrats that seek resource exchanges and work with who is available (graddy and chen, 2006; lee, lee, and feiock, 2012b; park and rethemeyer, 2014). in general, scholarship contends “there may be good political reasons to add actors to implementation arrangements to reduce threats or increase support for programs…or as part of an effort to distance state-centric institutions from contentious policy issues” (o’toole, 2015, p. 364). normatively, shifts in the intellectual approach to government’s role in society contribute some understanding as to why concepts such as hollowed-out state, joined-up service delivery, and multi-dimensional governance have emerged in practice (kettl, 2006; osborne, 2010; hill and hupe, 2014). as a whole, scholarship is cautiously optimistic about how partnerships, collaboration, and networks affect public services (mcguire, 2006; thomson and perry, 2006; mcguire and agranoff, 2011; o’toole, 1997, 2015). however, a common criticism of scholarship on both partnerships and networks is a lack of theoretical frameworks to integrate the array of nuanced findings (o’toole, 2015; kapucu, hu, and khosa, 2017). research on partnership formation suffers from the same limitation; although there is some 109 mailto:lukefowler@boisestate.edu 110 university of bamberg press existing scholarship that takes initial steps in this direction (graddy and chen, 2006; thomson and perry, 2006). one potential solution to this problem is to examine partnership formation as a special case of organizational choice, where distinct institutional settings shape how decision-makers interpret their responsibilities and actions (cohen, march, and olsen, 1972). as social relationships are a key component of collaboration, partnership formation is likely similar to other organizational choices made in organized anarchies, where there is a lack of linear logic or coherent structure. while hierarchy and markets surround public agencies as organizational principles, partnerships are principally driven by the need of an alternative to more conventional mechanisms by which public agencies interact with other organizations to solve public problems. on one hand, thinking about organizational choice in this way is in keeping with the environment in which partnership choices are made, which tend to be defined by complex administrative situations that function outside of conventional hierarchies (agranoff and mcguire, 2001; o’toole, 2015). on the other hand, it is also recognizes that partnerships are steeped in ambiguity (i.e., different and often competing ways of thinking about the same phenomena) (zahariadis, 2014). in other words, the pros and cons of partnerships can be thought of in a lot of different ways and there are no clear right or wrong answers, which influences how policy actors approach collaborating with others. for example, boswell and rodrigues (2016) find that cooperative relationships between local and national organizations are influenced by how interactions between policy, politics, and problems shape the way policy actors view the context of implementation. suffice it to say, more information may provide clarity to the questions underlying partnership choices, but it rarely provides answers. while there are alternative frameworks to examine organizational choices, the basic ideas of gct provide theoretical grounding to the pragmatic nature of partnerships and the unique circumstances that are required for them to materialize (cohen, march, olsen, 1972; bendor, moe, and shotts, 2001; olsen, 2001). as such, gct provides a way to integrate disjointed scholarship that examines both the systematic factors that increase the likelihood of partnership formation and the serendipitous nature of partnerships that appear to occur as much by chance as by design by focusing on a few central structural components that unpack the complex decisions that practitioners make. from this perspective, organizational decisions result from “a collection of choices looking for problems, issues and feelings looking for decision situations in which they might be aired, solutions looking for issues to which they might be the answer, and decision makers looking for work” (cohen, march, and olsen, 1972, p. 2). as such, organizational decisions are a function of several relatively independent streams related to problems, solutions, participants, and choice opportunities, which move through a primordial soup waiting to be coupled together at opportune times (cohen, march, and olsen, 1972; kingdom, 1995; bendor, moe, and shotts, 2001; olsen, 2001; zahariadis, 2014; herweg, zahariadis, and zohlnhofer, 2018). while there are many nuances, existing scholarship suggests these apply to partnership in the form of: problems, related to complex policy problems that may require collective action; solutions, related to needed capacities to overcome organizational limitations; and participants, related to organizations available for partnerships (mcguire, 2006; thomson and perry, 2006; mcguire and agranoff, 2011; o’toole, 2015). while highly interrelated, these represent distinct factors affecting partnership formation decisions, where the many ways in which managers can think about these problems, solutions, and participants cause ambiguity. in organized anarchies, ambiguity arises from three properties: problematic preferences, unclear technology, and fluid participation (cohen, march, and olsen, 1972; ostrom, 1998, 2000; bendor, moe, and shotts, 2001; olsen, 2001; peters, 2005; sager and rielle, 2013). specific to partnerships, ambiguity is caused by vague organizational and/or shared goals, uncertain organizational capacities, shifting availability in partners, or unclear collaboration technology (i.e., processes and techniques for collective action). furthermore, when partnerships include multiple organizations and become structurally more like networks, ambiguity increases as decisions parameters become more complex (huxham and vangen, complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 109-120 doi: http://dx.doi.org/10.20377/cgn-91 111 university of bamberg press 2000; teisman and klijn, 2002; kelsey and spanjers, 2004; thomson and perry, 2006; noordegraaf, 2007; stazyk and goerdel, 2011; davis and stazyk, 2015). additionally, there is an important temporal dimension that exists and “who pays attention to what and when is critical” (zaharidias, 2014, p. 68). often, problem parameters are shifting, causal pathways between solutions and results are unclear, and distinguishing between relevant and irrelevant information strains the limits of information processing capacity. thus, at opportune times, decision-makers find they at least partially comprehend an incomprehensible world, which allows them to make choices that are otherwise bogged down in ambiguity and uncertainty. in practice, managers act pragmatically in responding to both internal and external organizational environments in order to form inter-organizational ties. otherwise, partnerships fail to materialize. although the specific purposes for which actors design and implement goaldirected partnerships or networks likely reduces ambiguity compared to the serendipitous variety, practitioners are always grappling with how to forge concrete ideas from abstract circumstances (provan and lemaire, 2012). importantly, this creates a distinctive practical nature to partnerships, where they occur as much by chance as by design (bendor, moe, and shotts, 2001; olsen, 2001; sager and rielle, 2013; o’toole, 2015). as such, managers sift through ambiguous conditions to clarify what purpose partnership might serve, how it can help overcome organizational limitations, and what actions it would require. although these issues are complex, ambiguity in these streams can be effectively summed up by three key questions: 1) is there a problem that requires partnership?; 2) what new capacities are needed?; and, 3) what partnership opportunities are available? identifying literature while collaboration, partnership, and network (as well as a litany of other terms) are highly interrelated, they represent distinct academic concepts (lecy, mergel, and schmitz, 2014; o’toole, 2015; kapucu, hu, and khosa, 2017). as research in this area has increased, scholars have emphasized the need for consistency in terminology in order to clarify concepts. however, these efforts have been met with some resistance, as some scholars still tend to use terms interchangeably or alternatively define concepts. in turn, the lack of consistent terminology and conceptual applications creates barriers to comparing and compiling lessons from the literature, or even in identifying extant studies that are applicable (mcnamara, 2012; lecy, mergel, and schmitz, 2014; o’toole, 2015; kapucu, hu, and khosa, 2017). to this end, rather than attempt to systematically review an inconsistent and fluid body of literature, we identified seven review articles published in public administration review between 2001 and 2015 that serve as a guide through the literature. additionally, we also rely on two recent articles that provide a meta-analysis of the network research in public administration (see table 1). while there may be other studies in this area that are not included, we believe that these articles are generally representative of a set of core concepts, arguments, and factors that summarize the literature on partnership formation. review articles meta-analysis • agranoff (2006) • bryson, crosby, and stone (2006) • mcguire (2006) • o’toole (2015) • provan and lemaire (2012) • provan and milward (2001) • thomson and perry (2006) • lecy, mergel, and schmitz (2014) • kapucu, hu, and khosa (2017) table 1 articles used to guide literature review complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 109-120 doi: http://dx.doi.org/10.20377/cgn-91 112 university of bamberg press after identifying core concepts that consistently appeared, we grouped those into three categories as they related to factors that drive partnership choices in regards to problems, solutions, and participants. then, we used a snowball technique (i.e., tracing cited sources) to sift through the source material for the reviews and meta-analysis to drill down to the core research questions examined. we identified 54 journal articles and book chapters based on these steps. of those, 19 addressed aspects of problems, 26 solutions, and 19 participants, with seven addressing more than one. in general, this would indicate that solutions seem to be a more popular point of inquiry in comparison to problems and participants. additionally, it appears the vast majority of scholarship only focuses on one component, which would further suggest a need to account for how problems, solution, and participants interact during partnership formation. finally, after reviewing these articles, we distilled three key questions that serve as overarching points of inquiry that we use to frame our discussion of the literature. in addition to our broader review, we find two notable epistemological trends that appear across much of this body of scholarship. first, the unit of analysis tends to be the organization, rather than individuals. for instance, there is a focus on identifying systematic factors that drive partnership formation, which assumes there are a series of environmental or organizational factors that increase the probability of partnerships. on one hand, this overlooks the potential micro-level social interactions that play a role in partnerships. on the other hand, it lends itself to thinking about partnership formation as a function of organizational choice, as we do here. second, most scholars focus on partnerships that successfully form, as compared to partnership that were proposed but never came to fruition. while the latter may be difficult to identify, it may also hold some potentially significant insight to why partnerships form, especially where administrative logic and contextual factors would predict a partnership, but in reality, one never manifested (fowler, 2019). in interest of conceptual clarity, we bound our discussion of partnership in two ways. first, we define partnerships here as a formal arrangement between two organizations to collaborate on a specified set of functions based on explicit or implicit obligations and responsibilities and well-defined governance structures and/or organizing principles. second, in many cases, partnerships are the building blocks of formal governance networks. specifically, networks are a collection of interdependent organizations tied together with both formal and informal arrangements that may be based on unclear organizing principles, but that involve a set of shared interests, functions, or goals (thomson and perry, 2006; mcnamara, 2012; lecy, mergel, and schmitz, 2014; kapucu, hu, and khosa, 2017; o’toole, 2015). while previous scholars argue that partnerships are not always networks and networks are not always partnerships, these differences are more clearly identifiable during service delivery and policy governance than during the formation of collaborative arrangements (lowndes and skelcher, 1998; teisman and klijn, 2002). as such, existing research tends to use scholarship on partnership and network formation to inform one another (graddy and chen, 2006; lecy, mergel, and schmitz, 2014; kapucu, hu, and khosa, 2017). therefore, in the context of our discussion, partnerships and networks are so highly interrelated that any analysis of the former must consider the latter. consequently, network scholarship permeates our examination of partnerships in many ways and provides a broader perspective on how collaborative arrangements evolve with multiple organizations. key questions question 1: is there a problem that requires partnership? problem ambiguity occurs when determining if complex problems or organizational goals are manageable unilaterally or through existing inter-organizational mechanisms. if not, some form of partnership is essential to success. consequently, most scholars identify two distinctive features of partnerships used for public service delivery. first, in principle, partnerships are organized around collective action problems or shared goals, in one form or another (thomson and perry, 2006; feiock, complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 109-120 doi: http://dx.doi.org/10.20377/cgn-91 113 university of bamberg press 2013). typically, these occur when complex problems cross legal or geographic jurisdictions and require efforts from more than one organization, but also result from limited organizational capacities (bryson, crosby, and stone, 2006; thomson and perry, 2006; feiock and scholz, 2009). these complex problems take on many forms and may range from wicked problems to spillover effects (lozano and walles, 2007; feiock and scholz, 2009; head and alford, 2015). climate change is one of the more prominent illustrations of the need for partnerships, where both causes and effects transcend jurisdictions and require collective action (verweij, et al., 2006; gore, 2010). local economic development is another example. although individual local agencies can pursue development goals unilaterally, efforts are more effective if coordinated at a regional-level across multiple stakeholder organizations (gordon, 2007; feiock, steinacker, and park, 2009; lee, lee, and feiock, 2012a, 2012b). second, these problems cannot be managed through conventional inter-organizational relationships (e.g., hierarchies or markets) and require alternative mechanisms (agranoff and mcguire, 2001; o’toole, 2015). complex problems create dilemmas for public sector organizations where they can only exercise control over a portion of a problem that is insufficient to address the problem unilaterally (feiock and scholz, 2009; feiock, 2013). however, fragmented authorities create institutional barriers, transaction costs, and power struggles that limit the utility of existing inter-governmental ties and make alternative mechanisms of organization more utilitarian, adaptable, or sustainable. these alternative mechanisms (e.g., contracting services, public-private partnership) reduce obstacles to public agencies working with other organizations and increase their control over or capacity to manage problems (agranoff and mcguire, 2001; ling, 2002; thurmaier and wood, 2002; keast, mandell, brown, and woolcock, 2004; feiock and scholz, 2009). as such, practitioners should ask: 1) is the problem or goal too complex to manage alone? and 2) can relationships with other organizations be managed through conventional mechanisms? despite the collaboration fad in recent decades, not all problems require a partnership. in fact, partnerships may hinder managerial performance in some cases (longoria, 2005; hodge and greve, 2007; andrews and entwistle, 2010). practitioners must understand the problem they want to solve, as well as the existing institutional mechanisms that might support those efforts. only when they find that existing organizational capacities and inter-organizational relationships are inadequate to address a problem is partnership the right answer (longoria, 2005). for example, lundin (2007) finds that for simple policies that readily fit within local government missions and capacities, cooperation has no impact on policies outcomes but creates new transaction costs in coordinating organizations. on the other hand, implementation of more complex policies improves significantly when national and local governments cooperate. however, this requires managers to develop new mechanisms to cooperate in implementation, so partnerships are only an effective strategy under certain circumstances (lundin, 2007). as such, it is incumbent on managers to develop a sophisticated understanding of the circumstances surrounding problems and whether they are manageable either unilaterally or through existing intergovernmental mechanisms. question 2: what new capacities are needed? solution ambiguity occurs when determining what additional resources or capacities are needed to manage complex problems or achieve goals. in most cases, limitations on capacities boil down to political, economic, or institutional constraints that hamper the effectiveness of public sector agencies (bryson, crosby, and stone, 2006; feiock and scholz, 2009; andrews and entwistle, 2010). capacity is a nuanced concept with numerous dimensions that may be useful for public agencies, including procedural flexibility, access to special resources or other organizations, infrastructure, operational support, or leadership abilities. scholars typically discuss these issues within the context of resource exchanges or dependencies, which are also the most prominent explanations for partnerships. from this perspective, organizations seek ties with other organizations in order to gain access to new resources that would complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 109-120 doi: http://dx.doi.org/10.20377/cgn-91 114 university of bamberg press otherwise be unavailable (graddy and chen, 2006; lee, lee, and feiock, 2012b; park and rethemeyer, 2014). commonly, these are thought of as tangible resources, such as funding, manpower, or infrastructure (graddy and chen, 2006; provan and huang, 2012), but intangible resources, such as credibility, influence, or information, are also valued in partnership exchanges (andrew, 2009; feiock, lee, park, and lee, 2010; provan and huang, 2012). in general, resources can be thought of as a special category of capacities in as far as they are valuable to partnerships as a mechanism to address organizational limitations. however, there is not necessarily a clear distinction made between resources or capacities across the partnership scholarship as some refer to intangible resources in the same way that others refer to capacities (e.g., political capital, leadership). importantly, the character of exchanges, including which capacities or equity in distributions, affect how engaged organizations are in partnerships (park and rethemeyer, 2014). furthermore, not all dimensions may be needed or even advantageous, and capacity can be exchanged in whole or in part (fredericksen and london, 2000; boyne, 2003; brown and potoski, 2003; christensen and gazley, 2008; yang, hsieh, and li, 2009; andrews and entwistle, 2010; mcguire and silvia, 2010; andrews and boyne, 2011). generally, frameworks tend to distinguish partnership forms based on level of inter-organizational integration (mandell and steelman, 2003; bovaird, 2006; thomson and perry, 2006; mcnamara, 2012). for example, feiock and scholz (2009) place mechanisms for coping with institutional collective action dilemmas on a continuum based on whether they preserve autonomy (e.g., informal policy networks) or impose obligations (e.g., consolidated government). alternatively, teisman and klijn (2002) argues partnerships differ based on who is making decisions, and may include scenarios where goals are set by public agencies (e.g., contracting) as well as those that allow for joint decision-making. organizations may only need to integrate to the extent that they have access to the most utilitarian capacities (mandell and steelm, 2003; bovaird, 2006; thomson and perry, 2006). in practice, it is not always necessary or feasible to fully integrate two or more organizations, especially for simple service delivery functions that can be better accomplished with a contractual relationship (mandell and steelman, 2003; bovaird, 2006; mcguire, 2006). for example, a local government does not need to integrate with a garbage collection firm in order to access its expertise in garbage collection or flexibility in public service delivery. rather, a contractual agreement between a local government and garbage collection firm can provide needed capacities without creating unnecessary transaction costs. additionally, capacities can be obtained from more than one source, so multi-organizational partnerships (i.e., networks) may be a viable option as goal or problem complexity increases (provan, veazie, staten, and teufel-shone, 2005; agranoff, 2006; bryson, crosby, and stone, 2006). as such, practitioners should ask: 1) what capacities are limiting organizational efficacy? and 2) what capacities need to be gained from a partnership? not all capacities are offered from every partner, and different forms of partnership may be more or less suitable for specific problems, organizational capacities, or available partners (keast, brown, and mandell, 2007; andrews and entwistle, 2010; lofstrom, 2010). practitioners must determine which capacities are most advantageous, and which may be redundant in a partnership. for example, in an infrastructure development project, partnership with a financial services firm may provide financial resources as well as expertise on project or contract management. however, a financial services firm may not be capable of construction, so another firm may need to provide construction services. on the other hand, another public agency without specific capacities or resources in these areas would create more complexity without providing specific benefits (koppenjan, 2005; hodge and greve, 2007; garvin and bosso, 2008; koppenjan and enserink, 2009). nevertheless, it is incumbent on practitioners to sort through ambiguous advantages (and disadvantages) of capacities in relation to their own organizational limitations. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 109-120 doi: http://dx.doi.org/10.20377/cgn-91 115 university of bamberg press question 3: what partnership opportunities are available? participant ambiguity occurs when determining what partnership opportunities are available and viable. opportunities that emerge from unique contexts are an important driver of partnership formation and can be defined by two parameters: 1) availability of partners and 2) barriers to partnership. first, availability of partners is one of the strongest predictors of partnerships, as well as size and scope of networks. in the most practical sense, partnerships cannot form if organizations are not available to collaborate with or are unaware of each other’s availability. while the myriad of organizations now participating in public service delivery creates plenty of potential partners, there are numerous limitations to availability, including shared goals or problems, resources, collaborative capacity, partnership density, social clustering, organizational type, and geography (brinkerhoff, 2002a; graddy and chen, 2006; weber, lovrich, and gaffney, 2007; jasny, 2012; lee, lee, and feiock, 2012b; hill and hupe, 2014; o’toole, 2015). additionally, too many potential partners can lead to information overload, where managers cannot fully analyze every option (brinkerhoff, 2002b; ramiro, 2009; reynaers, 2013). importantly, for a partner to be “available,” managers have to be aware of the viability of potential partnership with that organization. second, there are a number of barriers to partnership with any organization. in the most basic sense, these include the “operational, performance, or legal barriers that prevent the next action step…in converting solutions into policy energy, assessing internal effectiveness, surmounting the inevitable process blockages, mission drift, and so on” (mcguire and agranoff, 2011, p. 265). internal capacities are also among these barriers, where issues such as structures, processes, and social norms related to risk management, problem-solving, shortand long-term planning, accountability, decision-making, and performance management affect the ability of organizations to effectively engage in collaborations (thomson and perry, 2006; weber and khademian, 2008; o’toole, 2015). furthermore, interorganizational ties create a tremendous amount of risk, where investing resources into a failed venture can be devastating to organizational missions, reputation, and political capital (akintoye, beck, and hardcastle, 2003; mcguire, 2006; feiock, lee, park, and lee, 2010). for example, cross-sectoral partnerships run inherent risks of principal-agent problems due to mission conflicts (gazley and brudney, 2007; hodge and greve, 2007; acar, guo, and yang, 2008). as a result, public agencies may find themselves investing tremendous resources into a partnership that may eventually collapse. barriers can also emerge from existing partnerships or networks, which create preconditions to facilitate (or not) further partnerships (fischer, ingold, sciarini, and varone, 2012). some scholars argue that a balance between loose structures and a foundation that embeds partnerships is necessary to attract new partners and expand networks (laven, kyrmkowski, ventriss, manning, and mitchell, 2010; robins, bates, and pattison, 2011; o’toole, 2015). when partners are too loosely associated, issues such as uncoordinated goals, defensive strategies, or unpredictable demands emerge (robins, bates, and pattison, 2011; o’toole, 2015). on the other hand, when partnerships are too embedded, organizations lose autonomy and there are high hurdles for new partners (o’toole, 2015). additionally, social clustering and network density may help maintain commitments between partners and increase desirability of partnerships (graddy and chen, 2006; jasny, 2012; lee, lee, and feiock, 2012b). however, when partnership mulitplexity (i.e., separate ties between organizations) increases too much, coordination costs overburden transactions and lead to collaborative inertia. essentially, collaborative capacity can only be stretched so far, and eventually maintaining ties creates more burdens than benefits (provan and milward, 2001; mcguire and agranoff, 2011; provan and lemaire, 2012). as such, practitioners should ask: 1) what other organizations are available for partnership? and 2) what barriers are there to those partnerships? public-sector agencies are surrounded by other public, private, and nonprofit organizations, but some are in a better position to engage in partnership than others (graddy and chen, 2006; o’toole, 2015). practitioners must identify which other organizations have the necessary attributes to make good partners, which includes the potential barriers to partnership (brinkerhoff, 2002a; mcguire, 2006; feiock, lee, park, and lee, 2010; fischer, ingold, sciarini, and complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 109-120 doi: http://dx.doi.org/10.20377/cgn-91 116 university of bamberg press varone, 2012; o’toole, 2015). although some organizations may appear available, they may have bad partnership habits or be over-extended with too many partners, which limits their ability to effectively collaborate (weber, lovrich, and gaffney, 2007; weber and khademian, 2008; mcguire and agranoff, 2011; provan and lemaire, 2012). however, the balance between availability and barriers is tied to the unique circumstances of partnership goals. for example, a joint venture for urban renewal creates significant limitations for identifying potential partners capable of overcoming complex barriers related to the project (hodge and greve, 2007; gordon, 2007; feiock and scholz, 2009). nevertheless, partnership opportunities are unique to every context, and practitioners should consider which organizations are best positioned for success. discussion and conclusions a key goal of research on partnerships should be to guide practitioners as they sift through ambiguous circumstances surrounding what purpose they serve, how they may enhance organizational efficacy, and what they look like in practice. while more information may not always eliminate the ambiguity in an inherently chaotic and complex environment, practitioners can better address challenges when more information is available (zahariadis, 2014). therefore, scholarship plays an integral role in helping practitioners understand the connections between their organizational missions and shared goals, capacities that improve specific policy outcomes, identifying organizations that make good partners, and processes by which partnerships form, among other things. while there are extensive analyses of these points in existing scholarship, much of it is inaccessible to practitioners. in this vein, one of our goals here is to synthesize theory and findings into a form that practitioners, non-academics, or non-subject area experts can consume, where many scholars present the same concepts with a solely academic audience in mind (bushouse, et al., 2011). while asking the key questions identified above may not lead to better partnerships or stave off disaster, it may help practitioners understand potential pitfalls from the beginning. from a theoretical perspective, using gct to synthesize key questions that actors may ask when making choices provides both practitioners and academics some guidance on how to think about partnership in the public service as organizational choice. the assumptions of gct largely align with the reality of forming partnerships in complex administrative environments that lack clear hierarchies and that are steeped in ambiguous circumstances. furthermore, gct also helps capture the distinctive practical nature of partnerships, where they succeed or fail as much by chance as by design. while existing literature tends to focus on goal-directed partnerships or networks, serendipitous partnerships are a common phenomenon in practice and play an important role in collaborative governance (provan and kenis, 2008; provan and lemaire, 2012). thus, developing analytical frameworks that account for components of decision-making that occur when faced with ambiguity and unanticipated opportunities provides new depth to scholarship on how and why partnerships and networks form. still, this analysis may be over-simplistic in many ways and may not fully account for all the nuances in partnership formation. while we draw our inferences from key literature that is consistently cited, this body of scholarship lacks coherence along several dimensions, so it is difficult to integrate competing epistemological and theoretical perspectives into a single framework of partnership formation (o’toole, 2015; kapucu, hu, and khosa, 2017). despite the inherent challenges, partnership formation is still an important point of inquiry. as policy and governance scholars argue, choices made during formulation are fundamental in understanding policy, including its intended purpose and the relationship between different organizations in its implementation (nakamura, 1987; hill and hupe, 2014). in that same vein, understanding choices made when forming partnerships can provide important insights into their design, mechanisms, and intended purposes. for instance, although both are public-private partnerships, service delivery partnerships (i.e., partners each contribute particular expertise to provide public services) and financing complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 109-120 doi: http://dx.doi.org/10.20377/cgn-91 117 university of bamberg press partnerships (i.e., private partners provide access to capital financing for public agencies) are built on different interpretations of problems, solutions, and participants. consequently, they require different managerial tools and strategies to be successful (bovaird, 2006). for both practitioners and academics, expanding our understanding of these connections can support more effective uses of scholarship in informing practical decisions about partnerships (mcguire and agranoff, 2011; o’toole, 2015). as such, further developing theoretical frameworks for partnership formation can provide deeper insights into the nuances of collaboration, and practitioner-oriented scholarship can guide decision-makers to choices that are more likely to yield success (o’toole, 1997, 2015). partnership formation is only a small part of public administration’s research agenda into collaborative governance (o’toole, 1997, 2015). scholarship focused on other aspects of partnerships, collaboration, networks, and governance provided a significant portion of the insights and observations above that account for which factors should be considered in partnership decisions. understanding what leads to better collaborative outcomes helps reduce the ambiguity around deciding what problems require partnership, what solutions partnerships can offer to organizational limitations, and what partnership opportunities are available. however, this is a two-way street, and partnership formation offers insight into managing collaborative relationships as well. although performance management, collaborative technologies, and inter-organizational accountability are at the center of the partnership research agenda, scholarship does not clearly account for how partnership formation affects these issues (mcguire, 2006; mcguire and agranoff, 2011; o’toole, 2015). specifically, scholarship does not always connect partnership formation to program outcomes, managerial strategies and tools, or network management. furthermore, there are implications for normative questions, including why public management is becoming more reliant on collaborations and how this fits into democratic institutions. references acar, m., c. guo, and k. yang. 2008. accountability when hierarchical authority is absent. american review of public administration 38(1): 3-23. agranoff, r. 2006. inside collaborative networks: ten lessons for public managers. public administration review 66(s): 56-65. agranoff, r. and m. mcguire. 2001. american federalism and the search for models of management. public administration review 61(6): 671-681. andrews, r. and g. boyne. 2011. corporate capacity and public service performance. public administration 89(3): 894-908. andrews, r. and t. entwistle. 2010. does cross-sectoral partnership deliver? an empirical exploration of public service effectiveness, efficiency, and equity. journal of public administration research & theory 20(3): 679-701. bendor, j., t.m. moe, and k.w. shotts. 2001. recycling the garbage can: an assessment of the research program. american political science review 95(1): 169-190. boswell, c. and e. rodrigues. 2016. policies, politics, and organisational problems: multiple streams and the implementation of targets in uk government. policy & politics 44(4). bovaird, t. 2006. developing new forms of partnership with the ‘market’ in the procurement of public services. public administration 84(1): 81-102. boyne, g.a. 2003. sources of public service improvements: a critical review and research agenda. journal of public administration research & theory 13(3): 367-394. brinkerhoff, j.m. 2002. assessing and improving partnership relationships and outcomes: a proposed framework. evaluation and program planning 25(3): 215-231. brinkerhoff, j.m. 2002. global public policy, partnership, and the case of the world commission on dams. public administration review 62(3): 324-336. brown, t.l. and m. potoski. 2003. contract-management capacity in municipal and county governments. public administration review 63(2): 153-164. bryson, j.m., b.c. crosby, and m.m. stone. 2006. the design and implementation of cross-sector collaborations: propositions from the literature. public administration review 66(s): 44-55. bushouse, b.k., w.s. jacobson, k.t. lambright, j.l. llorens, r.s. morse, and o. poocharoen. 2011. crossing the divide: building bridges between public administration practitioners and scholars. journal of public administration research & theory 21(s1):i99-i112. carpenter, d.p. and g.a. krause. 2012. reputation and public administration. public administration review 72(1): 26-32. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 109-120 doi: http://dx.doi.org/10.20377/cgn-91 118 university of bamberg press christensen, r.k. and b. gazley. 2008. capacity for public administration: analysis of meaning and measurement. public administration & development 28(4): 265-279. cohen, m.d., j.g. march, and j.p. olsen. 1972. a garbage can model of organizational choice. administrative science quarterly 17(1): 1-25. davis, r.s. and e.c. stazyk. 2015. developing and testing a new goal taxonomy: accounting for the complexity of ambiguity and political support. journal of public administration research & theory 25(3): 751-775. feiock, r.c. 2013. the institutional collective action framework. policy studies journal 41(3): 397-425. feiock, r.c. and j.t. scholz. 2009. self-organizing governance of institutional collective action dilemmas: an overview. in selforganizing federalism: collective mechanisms to mitigate institutional collective action dilemmas edited by r.c. feiock and j.t. scholz (pgs. 3-32). new york: cambridge university press. feiock, r.c., a. steinacker, and h.j. park. 2009. institutional collective action and economic development joint ventures. public administration review 69(2): 256-270. fischer, m., ingold, k., sciarini, p., and varone, f. 2012. impacts of market liberalization on regulatory network: a longitudinal analysis of the swiss telecommunications sector. policy studies journal 40(3): 435-457. fowler, l. 2019. obstacles and motivators for partnership formation in a multidimensional environment. politics & policy 47(2): 267-299. fredericksen, p. and r. london. 2000. disconnect in the hollow state: the pivotal role of organizational capacity in communitybased development organizations. public administration review 60(3): 230-239. garvin, m.j. and d. bosso. 2008. assessing the effectiveness of infrastructure public-private partnership programs and projects. public works management & policy 13(2): 162-178. gazley, b. and j.l. brudney. 2007. the purpose (and perils) of government-nonprofit partnership. nonprofit and voluntary sector quarterly 36(3): 389-415. gordon, v. 2007. partners or competitors? perceptions of regional economic development cooperation in illinois. economic development quarterly 21(1): 60-78. gore, c.d. 2010. the limits and opportunities of networks: municipalities and canadian climate change policy. review of policy research 27(1): 27-46. graddy, e.a. and b. chen. 2006. influences on the size and scope of networks for social service delivery. journal of public administration research & theory 16(4): 533-552. head, b.w. and j. alford. 2015. wicked problems: implications for public policy and management. administration & society 47(6): 711-739. herweg, n., n. zahariadis, and r. zohlnhofer. 2018. the multiple stream framework: foundations, refinements, and empirical applications. in theories of the policy process, 4th ed., edited by c.m. weible and p.a. sabatier. boulder, co: westview. hill, m. and p. hupe. 2014. implementing public policy, 3rd ed. thousand oaks, ca: sage. hodge, g.a. and c. greve. 2007. public-private partnerships: an international performance review. public administration review 67(3): 545-558. huxham, c. and s. vangen. 2000. ambiguity, complexity, and dynamics in the membership of collaboration. human relations 53(6): 771-806. jasny, l. 2012. baseline models for two-mode social network data. policy studies journal 40(3): 458-491. kapucu, n., q. hu, and s. khosa. 2017. the state of network research in public administration. administration & society 49(8): 1087-1120. keast, r., k. brown, and m. mandell. 2007. getting the right mix: unpacking integration meanings and strategies. international public management journal 10(1): 9-33. keast, r., m.p. mandell, k. brown, and g. woolcock. 2004. network structures: working differently and changing expectations. public administration review 64(3): 363-371. kelsey, d. and w. spanjers. 2004. ambiguity in partnerships. the economic journal 114(497): 528-546. kettl, d.f. 2006. managing boundaries in american administration: the collaboration imperative. public administrative review 66(s): 10-19. kingdon, j. 1995. agendas, alternatives, and public policies, 2nd ed. new york: harper collins. koppenjan, j.f.m. 2005. the formation of public-private partnerships: lessons from nine transport infrastructure projects in the netherlands. public administration 83(1): 135-157. koppenjan, j.f.m. and b. enserink. 2009. public-private partnerships in urban infrastructures: reconciling private sector participation and sustainability. public administration review 69(2): 284-296. laven, d.n., d.h. kyrmkowski, c.l. ventriss, r.e. manning, and n.j. mitchell. 2010. from partnerships to networks: new approaches for measuring u.s. national heritage area effectiveness. evaluation review 34(4): 271-298. lecy, j.d., i.a. mergel, and h.p. schmitz. 2014. networks in public administration: current scholarship in review. public management review 16(5): 643-665. lee, y., i. lee, and r.c. feiock. 2012a. competitors and cooperators: a micro-level analysis of regional economic development collaboration networks. public administration review 72(2): 253-262. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 109-120 doi: http://dx.doi.org/10.20377/cgn-91 119 university of bamberg press lee, y., i. lee, and r.c. feiock. 2012b. interorganizational collaboration networks in economic development policy: an exponential random graph model analysis. policy studies journal 40(3): 547-573. ling, t. 2002. delivering joined-up government in the uk: dimensions, issues, and problems. public administration 80(4): 615642. lofstrom, m. 2010. inter-organizational collaboration projects in the public sector: a balance between integration and demarcation. international journal of health planning and management 25(2): 136-155. longoria, r.a. 2005. is inter-organizational collaboration always a good thing? journal of sociology & social welfare 32(3): 123138. lowndes, v. and c. skelcher. 1998. the dynamics of multi-organizational partnerships: an analysis of changing modes of governance. public administration 76(2): 313-333. lozano, m. and j. valles. 2007. an analysis of the implementation of an environmental management system in a local public administration. journal of environmental management 82(4): 495-511. lundin, m. 2007. when does cooperation improve public policy implementation? policy studies journal 35(4): 629 -652. mandell, m.p. and t.a. steelman. 2003. understanding what can be accomplished through interorganizational innovations: the importance of typologies, context, and management strategies. public management review 5(2): 197-224. mcguire, m. 2006. collaborative public management: assessing what we know and how we know it. public administration review 66(s): 33-43. mcguire, m. and r. agranoff. 2011. the limitations of public management networks. public administration. 89(2): 265-284. mcguire, m. and c. silvia. 2010. the effect of problem severity, managerial and organizational capacity, and agency structure on intergovernmental collaboration: evidence from local emergency management. public administration review 70(2): 279288. mcnamara, m. 2012. starting to untangle the web of cooperation, coordination, and collaboration: a framework for public managements. international journal of public administration 35(6): 389-401. mcquaid, r.w. 2010. theory of organizational partnerships: partnership advantages, dis-advantages and success factors. in the new public governance? edited by s.p. osborne, 127-148. new york: routledge. nakamura, r.t. 1987. the textbook policy process and implementation research. policy studies review 7(1): 142-154. noordegraaf, m. 2007. from “pure” to “hybrid” professionalism: present-day professionalism in ambiguous public domains. administration & society 39(6): 761-785. olsen, j.p. 2001. garbage cans, new institutionalism, and the study of politics. american political science review 95(1): 191-198. osborne, s.p. 2010. the new public governance? emerging perspectives on the theory and practice of public governance. new york: routledge. ostrom, e. 1998. a behavioral approach to the rational choice theory of collective action. american political science review 92(1): 1-22. ostrom, e. 2000. collective action and the evolution of social norms. journal of economic perspectives 14(3): 137-158. o’toole, l.j. 1997. treating networks seriously: practical and research-based agendas in public administration. public administration review 57(1): 45-52. o’toole, l.j. 2015. networks and networking: the public administrative agendas. public administration review 75(3): 361-371. park, h. and r. rethemeyer. 2014. the politics of connections: assessing the determinants of social structure in policy networks. journal of public administration research & theory 24(2):349-379. peters, b.g. 2005. governance: a garbage can perspective. in complex sovereignty: reconstituting political authority in the twentyfirst century, edited by e. grande and l.w. pauly. toronto, ontario: university of toronto press. provan, k.g. and p. kenis. 2008. modes of network governance: structure, management, and effectiveness. journal of public administration research & theory 18(2): 229-252. provan, k.g. and r.h. lemaire. 2012. core concepts and key ideas for understanding public sector organizational networks: using research to inform scholarship and practice. public administration review 72(5): 638-648. provan, k.g. and h.b. milward. 2001. do networks really work? a framework for evaluating public-sector organizational networks. public administration review 61(4): 414-423. provan, k.g., m.a. veazie, l.k. staten, and n.i. teufel-shone. 2005. the use of network analysis to strengthen community partnerships. public administration review 65(5): 603-613. ramiro, b. 2009. processing complexity in networks: a study of informal collaboration and its effect on organizational success. policy studies journal 37(3): 521-539. reynaers, a. 2014. public values in public-private partnerships. public administration review 74(1): 41-50. robins, g., l. bates, and p. pattison. 2011. network governance and environmental management: conflict and cooperation. public administration 89(4): 1293-1313. sager, f. and y. rielle. 2013. sorting through the garbage can: under what conditions do governments adopt policy programs? policy sciences 46(1): 1-21. stazyk, e.c. and h.t. goerdel. 2011. the benefits of bureaucracy: public managers’ perceptions of political support, goal ambiguity, and organizational effectiveness. journal of public administration research & theory 21(4): 645-72. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 109-120 doi: http://dx.doi.org/10.20377/cgn-91 120 university of bamberg press teisman, g.r. and e. klijn. 2002. partnership arrangements: governmental rhetoric or governance scheme? public administration review 62(2): 197-205. thomson, a.m. and j.l. perry. 2006. collaboration processes: inside the black box. public administration review 66(s): 20-32. thurmaier, k. and c. wood. 2002. interlocal agreements as overlapping social networks: picket fence regionalism in metropolitan kansas city. public administration review 62(5): 585-598. verweij, m., m. douglas, r. ellis, c. engel, f. hendriks, s. lohmann, s. ney, s. rayner, and m. thompson. 2006. clumsy solutions for a complex world: the case of climate change. public administration 84(4): 817-843. vigoda, e. 2002. from responsiveness to collaboration: governance, citizens, and the next generation of public administration. public administration review 62(5): 527-540. weber, e.p. and a.m. khademian. 2008. wicked problems, knowledge challenges, and collaborative capacity builders in network settings. public administration review 68(2): 334-349. weber, e.p., n.p. lovrich, and m.j. gaffney. 2007. assessing collaborative capacity in a multidimensional world. administration & society 39(2): 194-220. yang, k., j.y. hsieh, and t.s. li. 2009. contracting capacity and perceived contracting performance: nonlinear effects and the role of time. public administration review 69(4): 681-696. zahariadis, n. 2014. ambiguity and multiple streams. in theories of the policy process, 3rd ed., edited by p.a. sabatier and c.m. weible (pgs. 25-58). boulder, co: westview. complexity, governance & networks – vol. 6, no 1 (2020) special issue: global governance in complex times: exploring new concepts and theories on institutional complexity, p. 109-120 doi: http://dx.doi.org/10.20377/cgn-91 superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 12 university of bamberg press superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets authors: victor galaza, jon pierreb a corresponding author: stockholm resilience centre, stockholm university, sweden email: victor.galaz@su.se and global economic dynamics and the biosphere programme, royal swedish academy of sciences, sweden bdepartment of political science, göteborg university, sweden and melbourne school of government, university of melbourne, australia email: jon.pierre@gu.se increased trading with financial instruments, new actors and novel technologies are changing the nature of financial markets making trade faster, more information dense and more globalized than ever. these changes in financial markets are not incremental and linear, but transformative with the emergence of a new “machine-ecology” with intricate system behavior and new forms of systemic financial risks. we argue that the nature of these changes pose fundamentally new challenges to governance as they require policy-makers to respond to system properties characterized by not only complex causality, but also extreme connectivity (i.e. global), ultra-speed (i.e. micro-seconds) and “hyperfunctionality”. governance can fail at the system level if a subsystem performs its function to such an extreme; this could jeopardize the efficiency of the system as a whole. we elaborate in what ways governance scholars can approach these issues, and explore the types of strategies policy-makers around the world use to address these new financial risks. we conclude by pointing out what we perceive as critical research fronts in this domain. keywords: complexity; financial markets; algorithmic trade; governance; network governance 1. introduction trade in financial markets relies fundamentally on information flows. financial trade is in fundamental ways, information. financial actors such as investors not only make their living by betting on the right stock, bond or futures contract, but also (amongst other strategies) by making quick investment decisions based on the latest financial or political news or shifting expectations amongst other market actors. the speed of these information flows has become increasingly important. as computing power continues to increase exponentially, and computer algorithms become progressively sophisticated through machine learning, the capacity of financial actors to process growing amounts of information, design complex trading strategies, and conduce extremely rapid trade in bonds, stocks, commodities and foreign exchange markets (sometimes at the scale of microseconds) also surges. http://dx.doi.org/10.20377/cgn-55 mailto:victor.galaz@su.se mailto:jon.pierre@gu.se complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 13 university of bamberg press algorithmic tradei (henceforth denoted at) involves several types of automated trading strategies, each with different ways of making profit with minimal human intervention, and at extremely high speeds. at has four main strengths compared to conventional trade, according to mcgowan (2010, p. 3), gomber, arndt, lutat, & uhle (2011, 21f), and narang (2013). first, algorithms have the ability to split up large orders, and execute them in such small portions over time so that market impacts (e.g. unwanted price increases) are minimized. second, they also have the ability to text-mine different sources of information for “weak signals” – such as the movement of interest rates, very small economic fluctuations, sentiments in financial news and in social media such as in the micro-blog service twitter – and rapidly execute trade before anyone else in the market is aware of them. third, algorithms benefit from identifying and rapidly trade on differences in prices between exchanges and platforms, known as “arbitraging between spreads”. fourth, algorithms are able to place “flash orders” – that is trade based on access to market information for a matter of milliseconds, which is not yet public.ii while the profit can be extremely small for each individual trade (pennies of a us dollar for example), algorithmic trade can be executed millions of times a day, thereby generating large gains in aggregate. for example, the quantitative trading firm renaissance technologies in the u.s. is famous for its impressive 35 percent average yearly returns since 1990 (narang, 2013, p. 4). in addition, despite the economic recession, algorithmic trade is seen as one of the most lucrative businesses on wall street, and is estimated to generate approximately 15-25 billion us dollars in revenue yearly (mcgowan, 2010, p. 3f). these figures are contested however. thus, at is far from a marginal phenomenon. on the contrary, large market players such as hedge, pension, and mutual funds increasingly execute large trades through algorithmic trading. early estimates show that algorithmic trading has increased from near zero in the mid-1990s to as much as 70 percent of the trading volume in the united states (hendershott et al., 2011, p.1), 30 percent in the uk, and between 30-50 percent in the european equities markets (government office for science, 2011). at also is making fast progression into commodity derivative markets (such as futures contracts for cocoa, oil and wheat) in the u.s., russia, india and china (galaz, gars, moberg, nykvist & repinski, 2015). a number of scholars and institutions have raised numerous concerns of systemic financial risks emerging from these technologies. these concerns are about large price swings and market instability triggered by unforeseen feedback-loops, malicious or incorrectly programmed algorithms and/or incorrect input data (gos, 2011, 3ff; lenglet, 2011, 59f; chaboud, chiquoin, hjalmarsson, & vega 2014). this element of “surprise” created by complexity can be exemplified with the 2010 “flashcrash”. in the course of about 30 minutes, u.s. stock market indices, stockindex futures, options, and exchange-traded funds experienced an unexpected sudden price drop of more than 5 percent, followed by a rapid rebound (us commodity futures trading commission & us securities & exchange commission, 2010). similar unexpected, and very rapid volatility in prices for oil and gas futures have been observed the last few years (gos, 2011, p. 36). one of the most extreme examples however, must be the ultrarapid $121 billion loss (and recover) of the value of the stock market index s&p500 in april 2013. in this case, the acclaimed reason was a hijacked twitter-account of the major news company associated press (ap) sending out false messages of a terror attack in the white house, thereby triggering rapid automated sales on the market. the progression of these technologies hence not only carries with them possible benefits (such as increased liquidity, reduced volatility, or reduced transactions costs), but also wider and highly debated financial risks.iii http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 14 university of bamberg press 1.1. aims of this paper in this paper, we argue that at pose a number of new interesting insights into the governance of complexity, but complexity of a very specific sort: superconnected, ultrafast and hyperfunctional. here we ask: 1. what new system properties make at in financial markets interesting from a complexity governance perspective? 2. how can these properties be studied from a governance perspective? 3. what are the most important emerging governance responses deployed by decisionmakers and private actors trying to respond to these new properties? question 1 is elaborated in detail part 2 below. in part 3 we discuss how these properties pose challenges to existing theoretical approaches. in part 4 we elaborate question 2 by unpacking key governance functions in the context of at. in part 5, we explore question 3 by investigating emerging generic governance strategies employed by policy-makers in different parts of the world with the intention to harness at. we conclude this article by pointing out a number of intriguing questions for scholars who are interested in the governance of these sort of ultrafast, superconnected and hyperfunctional systems. 2. superconnected, ultrafast and hyperfunctional financial markets are sometimes portrayed as complex adaptive systems – systems whose behavior is in constant flux, prone to quite intricate emergent patterns including unavoidable uncertainties, cascading failures, and surprise (e.g. sornette, 2009; may, levin & sugihara, 2008). continuous innovation and evolution are key aspects of financial markets as hardware, software and human innovation drive the system towards higher (perceived at least) efficiency, connectivity and speed over time (chesini & giaretta, 2014; galaz, 2014). the speed of innovation in financial markets and associated technologies however, has created new forms of system properties that until now remain to be explored by governance scholars. three properties are of specific interest in this context: connectivity, speed and hyperfunctionality. 2.1. connectivity during the past decade it has become increasingly clear that connectivity in financial markets had amplified over time, almost to the point that these could be denoted as superconnected. by “connectivity” we mean that flows of information and financial resources have expanded to cover not only larger geographical areas, but also an increasing number of economic sectors in society. for example, stock exchanges have merged across continents in the last decades, thereby creating information channels that link market places across continents in ways that where unimaginable just a few decades ago. the merger between the new york stock exchange and euronext in 2007 is one case in point (karmel, 2007). algorithm based trading platforms nowadays also allow for trade with all sorts of financial instruments (ranging from futures to stocks and commodity derivatives), in any market around the world thereby contributing to additional connectivity.iv recent analyses also show that there are increased connections between what historically used to be separated asset classes. examples include the increased merger between financial http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 15 university of bamberg press capital and economic sectors involving technological and basic materials (harmon, stacey & baryam, 2010); financial capital and commodity derivatives such as commodity futures (galaz et al., 2015); and financial capital and natural resources through the emergence of financial instruments such as biodiversity, forests, fisheries and weather bonds (i.e. “catastrophe” bonds, see business and biodiversity offsets programme & the unep finance initiative, 2010). 2.2. speed information flows in financial markets are so fast as to over-perform human information processing capacities by several orders of magnitude. and technological innovation continues to push speed limits. in 2011, for example, the firm fixnetix developed a microchip that can help algorithms execute trades in 740 nanoseconds – 740 billionth of a second (0.000000740 seconds).v in 2013, american news agency cnbc reported on a server farm situated in washington able to transmit economic data for trading at near light speed to new jersey, via a superfast microwave transmission service.vi these explosive increases in speed are tightly coupled to information technological advances, creating what some have a denoted a technological “arms race” between algorithmic trading actors (lewis, 2015; snider, 2011; chaboud et al., 2014). this race towards increasingly rapid trades has not only created intriguing phenomena such extremely rapid volatility in stock prices (denoted “subsecond extreme events”, johnson, zhao, hunsader, qi, johnson, meng & tivnan, 2013), and the emergence of “predator algorithms” (i.e. algorithms designed to spy on, or crack the codes of other algorithms, see rose, 2010; weber, 2011). this change is so drastic that some have proposed that global financial markets have evolved into a new “machine ecology” where changes in the system’s behavior are considerably faster than human response time (johnson et al., 2013). in short: the speed of information flows in modern financial markets effectively surpasses human capacity to detect and respond to it. 2.3. hyperfunctionality speed and connectivity in combination create what we call “hyperfunctionality” in financial markets. the notion of hyperfunctionality is widespread. in industrial design, hyperfunctionality is used to describe an optimal fitness for purpose or the “disappearing tool”, i.e. tools that are so well adapted to the user’s need that s/he is longer aware that s/he is actually using a tool such as a computer (kettley, 2012). in a different genre, baudrillard (2006) uses hyperfunctionality to describe a variety of phenomena, including consumers’ constant search for machines that work faster than the previous computer or smartphone. the most common application of the hyperfunctionality concept is however in medicine, describing pathologies where bodily organs such as for instance the hypothalamus, produce excessive amounts of a particular hormone. in the present context of financial markets and at, we use hyperfunctionality in a similar perspective as the latter. we see hyperfunctionality as a systemic pathology caused by an overperforming subsystem in the wider context of economic governance. the root cause of the problem is that while decision making is an essential component of financial management and market behavior, algorithms perform that function infinitely faster than decision makers in any other areas of economic governance (for examples of these areas see gamble, 2000). just as a hyperfunctioning hypothalamus causes problems not to the hypothalamus itself but rather to other bodily organs, so too can hyperfunctioning algorithms in highly connected financial http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 16 university of bamberg press markets cause problems in other areas of economic governance by undermining the trust and stability of financial systems (discussed in the introduction). 3. theoretical approaches to algorithmic trade how can governance scholars approach this sort of hyperfuctional, hyperfast and complex system? a number of scholars have explored the challenges posed by the recent transformation of financial markets from a legal perspective (e.g. gomber & gsell, 2006; chesini & giaretta 2014; macnamara, 2015). however, few if any, have elaborated how this dramatic change in financial markets challenges more general principles or functions of governance (sensu pierre & peters, 2005; koppenjan & klijn, 2004; hooghe & marks, 2003; bell & hindmoor, 2009). a number of theoretical approaches or lenses are possible to apply in this context. existing theories of economic and financial governance provide important insights. financial markets, like all markets, have regulatory frameworks defining proper conduct and methods of exchange in that market. these regulatory frameworks can either be imposed by a higher-level authority— typically the state—or be deliberated among the market actors into a self-regulating regime (lindberg, campbell & hollingsworth, 1991). similar helpful academic discussions can be found in the governance literature exploring market regulation. regulations articulate the public interest and address collective action problems in markets, but excessive or poorly designed regulation jeopardizes the allocative efficiency of the market (eisner, 2000; levi-faur, 2006). therefore, regulators typically tread carefully, searching for a “sweet spot” that regulates the market according to political preferences without choking the efficiency of the market. globalization has however to some degree undermined the strength and effectiveness of domestic regulatory frameworks (eisner, 2000). an alternative approach to understanding governance dimensions of at could be theories of network governance. networks offer opportunities for informal collaboration towards specific collective goals among actors who may be pitted in competition with each other in other respects (torfing & sørensen, 2007). the idea that market actors form networks to overcome collective action problems or to generate economies of scale is not new (e.g. johnson, 1982). in this context, theories of network governance become relevant since financial regulation with trade at super-speed and scale require transnational and highly flexible network collaborations between international, state and non-state actors. to our knowledge, such an analysis has not been assumed yet in this domain. a last possible avenue to a conceptual understanding of at would be to bring in the literature on crisis governance. again, space will only allow a very brief and condensed account of a very rich literature. a common theme in this literature of relevance in this context, is the fact that crises tend lead to calls to centralize decision-making and control, and to emphasize the role of leadership to coordinate prompt responses (boin, 2004; boin, t’hart & stern, 2006; peters, pierre & randma-liiv, 2011). this literature is relevant since at has brought about crisis-like situations, however putting crisis decision-making in the hands of actors that normally are not studied as crisis managers. two important conclusions come out from this very brief review. one is that several theoretical approaches are useful to understand the governance challenges posed by at in financial markets. the other general conclusion is that all the theories elaborated briefly above tacitly assume either the possibility for centralized decision-making and control, or some form of exchange, communication and deliberation. this action could emerge either among sovereign agents pursuing a joint regime or between those agents and some higher-level authority (say, http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 17 university of bamberg press national regulators and private financial market actors). as we explore in more detail below, at does not fit neatly into any of these assumptions due to its inherent system properties and complex governance setting. hence there is a need for governance scholars to elaborate in what ways at challenges basic functions of governance (see section 4), and the ways policy-makers and private actors try to cope with the intriguing governance challenges posed by superconnected, ultrafast, and hyperfunctional systems (see section 5). 4. governance functions and challenges in order to better understand the governance problems and responses that are associated with at (explored in the next section) we need to differentiate between phases of the process of governing and policy-making. the stages model of policy-making dates back to harold lasswell’s work in the 1950’s, but it continues to be a useful way to think about the policy-making process even in these settings (for an overview see hupe & hill, 2006). in a governance perspective these stages roughly translate into what peters and pierre (2016) refers to as governance functions. they include goal definition, resource mobilization, decision-making and implementation. we will use these functions to elaborate on how at challenges conventional governance analysis, and as a framework for the assessment of the impact of at on economic governance more specifically. 4.1. goal definition defining goals for collective action is customarily seen as the province of government since it is a core democratic function that actors and institutions that can be held to democratic account should exercise, normatively speaking. however, markets are known to have a strong indirect influence on these goals, not least due to the economic growth that the market generates (hall, 1986; lindblom, 1977). at could be seen as making a positive contribution to this function by making financial markets efficient by e.g. improving the price discovery mechanism, accelerating the number of trades, reducing transaction costs and increasing liquidity (e.g. hendershott, jones & menkveld, 2011; vuorenmaa, 2013). however, it has also been suggested that at leads to sluggish and volatile trade with more extensive ramifications within the market and beyond (gerig, 2015; johnson et al., 2013; kirilenko, kyle, samadi & tuzun 2014; mcgowan, 2010). the speed and hyperfunctionality of at may thus contribute to financial instabilities not only within the financial sector, but also in other sectors and regions through increased connectivity (see previous section). one such example of a financial cascade was widely debated in the aftermaths of the 20082009 “food crises” (clapp, 2009), and more recent concerns that at is contributing to excessive price volatility in commodity financial derivative markets such as those for sugar and cocoa (galaz et al., 2015). such developments would greatly complicate goal setting as the nature of the problem would constantly mutate. effective, specified goal setting requires some degree of continuity in the nature of the problem which the policy seeks to address. the fundamental problem here is that while market developments and forecasts are essential to political goal setting—e.g. expected growth rates are a key variable in assessing the government’s financial capabilities over some future time period—at is likely to increase the uncertainty in such forecasting, and therefore increase the uncertainty in goal setting (see for example discussion on http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 18 university of bamberg press decision-making challenges posed by cascading crises with multiple uncertainties, galaz, moberg, olsson, paglia & parker 2011). 4.2. resource mobilization financial markets are integral to the economic performance of a country. for example, the u.s. subprime mortgage crisis in 2007-2009 was not isolated to the financial sector, but had severe repercussions on economic growth, unemployment rates and fiscal policies in general. the complex interdependency between the state and financial markets has been demonstrated on several occasions. the influence of the market on public policy and government action increased dramatically after the deregulation of financial markets in the 1980s and 1990s (reinhart & rogoff, 2008; strange, 1997, 1998). even so, interdependency, not dominance of one over the other, remains the general pattern. as an example, private financial institutions often ascribe a fundamental role to governments, even in what could be seen as largely deregulated international financial markets (mosley, 2003). what is the contribution of at to resource mobilization? here, a time perspective and attention to diffusion is important. by virtue of considerably lower transaction speed, and a superior capacity to exploit trends and developments in the market, at may generate stronger short-term yields and growth in the market compared to conventional financial management. the key drawback of at is that it may significantly amplify movements in the market, often beyond human control due to its speed and complex patterns (johnson et al., 2013). hence it is possible that at entails market instability and volatility which may be counter-productive to resource mobilization. thus, from a wider economic governance perspective, this means possible yields will benefit the owners of the most sophisticated and successful algorithms whereas the downfall of volatility may impact on the market and society in general as a whole (lewis, 2015; gos, 2011; rose, 2010). this suggests that the contribution of at in resource mobilization may potentially not only be skewed towards benefitting certain high-technologically potent financial actors (snider, 2014; chesini & giaretta, 2014), but also be uncertain from the perspective of policy-makers. 4.3. decision-making as is the case with goal setting, political decision-making is normally seen as reserved for democratic institutions as they are responsive and accountable to the electorate. financial markets and their actors are however rarely present in political decision-making. indeed, actors in financial markets emphasize that although the market’s development may have political causes or consequences, they do not see themselves as political actors (mosley, 2003; pierre, 1999). thus, in a formal sense decision making is reserved for senior elected officials. the relationship between decision-makers and financial actors is much more complex that this of course. markets are known to have strong indirect influence on political goals, not only due to the economic growth that they generate, but also through lobbying efforts (elaborated in the next section). market actors in general prefer to reduce perceived market dysfunctionality through self-regulation rather than through government regulation since it would minimize government intrusion (for a critical account see pirrong, 1995). at the same time however, financial market leaders are sensitive to even subtle signals from government suggesting, for instance, a regulatory review of the financial system. this phenomenon is of course, not unique http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 19 university of bamberg press for financial markets (examples from the literature include johnson’s [1982] notion of the “iron fist in a velvet glove” and scharpf, 1997 on the “shadow of hierarchy”. see also pierre, 1999 for an example from swedish financial policy after the severe banking crises in the 1990’s). as we will explore more in detail below, the arenas and pathways for democratic decisionmaking about hyper-fast financial markets are fragmented and opaque and hence they lack a clear leverage for institutional change that would support more transparent decision-making processes. 4.4. implementation deregulation in the 1980s and 1990s de facto dis-embedded the market from its national context albeit without reducing its significance in the domestic economy. the ramifications of this disjuncture became obvious during the series of financial crises in the 1990s and early 2000s, particularly the 2007-8 financial crisis in the united states. since then, governments in most western countries have increased regulation of financial markets. financial actors have not been passive bystanders in that process however (see pierre, 2013; see also bó, 2006 on “regulatory capture”). thus, financial markets are to some extent both the targets and the agents of the implementation of regulatory reform. 4.5. feedback, evaluation and learning financial markets may indirectly provide important feedback on governance and public policy albeit of a rather biased sort. governance actions or policy programs which are seen as hostile to the financial market for example, may result in declining flows of capital and a reallocation of investment capital. such action, whether actual or anticipated, can exert tremendous influence on policy choices (pierre, 1999). this type of feedback is not uncontroversial; although the market is sometimes said to have a “privileged position” in domestic policy making (lindblom, 1977), other social constituencies may well argue that government and public policy must take a broader view and not cater unilaterally to the interests of financial markets. this simple dichotomy between financial interests versus the public interest and is however blurred by the fact that clients with stakes in financial markets, are also voters and hence represent a powerful political constituency. speed is also an issue in this regard. the temporal scales at which at operates are so fast, as to over-perform human information processing capacities by several orders of magnitude. as we will explore more below, this poses decision-makers with difficult challenges as they attempt to make sense of rapidly changing circumstances, evaluate ongoing financial activities and learn from surprise or mistakes. let us now turn to the flipside of that discussion and see how political actors and institutions are trying to seek to gain control over at despite these severe challenges to well-known governance functions. this exploration works as a means to explore emerging governance strategies deployed to harness these highly complex and fast systems. 5. governing the algorithm – current governance responses the emergence of at hence creates difficult challenges for decision-makers as they try to set goals, mobilize resources, identify clear decision-making pathways, implement decisions, and http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 20 university of bamberg press continuously learn from changing circumstances. however, political decision-makers are far from powerless. they can produce governance responses that differ from steering functions normally explored by governance scholars since they focus on getting a grip of phenomena characterized by complexity, speed and the potential for propagating failure. in addition, private actors such as stock exchanges have strong (economic) incentives to contribute to improved oversight to maintain public trust in the inner workings of the financial system.vii below we explore these governance responses summarized under the headings increasing transparency, slowing down, and decoupling. these are intended to capture an important set of currently unfolding governance activities assumed by state (such as government agencies) and non-state actors (such as stock exchanges). as fleckner (2015) notes however, regulation in this domain takes place through a “bouquet of rules, norms and standards (p. 15). the presentation below should therefore be seen as a selection of important and illustrative governance responses in the context of ultrafast and complex systems worth further exploration.viii 5.1. increasing transparency it is well known that the effective governance of complexity requires adaptive learning and flexible responses to changing circumstances (pierre & peters, 2005; folke, hahn, olsson & norberg 2005; weick & sutcliffe, 2007; see also previous section on feedback, evaluation, and learning). this, however, requires continuous flows of correct information and changes in key parameters as a means to support sense making processes, mechanism for overview, and selfregulation. as a result, transparency becomes a critical issue for the governance of at. the lack of transparency in financial markets has been a hot topic globally in the last decade, and especially in the u.s. after the latest financial crisis in 2007-2009 (hellwig, 2009). in the case of at, transparency problems emerge through the combination of three factors – technological complexity, ultra-speed and financial innovation. in the first case, many have noted the “black box” character of financial algorithms. that is, while some financial algorithms are well known and widely used (e.g. the “volume weighted average price” or ”vwap” algorithm), others are protected as precious trade secrets (lenglet, 2011; pasquale, 2015). in addition, the speed and volume of information in addition also creates severe issues for transparency, and therefore rapidly increase the risks for governance failure. zhang (2010), for example, notes: […] the non-transparency that stems from high-frequency trades (which can happen in milliseconds) makes tracking the trades virtually impossible. it took sec [the securities and exchange commission, our addition], which has unlimited access to data, more than five months to determine what caused the flash crash. (pp. 22) see also golumbia (2013) for a similar comment. as the financial data company nanex notes, the faster financial information moves, the bigger data processing requirements are, and the slower surveillance activities becomes. or put much more drastically: trading at the microsecond level: in one millionth of a second, nothing a human does consciously, can be measured. a quote can travel about 1000 feet (300m). a 1 terabyte hard disk will store about 10 seconds worth of market data, which means you need over 2,000 disks to store 1 day. to receive market pricing information, you would need a terabit network, which is http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 21 university of bamberg press currently a bleeding edge technology. you will be dead before the sec finishes investigating an hour worth of data. quote from nanex (2015, february 4). hft doesn't harm investors and other absurdities, retrieved october 11th, 2016 from http://www.nanex.net/aqck2/3532.html lack of transparency is also the result of ingenious creations of new financial instruments created by financial actors themselves. in some cases, these are created to avoid political oversight. one clear example is the development of “swaps” in the 1990s in the u.s. as a means to avoid stricter government regulation of stocks and insurance products (golumbia, 2013:12, faiola, nakashima & drew, 2008). hence even though trades take place at hyperspeed, the secretive nature of certain types of financial algorithms, the massive amounts of data processed through financial systems, and continuous financial innovation, ironically makes oversight processes slow and challenging. as a response, policy makers and market actors try to gain control, implement decisions and learn from changing circumstances, by creating policies with the ambition to increase transparency.ix as snider (2011) notes, real-time monitoring of trades through automated systems began already in the 1990s in the us. systems such as these not only track trading activity, but are also designed to detect suspicious trading activities. the so called 2010 dodd–frank wall street reform and consumer protection act brought considerable changes to the u.s. financial infrastructure, including policies to increase transparency and government supervision of financial instruments and markets across asset classes including commodities (snider, 2011; scopino, 2015). x the 2010 “flash crash”xi, led to additional proposals to further increase transparency from the u.s. commodity futures trading commission, and u.s. securities and exchange commission (sec)– two u.s. agencies whose actions have global financial repercussions. in 2012, the u.s. securities and exchange commission voted for the incorporation of a new system that will allow for better overview of trade data and hence help audit the extent and behavior of algorithmic trade.xii failure to comply has led to several occasions of punishment through fines. for example, on november 25, 2011, both cme and nymex fined infinium capital management (infinium) a total of $850,000 for failing to supervise its alternative trading system (ats), hence making disruptive high-speed trading more difficult to monitor (scopino, 2015, p. 241). on march 25th 2015 sec proposed a rule that would require high-frequency trading firms in “dark pools”xiii to register with the financial industry regulatory authority, and submit a daily paper trail on their trading activities.xiv the sec has also proposed a consolidated audit trail (cat), a multibillion dollar monitoring system (keller, 2012, pp. 1479-1480). the european union has made several important advances in this regard as well. the markets in financial instruments directive (mifid) and recent revisions in the new directive (mifid ii) and a new regulation (mifir) play key roles on this regard.xv new legislation intends to increase transparency in so-called over-the-counter (otc) derivatives markets by supporting the creation of a new breed of trading platform, known as an organised trading facility (or otf). otc derivatives markets cover trade with credit default swaps and interest rate swaps that are mainly traded bilaterally between the 15 top banks, with a value of up to $640 trillion usd.xvi in addition, firms engaging in at will be required to notify its “member state competent authority” of the firm’s activities and may, at any time, be required to provide details of the systems, strategies and controls it has in place. trading venues will also be required to provide relevant authorities access to their order book on request so that the latter are able to monitor trading.xvii http://dx.doi.org/10.20377/cgn-55 http://www.nanex.net/aqck2/3532.html complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 22 university of bamberg press in australia, increased transparency and hence overview has been assumed through the creation of a new market surveillance system with the capacity to analyze very large qualities of data as a means to avoid “flash crashes” and similar market events. it is notable that the monitoring system builds on the same sophisticated technologies and algorithms it is intended to monitor (chesini & giaretta, 2014). transparency is also seen as critical by actors at the supranational level, such as by the international organization of securities commission (iosco). after a direct request from the g20, iosco presented a number of recommendations to market actors such as financial centers that would increase transparency and monitoring, including real-time monitoring and automated alerts (chesini & giaretta, 2014, p. 168; for details see iosco, 2011, 2013). these new monitoring systems again build on the same sort of technologies it is intended to monitor, such as data mining, sophisticated statistics and artificial intelligence (chesini & giaretta, 2014, p. 182). the “algorithm arms race” in the financial sector hence unravels not only between financial firms, but also between the financial sector and policy-makers as the latter strive for transparency and oversight. 5.2. slowing down another form of governance intervention in hyperfunctional systems seems to focus on reducing speed (as elaborated in partnoy, 2012). slowing down hyperfunctional systems allows regulating actors to not only better monitor trading activities, but also to intervene by disconnecting parts of the systems of interest if considered needed (see more about this last step under “decoupling”). one example of this strategy can be found amongst private actors, such as the brazilian stock market bolsa de valores, mercadorias & futuros de são paulo (or bm&f bovespa). this market requests at-firms to route their orders through a “pretrade risk tool” (stafford, 2011; mellow, 2014). a requirement such as this not only makes sure that errant buy or selling orders get filtered out before entering trade platforms, but also adds considerable latency (i.e. slows downs the trading speed) for actors acting normally acting on the scale of microseconds (holley, 2013). as at has made progression in other countries such as india, financial legislation has been changed to introduce “speed-bumps”, including leverage limits, securities transaction taxes (in 2004), and a basic levy for futures and options trade.xviii in 2013 the forward markets commission (fmc) in india announced its plans to put in place guidelines to oversee potentially adverse effects of algorithmic trade in commodity exchanges. some of the measures include limitations on the possibilities for at-firms to co-locate their servers next to the exchange’s servers, and fines in cases where orders do not lead into actual trade thereby reducing both speed and the volume of information entering the exchanges.xix policy-makers in china have implemented similar measures that not only put a fee on transactions, but also limit the velocity of trade by for example forbidding the sale of a stock that was bought the same day (the so called t+1 rule).xx each of these mechanisms contributes to a reduction of trading volume information, and increase the average trading time, a critical parameter for at-actors that tend to trade in very large volumes quickly. european legislators have explored other strategies to slow down financial markets – e.g. the possible creation of a “half-second rule”. this proposed rule would force at actors to maintain their market orders for at least 0.5 seconds. such a rule would in principle filter away at-driven orders that move in and out of stock exchanges (note that the current time span for such http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 23 university of bamberg press movements is below 5 microseconds [0.005 seconds]). the european parliament supported the rule officially in september 2012, but the rule was later dropped after negotiations between member states and private actors.xxi however, new pieces of european legislation in mifid2 require algorithmic traders to register their formulae with regulators. it should be noted that the final approval of mifid2 was only achieved after tough negotiations between european states with clearly conflicting views on the matter.xxii key agencies in the united states have also at several occasions attempted to advance policies that would slow down the flow of financial information. a joint sec and cftc committee in 2011 (cftc-sec advisory committee 2011) proposed restrictions on co-location (pp. 7f) and penalties for rapid order cancellation (pp. 11). these proposals in combination would help slow down financial information flows. however, these have yet to be implemented and seem to be halted by existing tensions between legislators and the private sector.xxiii 5.3. decoupling the last strategy is related to policy interventions aiming to strategically reduce connectivity by decoupling different parts of the system (sometimes denoted as “circuit breakers”, or “breaking points”). the goal here is to mitigate networked vulnerabilities, and reduce the possibilities of cascading failures as different parts of financial markets become increasingly connected (e.g. stock and commodity markets), or as financial innovations and actors start to penetrate other parts of the economic system (for an elaboration of the implications of such networked risks, see helbing, 2013; may et al., 2008). in 2014 the european parliament adopted a first draft legislation based on the markets in financial instruments derivative (mifid), attempting to constraint what is perceived as “harmful speculation” by proposing so called “position limits”. this refers to limits to the number of contracts in a particular commodity that can be held by a trader, or group of traders. the intention is to keep certain types of financial actors away from agricultural commodity markets as they may contribute to price volatility.xxiv legislation targeting algorithmic trade in particularly has also been adopted by the parliament in october 2012, and january 2014 (mifid2).xxv mifid2 also requires trading venues to introduce a so called a ‘kill button’ – mechanisms that temporarily halt trade in an exchange or in one stock if it falls a certain pre-defined percentage. the joint u.s. sec and cftc committee in 2011 not only proposed interventions to slow down flows of financial information, but also updated protocols for “circuit breakers”xxvi targeting at. the committee based this proposal on a previous private-public pilot project including only a subset of securities (e.g. securities included in the s&p 500 index), and called for an extension for larger segments and financial instruments in the market (sec-cftc advisory committee 2011, pp. 3f). again, these measures have yet to be implemented. 5.4. fighting back needless to say, all these initiatives with the ambition to increase transparency, decouple and slow down are highly conflictive, and have led to intense debates between policy-makers, representatives from a diversity of financial actors including banks and stock exchanges, and nongovernmental organizations. in the u.s. for example, lobbying efforts by wall street banks such as goldman sachs have been intense with the ambition to influence the details of the rules in such a way to keep lucrative trading practices unregulated or at least self-regulated (snider, 2011, http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 24 university of bamberg press p. 12; see also valdez, 2015). the number of european policy initiatives elaborated above have also led to intense negotiations between member states, but also concerns from associations such as the british bankers’ association who described parts of the legislation as “particularly onerous”.xxvii xxviii in contrast, u.s. federal courts have rejected the proposal on so called “position limits”, inducing new rounds of still ongoing debates and legislative processes. even the mere definition of what actually entails as “high frequency trading” is debated and contested (patterson & strasburg, 2012). hence while these measures might seem logical from a governance point of view, they do not come easy in a sector here decision-making is diffuse, large amounts of capital are at stake, and where causality is complex and highly contested. 6. conclusion: complexity governance in an ultrafast world as we have elaborated in detail in this article, the progression of at in financial markets provides a number of interesting insights into the governance of complexity, but complexity of a very specific sort: superconnected, hyperfunctional, and ultrafast. as a response to the first question posed in the opening of this paper, at is a new type of phenomenon in financial and commodity derivative markets, defined by atomistic behavior resulting from human-computer interactions, extremely high speed of transactions and a global reach. it is interesting to note that some accounts of financial management dysfunctional behavior in the 1980s and early 1990s (e.g. davis, 1998; krugman, 1994) appear as rather innocent financial extravagancies compared to some of the more extreme pathologies that can be associated with uncontrolled at. this calls for theoretical approaches that acknowledge this complexity, emerging phenomena such as hyperfunctionality, surprise and propagating shocks, and that explore novel ways to address these risks. as we have elaborated, at challenges simple notions of agency in financial market decisionmaking. for example, the fact that the dynamics of financial systems are to such large extent driven by the interplay between human action and algorithms flies in the face of most conventional models of governance, where deliberation, interaction, reciprocity and consent are seen as integral to the organization and pursuit of collective action (e.g. koppenjan & klijn, 2004; colander & kupers, 2014; duit & galaz, 2008; pierre & peters, 2005). the ‘”ormal state of affairs” in economic governance, as it has been conducted in the advanced western democracies over the past several decades, is characterized by the pursuit of the “sweet spot” between articulating collective interests in the market without distorting the allocative mechanisms in that market. in many countries this governance is conducted through extensive social consultation and deliberation in order to allow for external input, and generate consent and legitimacy for the economic policy. as we have discussed in this article, at is alien to this conventional governance arrangement. focusing on the governance aspects (and as a response to our second question), at forces us to re-evaluate existing theoretical approaches and what we perceive as key functions in governance. at seemingly changes the nature of economic governance from a pro-active to a reactive (“negative”) governance where accommodation and damage control seems to take precedence over future-oriented policy. whether this hypothesis holds over time, and in the face of additional empirical analysis from more countries and markets, is a critical research question for the future. hypefunctionality is an additional key issue to explore in this regard. governance failure is usually associated with non-functioning or malfunctioning systems or institutions (peters & http://dx.doi.org/10.20377/cgn-55 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 25 university of bamberg press pierre, 2015). in our view, at’s hyperfunctionality is a governance problem, more than it is a technical or legal problem. at thus provides an excellent example of globalization processes combined with rapid technological development, and a lagging regulatory response. once the regulatory process has caught up with the dysfunctions of at, new technological advances will most likely have created new technologies or financial innovations that evade the regulations. governance hence remains a central issue, and ultrafast systems of this sort forces us to ask how to ensure that the financial market is brought back into “sync” with other components and subsystems in the economy. however, such governance will by necessity be networked. the reason is the distributed nature of the issue spanning over multiple sectors, types of actors, and geographical political boundaries. this brings us to our third question about current governance responses. the responses explored here – transparency, slowing down and decoupling – show the emergence of new strategies assumed by multiple governmental and non-governmental decision-makers facing the risks entailed with ultrafast and superconnected systems. these responses are distributed across a number of governmental and private actors, at times converging into networked or loosely coupled polycentric responses as actors continuously adjust to each other’s behavior (as explored by galaz, crona, österblom, olsson & folke 2012 for polycentric coordination at large scales). while distributed, these initiatives strategically target key parameters in the system of interest as a means to make the at at all governable, by influencing information content and volume and the technological infrastructures that create connectivity and superspeed. whether these strategies will prove successful, to what extent these governance networks will emerge and prove robust, and whether political actors have the resources and technology to enforce these rules in multilevel governance settings (i.e. ranging from national to transnational settings), remain critical research questions. lessons from the financial sector explored here could, in the meantime, prove very valuable as governance scholars continue to explore emerging governance responses and failures in a hyperconnected world. references banks, e. (2010). dark pools: the structure and future of off-exchange trading and liquidity (basingstoke and new york: palgrave). baudrillard, j. (2006). the system of objects (london: verso). bbop & unep (2010). biodiversity offsets and the mitigation hierarchy: a review of current application in the banking sector. a study completed on behalf of the business and biodiversity offsets programme and the unep finance initiative by pricewaterhousecoopers llp. business and biodiversity offsets programme, online: http://www.foresttrends.org/documents/files/doc_2399.pdf bell, s. & a. hindmoor (2009). rethinking governance: the centrality of the state in modern society. cambridge: cambridge university press. boin, a. (2004). lessons from crisis research. international studies review, 6,165-94. boin, a., p. t’hart & e. stern (2005). the politics of crisis management: political leadership under pressure. cambridge: cambridge university press. campbell, j., j. r. hollingsworth & l. n. lindberg (eds) (1991). governance of the american economy. cambridge: cambridge university press. chaboud, a. p., chiquoine, b., hjalmarsson, e., & vega, c. (2014). rise of the machines: algorithmic trading in the foreign exchange market, the journal of finance, 69(5), 2045-2084. chesini, g. & e. giaretta (2014). regulating high-frequency trading: an examination of european, us and australian equity market structures. in lindblom, t., s. sjögren & m. willesson (eds.). financial systems, markets and institutional changes (pp. 152-187). london: palgrave macmillan. http://dx.doi.org/10.20377/cgn-55 http://www.forest-trends.org/documents/files/doc_2399.pdf http://www.forest-trends.org/documents/files/doc_2399.pdf complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 26 university of bamberg press cftc-sec advisory committee (2011). recommendations regarding regulatory responses to the market events of may 6, 2010. summary report of the joint cftc-sec advisory committee on emerging regulatory issues. clapp, j. (2009). food price volatility and vulnerability in the global south: considering the global economic context. third world quarterly, 30(6), 1183–1196. colander, d. & r. kupers (2014). complexity and the art of public policy: solving society’s problems from the bottom up. princeton: princeton university press. davis, h. (1998). a million a minute: inside the world of traders. london: nicholas brealey publishing. duit, a., & galaz, v. (2008). governance and complexity—emerging issues for governance theory. governance, 21(3), 311-335. eisner, m. a. (2000), regulatory politics in transition (2nd edn). baltimore: johns hopkins university press. faiola, a., e. nakashima, & j. drew (2008, october 15). what went wrong? washington post. retrieved october 13, 2016 from http://www.washingtonpost.com/wp-dyn/content/article/2008/10/14/ar2008101403343.html fleckner, a.m. (2015). regulating trading practices. max planck institute for tax law and public finance working paper 2015 – 08. available at ssrn: http://ssrn.com/abstract=2476950 folke, c., hahn, t. olsson p. & norberg j. (2005). adaptive governance of social-ecological systems. annual review of environmental resources, 30, 441–473. frank, a. b., collins, m. g., levin, s. a, lo, a. w., ramo, j., dieckmann, u. et al. (2014). dealing with femtorisks in international relations. proceedings of the national academy of sciences of the united states of america, 111(49), 17356-17362. galaz. v. (2014). global environmental governance, technology and politics: the anthropocene gap. cheltenham: edward elgar. galaz, v., moberg, f., olsson, e. k., paglia, e., & parker, c. (2011). institutional and political leadership dimensions of cascading ecological crises. public administration, 89(2), 361-380. galaz, v., b. crona, h. österblom, p. olsson, c. folke (2012). polycentric systems and interacting planetary boundaries—emerging governance of climate change–ocean acidification–marine biodiversity. ecological economics, 81, 21-32. galaz, v., gars, j., moberg, f., nykvist, b., & repinski, c. (2015). why ecologists should care about financial markets. trends in ecology & evolution, 30(10):571-580. gamble, a. (2000). economic governance. in j. pierre (ed.). debating governance (pp. 110-37). oxford: oxford university press. gerig, a. (2015). high-frequency trading synchronizes prices in financial markets. retrieved october 16, 2016, from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2173247 goldin, i. & m. mariathasan (2014). the butterfly defect: how globalization creates systemic risks, and what to do about it. princeton: princeton university press. golumbia, d. (2013). high-frequency trading: networks of wealth and the concentration of power. social semiotics, 23(2), 278-299. gomber, p., & gsell, m. (2006), catching up with technology the impact of regulatory changes on ecns/mtfs and the trading venue landscape in europe. competition & reg. network indus., 7(4), 535-558. gomber, p., arndt, b., lutat, m., & uhle, t. (2011). high-frequency trading. goethe universität frankfurt am main, germany. retrieved october 13, 2016 from doi:10.2139/ssrn.1858626 government office for science (2011). the future of computer trading in financial markets: an international perspective. government office for science, uk. hall, p. a. & soskice, d. (eds) (2001). varieties of capitalism: the institutional foundations of comparative advantage. oxford: oxford university press. harmon, d., stacey, b., & bar-yam, y. (2010). networks of economic market interdependence and systemic risk. retrieved october 13, 2016 from http://arxiv.org/abs/1011.3707 helbing, d. (2013). globally networked risks and how to respond. nature, 497(7447), 51-59. hellwig, m. f. (2009). systemic risk in the financial sector: an analysis of the subprime-mortgage financial crisis. de economist, 157(2), 129-207. hendershott, t., c. m. jones & a. j. menkveld (2011). does algorithmic trading improve liquidity? the journal of finance, 68(1), 1–33. héritier, a. & m. rhodes (eds.) (2011). new modes of governance in europe: governing in the shadow of hierarchy. basingstoke: palgrave. holley, e. (2013, october 8). race for speed intensifies as hft goes global. banking technology, retrieved october 13, 2016 from http://www.bankingtech.com/173902/race-for-speed-intensifies-as-hft-goes-global/ hooghe, l. & g. marks (2003). unraveling the central state, but how? types of multi-level governance. american political science review, 97(2), 233-243. http://dx.doi.org/10.20377/cgn-55 http://www.washingtonpost.com/wp-dyn/content/article/2008/10/14/ar2008101403343.html http://ssrn.com/abstract=2476950 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2173247 http://arxiv.org/abs/1011.3707 http://www.bankingtech.com/173902/race-for-speed-intensifies-as-hft-goes-global/ complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 27 university of bamberg press hupe, p. l. & m. j. hill (2006). the three action levels of governance: re-framing the policy process beyond the stages model. in b. g. peters & j. pierre (eds.), handbook of public policy (pp. 13-30). london: sage. iosco (2011). regulatory issues raised by the impact of technological changes on market integrity and efficiency. madrid: international organization of securities commission. iosco (2013). technological challenges to effective market surveillance: issues and regulatory tools. madrid: international organization of securities commission. johnson, c. a. (1982). miti and the japanese miracle. stanford: stanford university press). johnson, n., zhao, g., hunsader, e., qi, h., johnson, n., meng, j., et al. (2013). abrupt rise of new machine ecology beyond human response time. scientific reports, 3, 2627. doi:10.1038/srep02627 karmel, r. s. (2007). the once and future new york stock exchange: the regulation of global exchanges. brooklyn journal of corporate, financial & commercial law, 1, 355–410. keller, a. j. (2012). robocops: regulating high frequency trading after the flash crash of 2010. ohio st. lj, 73, 1457– 1485. kettley, s. (2012). interrogating hyperfunctionality. in p. breedon (ed.), smart design (pp. 65-73). london: springer. kirilenko, a. a., kyle, a. s., samadi, m., & tuzun, t. (2014). the flash crash: the impact of high frequency trading on an electronic market. retrieved october 13, 2016 from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1686004 koppenjan, j. f. m., & klijn, e. h. (2004). managing uncertainties in networks: a network approach to problem solving and decision making. london: routledge. krugman, p. (1994). currencies and crises. cambridge, ma: mit press. lenglet, m. (2011). conflicting codes and codings: how algorithmic trading is reshaping financial regulation. theory, culture & society, 28(6), 44–66. levi-faur, d. (2006). regulatory capitalism: the dynamics of change beyond telecoms and electricity. governance: an international journal of policy, administration, and institutions, 19, 497-525. lewis, m. (2015). flash boys – a wall street revolt. new york: w.w. norton. lindberg, l. n., campbell, j. l., & hollingsworth, j. r. (1991). economic governance and the analysis of structural change in the american economy. in lindberg, l. n., campbell, j. l., & hollingsworth, j. r. (eds.). governance of the american economy (pp. 3-34). cambridge: cambridge university press. lindblom, c. e. (1977). politics and markets. new york: basic books. mcnamara, s. (2015). hft class action defeats and the challenge to regulators. retrieved october 13, 2016 from columbia law school blog http://clsbluesky.law.columbia.edu/2015/11/12/hft-class-action-defeats-and-the-challenge-to-regulators/ may, r. m., levin, s. a., & sugihara, g. (2008). complex systems: ecology for bankers. nature, 451, 893–895. mcgowan, michael j. (2010). the rise of computerized high frequency trading: use and controversy. duke law & technology review, 16, 1. mellow, c. (2014, june 9). high frequency trading gets a mixed reception in emerging markets, institutional investor. retrieved october 13, 2016 from http://www.institutionalinvestor.com/article/3349687/asset-managementequities/high-frequency-trading-gets-a-mixed-reception-in-emerging-markets.html#.vrwifgtf_7c mosley, l. (2003). global capital and national governments. cambridge: cambridge university press. narang, r.k. (2013). inside the black box – a simple guide to quantitative and high-frequency trading. new york: wiley. partnoy, f. (2012). wait: the art and science of delay. new york: public affairs. pasquale, f. (2015). the black box societythe secret algorithms that control money and information. harvard: harvard university press. patterson, s. & j. strasburg (2012, june 19). cftc moves to brake high-speed traders. wall street journal. retrieved october 13, 2016 from http://www.wsj.com/articles/sb10001424052702303379204577477003146372804 peters, b.g. & pierre, j. (2016). comparative governance: rediscovering the functional dimension of governing. cambridge: cambridge university press. pierre, j. (1999). marknaden som politisk aktör – politik och finansmarknad i 1990-talets sverige. swedish government inquiry, stockholm. [statens offentliga utredningar, sou 1999:131]. (60pp). pierre, j. (2013). globalization and governance. cheltenham: edward elgar. pierre, j., & peters, b. g. (2005). governing complex societies: trajectories and scenarios. basingstoke: palgrave macmillan. pierre, j. (2015). varieties of capitalism and varieties of globalization: comparing patterns of market deregulation. journal of european public policy, 22, 908-26. peters, b. g., j. pierre & t. randma-liiv (2011). global financial crisis, public administration and governance: do new problems require new solutions?. public organization review, 11, 13-27. pirrong, s. c. (1995). the self-regulation of commodity markets: the case of market manipulation. journal of law and economics, 38(1), 141-206. http://dx.doi.org/10.20377/cgn-55 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1686004 http://clsbluesky.law.columbia.edu/2015/11/12/hft-class-action-defeats-and-the-challenge-to-regulators/ http://www.institutionalinvestor.com/article/3349687/asset-management-equities/high-frequency-trading-gets-a-mixed-reception-in-emerging-markets.html#.vrwifgtf_7c http://www.institutionalinvestor.com/article/3349687/asset-management-equities/high-frequency-trading-gets-a-mixed-reception-in-emerging-markets.html#.vrwifgtf_7c http://www.wsj.com/articles/sb10001424052702303379204577477003146372804 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 28 university of bamberg press reinhart, c. m. & k. s. rogoff (2008). this time is different: a panoramic view of eight centuries of financial folly. princeton: princeton university press. rose, c. (2010). dark pools and flash orders: the secret world of automated high-frequency trading. journal of business & economics research, 8(8), 11–16. scharpf, f. w. (1997). games real actors play: actor-centered institutionalism in policy research. boulder, co: westview press. scopino, g. (2015). do automated trading systems dream of manipulating the price of futures contracts? policing markets for improper trading practices by algorithmic robots. florida law review, 67, 221–294. snider, l. (2011, september). criminalizing the algorithm? stock market crime in the 21st century, plenary address delivered at the anz conference of criminology. snider, l. (2014). interrogating the algorithm: debt, derivatives and the social reconstruction of stock market trading. critical sociology, 40, 747–761. sornette, d. (2009). why stock markets crash: critical events in complex financial systems. princeton: princeton university press. stafford, p. (2011, august 10). bm&fbovespa high-frequency trading rises, financial times. retrieved october 13, 2016 from http://www.ft.com/intl/cms/s/0/a6dbd6d4-c32c-11e0-9109-00144feabdc0.html#axzz3vrnkf5oo strange, s. (1997). casino capitalism. manchester: manchester university press. strange, s. (1998). mad money: when markets outgrow governments. ann arbor: university of michigan press. torfing, j. & e. sörensen (eds.) (2007). theories of democratic network governance. basingstoke: palgrave. us commodity futures trading commission & us securities and exchange commission (2010). preliminary findings regarding the market events of may 6, 2010 report of the staffs of the cftc and sec to the joint advisory committee on emerging regulatory issues. retrieved october 13, 2016 from http://www.cftc.gov/idc/groups/public/@otherif/documents/ifdocs/opa-jointreport-sec-051810.pdf us commodity futures trading commission & us securities and exchange commission (2011). recommendations regarding regulatory responses to the market events of may 6, 2010. summary report of the joint cftc-sec advisory committee on emerging regulatory issues. retrieved october 13, 2016 from http://www.cftc.gov/idc/groups/public/@aboutcftc/documents/file/jacreport_021811.pdf valdez, j. m. (2015). should more be done to address high-frequency trading? the journal of trading, 10(2), 79–80. vuorenmaa, t. a. (2013). the good, the bad, and the ugly of automated high-frequency trading. the journal of trading, 8(1), 58–74. weber, b. (2011). the growing threat of rogue trading. business strategy review, 22(2), 50-53. weick, k. e. & k. m. sutcliffe (2007). managing the unexpected: resilient performance in an age of uncertainty. new york: wiley. zhang, f. (2010). high-frequency trading, stock volatility, and price discovery. retrieved october 13, 2016 from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1691679 endnotes i it should be noted that the definition of at in itself is highly contested. here we define algorithmic trade (at) as trade that relies on high-speed computer linkages; uses of sophisticated statistical, econometric, machine learning, and other quantitative techniques; and cases where traders hold positions for very short periods of time (micro-seconds) (government office for science (gos), 2011, p. 28; us securities and exchange commission, 2010; hendershott, jones, & menkveld, 2011, p. 1). ii note that these four examples are simplifications. for a more detailed description of different trading strategies and types of models, see narang (2013). iii the recent debate spurred by lewis’ book ”flashboys” (2014) is a good illustration of this challenge. the book triggered intense criticism from financial experts and commentators, see for example narang, m. (2014, april 4). a much-needed hft primer for 'flash boys' author michael lewis, institutional investor. retrieved october 11, 2016, http://dx.doi.org/10.20377/cgn-55 http://www.ft.com/intl/cms/s/0/a6dbd6d4-c32c-11e0-9109-00144feabdc0.html#axzz3vrnkf5oo http://www.cftc.gov/idc/groups/public/@otherif/documents/ifdocs/opa-jointreport-sec-051810.pdf http://www.cftc.gov/idc/groups/public/@aboutcftc/documents/file/jacreport_021811.pdf https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1691679 complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 29 university of bamberg press from http://www.institutionalinvestor.com/article/3328158/banking-and-capital-markets-trading-andtechnology/a-much-needed-hft-primer-for-flash-boys-author-michael-lewis.html#.u6ygpi1dwjg ; weil, jonathan (2014, april 1). weil on finance: fbi hops on michael lewis bandwagon, bloomberg news. retrieved october 11, 2016, from http://www.bloombergview.com/articles/2014-04-01/weil-on-finance-fbi-hops-on-michaellewis-bandwagon ; bradford, harry (2014, april 1). fbi investigating high-frequency traders. huffington post. retrieved october 11, 2016, from http://www.huffingtonpost.com/2014/03/31/fbi-high-speedtrading_n_5065622.html iv see for example http://www.algotrader.ch/ or https://www.quantopian.com/ which even hosts competitions as a means to crowd source hedge funds’ algorithm based strategies. v bbc news (2011, november 18). 'lightning fast' future traders working in nanoseconds, retrieved october 11, 2016, from http://www.bbc.com/news/business-15722530 vi cnbc (2013, july 24). buying an edge: the high-tech race for economic data pays. retrieved october 11, 2016, from http://www.cnbc.com/id/100910912 vii it should be noted that the actor landscape is considerably more complex and also involve actors such as commercial banks, investment banks, mutual funds, pension funds, venture capital, and others. for simplicity, we only focus on major governmental and private responses. viii the summary builds on a combination of secondary literature, reports from prominent organizations, and news articles as a way to get as rich of a picture as possible of unfolding and highly fluid governance efforts. it should be noted however that our ambition is not to present all private and public regulatory responses to at the last decade. not only would such a synthesis exceed the word limitations of an article, but it would also distract us from making our key argument. ix it should be noted is that our summary below includes policies that are proposed, and others that already are in effect. we have included both these categories to show the rapid evolution of policies, and how these match the identified governance activities (increasing transparency, slowing down, decoupling). our summary also includes policies and proposals from different parts of the world to illustrate the global scope of these governance attempts. however, our ambition has not been to create an all-encompassing list of regulation, but showcase a number of important policies as identified in secondary literature. x bloomberg business (2010, october 12). cftc to scrutinize algorithmic trading after may market plunge. retrieved october 11, 2016 from http://www.bloomberg.com/news/2010-10-12/commodity-regulator-to-review-algorithmictrading-after-may-market-plunge.html xi the “flash crash” is the term for a rapid crash on may 6th, 2010, when the dow jones industrial average dived about 1000 points, only to recover within minutes. (us commodity futures trading commission, & us securities and exchange commission, 2010). xii high frequency traders (2012, july 13). sec votes in favour of high frequency trading audits. retrieved october 11, 2016 from http://highfrequencytraders.com/news/2379/sec-votes-favour-high-frequency-trading-audits xiii “dark pools” are electronic trading venues that don’t display public quotes for stocks. these “pools” anonymously match orders of bigger investors with the ambition not alert the wider market of their intentions. estimates show that there are about 40 dark pools operating today, and include venues such as luquidnet, goldman sach’s sigma x and credit suisse’s advanced execution services (from mcgowan, 2010, see also banks, 2010). xiv new york post (2015, march 25). sec chair takes aim at predatory traders in ‘dark pools’, retrieved october 11, 2016 from http://nypost.com/2015/03/25/sec-chief-takes-aim-at-predatory-traders-in-dark-pools/ xv formally the final mifid ii and mifir texts were published in the official journal of the eu (oj) on 12 june 2014 and entered into force 20 days later on 2 july 2014. entry into application will follow on 3 january 2017. source: norton rose fulbright (2014, october). mifid ii / mifir series. retrieved october 11, 2016 from http://www.nortonrosefulbright.com/knowledge/publications/115236/mifid-ii-mifir-series xvi source: reuters (2014, january 14). eu reaches deal on sweeping securities reform. retrieved october 11, 2016 from http://uk.reuters.com/article/2014/01/14/uk-eu-mifid-idukbrea0d1ma20140114 xvii source: norton rose fulbright (2014, october). mifid ii / mifir series. retrieved october 11, 2016 from http://www.nortonrosefulbright.com/knowledge/publications/115236/mifid-ii-mifir-series http://dx.doi.org/10.20377/cgn-55 http://www.institutionalinvestor.com/article/3328158/banking-and-capital-markets-trading-and-technology/a-much-needed-hft-primer-for-flash-boys-author-michael-lewis.html#.u6ygpi1dwjg http://www.institutionalinvestor.com/article/3328158/banking-and-capital-markets-trading-and-technology/a-much-needed-hft-primer-for-flash-boys-author-michael-lewis.html#.u6ygpi1dwjg http://www.bloombergview.com/articles/2014-04-01/weil-on-finance-fbi-hops-on-michael-lewis-bandwagon http://www.bloombergview.com/articles/2014-04-01/weil-on-finance-fbi-hops-on-michael-lewis-bandwagon http://www.huffingtonpost.com/2014/03/31/fbi-high-speed-trading_n_5065622.html http://www.huffingtonpost.com/2014/03/31/fbi-high-speed-trading_n_5065622.html http://www.algotrader.ch/ https://www.quantopian.com/ http://www.cnbc.com/id/100910912 http://www.bloomberg.com/news/2010-10-12/commodity-regulator-to-review-algorithmic-trading-after-may-market-plunge.html http://www.bloomberg.com/news/2010-10-12/commodity-regulator-to-review-algorithmic-trading-after-may-market-plunge.html http://highfrequencytraders.com/news/2379/sec-votes-favour-high-frequency-trading-audits http://nypost.com/2015/03/25/sec-chief-takes-aim-at-predatory-traders-in-dark-pools/ http://www.nortonrosefulbright.com/knowledge/publications/115236/mifid-ii-mifir-series http://uk.reuters.com/article/2014/01/14/uk-eu-mifid-idukbrea0d1ma20140114 http://www.nortonrosefulbright.com/knowledge/publications/115236/mifid-ii-mifir-series complexity, governance & networks – vol. 3, no 2 (2017), p. 12-28 victor galaz, jon pierre: superconnected, complex and ultrafast: governance of hyperfunctionality in financial markets doi: http://dx.doi.org/10.20377/cgn-55 30 university of bamberg press xviii institutional investor (2015, june 9). high frequency trading gets a mixed reception in emerging markets, 2015-0609. retrieved october 11, 2016 from: http://www.institutionalinvestor.com/article/3349687/asset-managementequities/high-frequency-trading-gets-a-mixed-reception-in-emerging-markets.html#.vrwifgtf_7c xix the economic times (2013, february 1). forward markets commission lays down stringent norms for algorithm trades, retrieved october 11, 2016 from: http://articles.economictimes.indiatimes.com/2013-0201/news/36684553_1_forward-markets-commission-algos-high-frequency-trades xx institutional investor (2015, june 9). high frequency trading gets a mixed reception in emerging markets, 2015-0609. retrieved october 11, 2016 from: http://www.institutionalinvestor.com/article/3349687/asset-managementequities/high-frequency-trading-gets-a-mixed-reception-in-emerging-markets.html#.vrwifgtf_7c xxi reuters (2013, october 22). eu agrees preliminary deal to rein in speed traders. retrieved october 11, 2016 from http://www.reuters.com/article/2013/10/22/europe-speedtrading-idusl5n0ic1nh20131022 xxii see e.g. the economist (2014, april 26). a bigger bang. 2014-04-26, retrieved october 11, 2016 from http://www.economist.com/news/finance-and-economics/21601294-bold-new-law-will-reshape-europes-capitalmarkets-bigger-bang ; the bureau of investigative journalism (2012, september 17). inside the european negotiations that could rein in hft, retrieved october 11, 2016 from: http://www.thebureauinvestigates.com/2012/09/17/inside-the-european-negotiations-that-could-rein-in-hft/ ; norton rose fulbright (2014, october). mifid ii / mifir series. retrieved october 11, 2016 from http://www.nortonrosefulbright.com/knowledge/publications/115236/mifid-ii-mifir-series xxiii see for example mcnamare, s. (2015, november 12). hft class action defeats and the challenge to regulators. columbia law school blog, retrieved october 11, 2016 from http://clsbluesky.law.columbia.edu/2015/11/12/hft-classaction-defeats-and-the-challenge-to-regulators/ xxiv the guardian (2014, january 15). eu curb on food speculation gets qualified welcome from activists. 2014-01-15. retrieved october 11, 2016 from http://www.theguardian.com/global-development/2014/jan/15/eu-curb-foodspeculation-activists xxv financial news (2011, november 21). mifid ii: in a nutshell, retrieved october 11, 2016 from http://www.efinancialnews.com/story/2011-10-21/mifid-two-in-a-nutshell ; reuters (2014, january 14). eu reaches deal on sweeping securities reform. 2014-01-14. retrieved october 11, 2016 from http://uk.reuters.com/article/2014/01/14/uk-eu-mifid-idukbrea0d1ma20140114 xxvi circuit breakers are measures that temporarily halt trading on an exchange or in individual securities when prices hit pre-defined volatility boundaries. xxvii from the bureau of investigative journalism (2012, september 17). inside the european negotiations that could rein in hft. retrieved october 11, 2016 from http://www.thebureauinvestigates.com/2012/09/17/inside-theeuropean-negotiations-that-could-rein-in-hft/ xxviii financial times (2012, september 29). us court scraps cftc position limits rule. retrieved october 11, 2016 from http://www.ft.com/cms/s/0/be191d8e-09a8-11e2-a424-00144feabdc0.html#axzz2rbeylmup http://dx.doi.org/10.20377/cgn-55 http://www.institutionalinvestor.com/article/3349687/asset-management-equities/high-frequency-trading-gets-a-mixed-reception-in-emerging-markets.html#.vrwifgtf_7c http://www.institutionalinvestor.com/article/3349687/asset-management-equities/high-frequency-trading-gets-a-mixed-reception-in-emerging-markets.html#.vrwifgtf_7c http://articles.economictimes.indiatimes.com/2013-02-01/news/36684553_1_forward-markets-commission-algos-high-frequency-trades http://articles.economictimes.indiatimes.com/2013-02-01/news/36684553_1_forward-markets-commission-algos-high-frequency-trades http://www.institutionalinvestor.com/article/3349687/asset-management-equities/high-frequency-trading-gets-a-mixed-reception-in-emerging-markets.html#.vrwifgtf_7c http://www.institutionalinvestor.com/article/3349687/asset-management-equities/high-frequency-trading-gets-a-mixed-reception-in-emerging-markets.html#.vrwifgtf_7c http://www.reuters.com/article/2013/10/22/europe-speedtrading-idusl5n0ic1nh20131022 http://www.economist.com/news/finance-and-economics/21601294-bold-new-law-will-reshape-europes-capital-markets-bigger-bang http://www.economist.com/news/finance-and-economics/21601294-bold-new-law-will-reshape-europes-capital-markets-bigger-bang http://www.thebureauinvestigates.com/2012/09/17/inside-the-european-negotiations-that-could-rein-in-hft/ http://www.nortonrosefulbright.com/knowledge/publications/115236/mifid-ii-mifir-series http://clsbluesky.law.columbia.edu/2015/11/12/hft-class-action-defeats-and-the-challenge-to-regulators/ http://clsbluesky.law.columbia.edu/2015/11/12/hft-class-action-defeats-and-the-challenge-to-regulators/ http://www.theguardian.com/global-development/2014/jan/15/eu-curb-food-speculation-activists http://www.theguardian.com/global-development/2014/jan/15/eu-curb-food-speculation-activists http://www.efinancialnews.com/story/2011-10-21/mifid-two-in-a-nutshell http://uk.reuters.com/article/2014/01/14/uk-eu-mifid-idukbrea0d1ma20140114 http://www.thebureauinvestigates.com/2012/09/17/inside-the-european-negotiations-that-could-rein-in-hft/ http://www.thebureauinvestigates.com/2012/09/17/inside-the-european-negotiations-that-could-rein-in-hft/ http://www.ft.com/cms/s/0/be191d8e-09a8-11e2-a424-00144feabdc0.html#axzz2rbeylmup communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 31 university of bamberg press communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach authors: danny schippera, lasse gerritsb a department of public administration, erasmus university rotterdam, the netherlands e-mail: schipper@fsw.eur.nl b department of political science, otto-friedrich university of bamberg, germany e-mail: lasse.gerrits@uni-bamberg.de early in 2014, the dutch railway system spiralled out of control after traffic management was confronted with the decision to take four double switches and two rail tracks out of service. a lack of coordination between the responsible teams resulted in the decision to stop all traffic in one of the busiest parts of the network during the rush hour. in this study we aim to understand why the teams in the dutch railway system were not able to adopt a coordinated approach to reschedule rail services. to answer this question, we used a mixed-method approach by combining dynamic network analysis (dna) with sensemaking. our analyses show that a diverging framing of the situation accumulated over time, leading to inconsistent actions, incorrect assumptions and a lack of effective communication. informal and indirect communication spurred uncertainty and promoted negative emotions, which eventually resulted in a conflict between the actors. we discuss the difficulties of managing ambiguous events in multi-team systems. keywords: coordination; sensemaking; communication; railway; dynamic network analysis; complex system disruptions 1. introduction early in 2014 the dutch railway system spiralled out of control after the rail infrastructure manager decided to take four double switches and two rail tracks out of service. this decision was taken early in the afternoon by the responsible track team who thought that the equipment no longer met the authority’s safety standards. these switches and tracks are located at three of the busiest train stations in the netherlands in two different rail traffic control areas (from here on: areas a and b). since their decision would have a huge impact on the train service, the track team decided to give the other teams several hours of preparation time. regardless, what should have been a coordinated procedure, resulted in a loss of control. while the train dispatchers in area b managed the process adequately, train dispatchers in area a decided to stop the entire service during peak hours. many passengers were stranded, which drew negative attention from the media and politicians. we aim to answer the following research questions: why were the actors in http://dx.doi.org/10.20377/cgn-57 mailto:schipper@fsw.eur.nl mailto:lasse.gerrits@uni-bamberg.de complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 32 university of bamberg press the railway system unable to adopt a coordinated approach in order to adjust operations, and what explains the difference in response between the traffic control centres of areas a and b? like many other critical infrastructures (cis) in europe, the dutch railway system has undergone major changes over the past decades under the influence of eu policies. the dutch railway system has changed from a mostly large-scale integrated monopoly into a networked system consisting of multiple private and public organizations with diverging goals and specialized tasks, which may be conflicting (de bruijne & van eeten, 2007; schulman, roe, eeten, & bruijne, 2004). providing reliable services therefore requires multiple teams, who are separated by organizational and geographical boundaries, to continuously negotiate and renegotiate issues related to reliability (de bruijne, 2006). however, this network of teams poses additional challenges in terms of coordination and communication. for example, studies have found that geographically dispersed teams have difficulties in distributing information evenly, accurately, and in time (hinds & mcgrath, 2006). we are therefore interested in how actors in such networks communicate. to study the flows of information, we used social network analysis tools, more specifically dynamic network analysis (dna). the premise for this research is that coordination problems aren’t just the result of deficiencies in the quantity of information flows. given the contextual differences in which people are working and their differences with regard to knowledge, goals and expertise, there is a considerable chance that information will be interpreted differently (vlaar, van fenema, & tiwari, 2008). this could lead to different understandings of specific situations and therefore to potential conflicts regarding the course of actions to be followed. successful coordination stems from a congruent framing of a situation (cornelissen, mantere, & vaara, 2014). consequently, the structural dimensions of communication need to be studied in conjunction with the attribution of meaning in order to understand coordination in a network of diverse teams. in this study we therefore combined dna with theories of sensemaking to understand how information is processed within and among actors in order to understand how their actions made sense to them at that time (dekker, 2006; muhren, eede, & van de walle, 2008). this research contributes to the literature in three areas. first, we extend coordination beyond individual actors or collocated groups and look at coordination on the level of the whole network of geographically and organizationally separated teams, something which has received little attention (gittell & weiss, 2004; zaccaro, marks, & dechurch, 2012). second, we use a mixed-method approach by combining dynamic network analysis with an analysis of sensemaking. third, we answer the call for the integration of time dynamics into network studies using dna, to see how the structure of the inter-organizational network changes over time and how the relative importance of actors within the network changes (abbasi & kapucu, 2012; wolbers, groenewegen, mollee, & bím, 2013). the remainder of the paper is structured as follows. we will discuss the dynamics of communication and sensemaking in network coordination in section 2. the mixed methodology is explained in section 3, followed by an introduction to the case in section 4. we will identify key moments and actors by looking at communication patterns in section 5. the case is analyzed in section 6. the discussion and conclusions are presented in sections 7 and 8. 2. coordination and sensemaking between teams coordination can be achieved through pre-defined plans and procedures, but these formal modes of coordination are not always able to deal with the dynamics and uncertainty of specific http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 33 university of bamberg press situations and may severely limit the flexibility of organizations (bigley & roberts, 2001; johansson & hollnagel, 2007). adjusting to uncertain situations requires actors to mutually adapt and collectively improvise (cornelissen et al., 2014). communication plays a crucial role in the coordination of actions. especially rich informal communication has been identified as one of the most important sources of resilient system performance (roe & schulman, 2008). regular information updates to other team members help to create and maintain a shared understanding of problems and the actions needed to tackle them (kontogiannis & malakis, 2013). thus, accurate information updates should be provided regularly and on time (gittell & weiss, 2004). while previous researchers emphasized the importance of effective communication for successful coordination, they also found that geographically distributed teams face greater obstacles in sharing information effectively. since communication between distributed teams is often technology-mediated, the information flows in these processes are restricted (e.g. number of communication lines) and the updating of information suffers from delays (salas, burke, & samman, 2001). as hinds & mcgrath (2008) describe, distributed work settings lead to less informal and spontaneous communication in comparison to teams that are collocated. people working at different locations will also have different information assumptions, preferences and constraints (vlaar et al., 2008). consequently, information is often distributed unevenly and communication patterns can be quite unpredictable (powell, piccoli, & ives, 2004). moreover, since information flows and format are mediated by technology, important visual and social cues associated with traditional face-to-face interaction methods that help to interpret communication and team members’ actions are absent (fiore, salas, cuevas, & bowers, 2003). in short, challenges of understanding and communication are more salient in a network of diverse teams, which could result in the development of a different framing of situations. it is therefore important that the communicating parties can reach at least a congruent or compatible shared understanding during non-routine situations, (fiore et al., 2003). following wolbers & boersma (2013), we see this act of creating a sufficiently shared understanding as a process of sensemaking. sensemaking means that actors try to understand events that are novel, ambiguous or contrary to expectations. they deal with this ambiguity or uncertainty by creating plausible interpretations of reality through the extraction of cues from their environment to create an initial sense of the situation (maitlis, 2005; maitlis & christianson, 2014; sandberg & tsoukas, 2014; weick, sutcliffe, & obstfeld, 2005). cues trigger sensemaking as they indicate a discrepancy in the ongoing flow of events, which creates uncertainty about how to act. actors then try to interpret and explain these surprising events by placing these cues in a mental model or frames of roles, rules, procedures and authority relations (weick, 1993). these frames thus play an important role in terms of coordination as they trigger specific activities and expectations regarding the behaviour of others (cornelissen et al., 2014). finally, the sense made of the situation has to be put into action, to see whether it restores the interrupted event or if it is necessary to revise the interpretation. sensemaking is therefore essentially an episodic process that occurs from the moment organizational activities are interrupted until they are restored or permanently interrupted and contains three recurrent steps: noticing and bracketing cues from the environment, creating interpretations and action taking (maitlis & christianson, 2014; sandberg & tsoukas, 2014). sensemaking is a social process because actors interpret their environment in and through interactions with each other, thereby constructing shared accounts that allow them to comprehend the world and act collectively (maitlis, 2005). actors can therefore influence the sensemaking and meaning construction of others, a process that is called sensegiving (gioia & chittipeddi, 1991). actors can also actively demand information and clarification, which is called http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 34 university of bamberg press sensedemanding (vlaar et al., 2008). as cornelissen et al. (2014) put it, successful or failed coordination depends on how actors individually and collectively frame and reframe situations as a basis for action. these accounts don’t have to be completely overlapping, but they should be equivalent enough to allow coordinated action (maitlis & christianson, 2014). however, the creation of a shared understanding is a difficult task that requires much effort and interaction (bechky, 2003). the responses to violated expectations or ambiguous events depend on a variety of factors, e.g. individual, social, or organizational identity and personal and strategic goals (maitlis & christianson, 2014). this means that sensemaking is tied to individuals and that meaning in organizations is often contested because of the different positions, interests and backgrounds of actors (brown, stacey, & nandhakumar, 2008). to sum up, coordination in a network of diverse teams requires both effective information sharing and acts of collective sensemaking in order to create a sufficient shared understanding of the task situation. in this study we therefore combined dynamic network analysis to capture the flows of information with a qualitative analysis of how this information is processed by the actors. in the next section we will explain how we gathered and analyzed the data for both the dna and the sensemaking process and how we combined both methods. 3. research methodology we obtained recordings of all telephone conversations between all actors involved in the disruption. from these recordings, we selected the calls in which information on the switches and tracks was shared between actors. these recordings were transcribed (156 telephone calls in total). in addition, we carried out 9 in–depth interviews with actors involved in the case. respondents were selected on the basis of their different roles and their geographically different locations. the interviews were used to reconstruct the events of the day from their respective locations. all interviews were transcribed. we also studied all relevant written documents, such as shift reports, e-mail conversations and logs. finally, we attended a meeting during which actors reconstructed the day and shared their perspectives on the events. we observed this meeting and took detailed notes. in order to reconstruct the network, it is necessary to know “who talks to whom and at what time”. the telephone recordings offer rich and complete network data. most of the files included information on the specific actors communicating and the time of communication. we transcribed the recordings and then translated them into numerical data. using these data, we created an edge list containing the sources and targets of information flows and the time of communication. the telephone conversations don’t cover the communication between actors located in the same room. the interviews and documents fill in this data-gap. next, we created six time slices, each lasting half an hour, following the example of wolbers et al. (2013). comparison of the different time slices shows how the network evolved over time. we found that a thirty-minute time interval offered us enough detail to show the general communication dynamics. we also created a two-mode network that shows which actors (mode 1) were involved during certain time periods (mode 2) of the process. the two-mode network was recorded as an incidence matrix, marking the presence (1) or absence (0) of actors in the different time periods (seen as events). the relationship between the two modes shows how many actors were actively communicating during each specific period of time. to show the relative importance of a time slice, we divided the number of actors in a time slice by the maximum number of actors (borgatti & halgin, 2011). this two-mode analysis required us to use time slices of fifteen http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 35 university of bamberg press minutes in order to obtain a more detailed picture of the network development. we used the software package ora1 to structure the data. the metrics from the dna form the backbone for the analysis of the sensemaking process. to this end, we performed a qualitative analysis of the telephone conversations. this allowed us to identify which frames emerged, persisted or disappeared throughout the day, and how that was caused by both sensegiving and sensedemanding activities. first, we coded cues or occurrences that interrupted the expectations of actors regarding normal work practices. this is where actors collect and bracket information to get an initial sense of the interrupted situation (sandberg & tsoukas, 2014). we then focused on how these events were categorized as interruptions to define a specific situation. following cornelissen et al (2014), we coded words and expressions within communication that cued or prompted a cognitive or schema of interpretation. in this step we also looked for the factors influencing sensemaking as mentioned in the literature, i.e. identity and emotions. in the final step, we identified the actions taken by actors based on the framing of the interrupted situation and how this fed into the next phase. this allowed us to detect whether and how frames are updated with the help of new information. 4. introduction to the situation the dutch railway system is managed by the government-owned organization prorail, which manages the maintenance of the railway network, assigns capacity to the train operating companies (toc) and monitors and controls all train movements. maintenance has been outsourced to contractors, but is monitored by prorail’s track managers and track inspectors. railway traffic is controlled by thirteen regional traffic control centres. regional traffic controllers optimize traffic flows within their own region and train dispatchers are responsible for safe rail traffic on the sections assigned to them. the management of the railway system is decentralized, with considerable local autonomy. over the years, this has led to problems with local optimization and working at cross-purposes. a national control room, the operational control centre rail (occr), was established in 2010 to overcome such problems. in the occr, prorail and dutch railways (by far the largest toc in the netherlands) monitor the railway system at the national level. we present the main actors in the dutch railway system in table 12. table 1. the main actors involved in the case and a description of their role role abbreviation role description track manager wd 2 am is responsible for the quality of the railway infrastructure in order to assure safe usage track inspector supports the track manager by monitoring deviations and consulting him on corrective measures http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 36 university of bamberg press regional traffic controller traffic control is responsible for the optimization of rail traffic flows in specific geographical areas. train dispatcher is responsible for the safe allocation of railway tracks to trains, primarily by using signalling and controlling switches national asset manager coordinator riib monitors incidents, malfunctions and maintenance work and their impact on railway traffic at national level in the occr national traffic coordinator rlvl manages railway traffic at national level in the occr national traffic controller lvl monitors the railway traffic on the main corridors of the railway system in the occr power switching and monitoring centre smc takes calls on malfunctioning infrastructure and reports them to the contractors national train operating coordinator lbc manages the train operations for dutch railways at national level in the occr. team leader tl team leader of the regional traffic control centres node coordinator nc manages the shunting of trains at the large stations we began our study when a regular inspection was conducted by the contractor in area a on february 19th, 2014. the contractor noticed that two rail tracks did not meet the safety standards defined by prorail. this did not mean that there was an immediately unsafe situation, but action was required to assess the situation. the contractor informed the track inspector about these deviations at 8:30. at the same moment the track inspector was reading his monthly reports on four double switches (2 in area a and 2 in area b), which had been showing deviations from safety standards for some time. these switches were being monitored on a monthly basis. large-scale renewal of switches had been scheduled for some years, but they had been postponed due to a lack of funds. subsequently, the track team started to deviate from their own safety rules and had to increasingly rely on their own expert judgments. audits conducted by the human environment and transport inspectorate, in which they rebuked the maintenance team for not following their own rules, served as a wake-up call. this made the team aware of their own behaviour and changed their perspective on how to apply safety standards. consequently, the track inspector and track manager decided that the switches and tracks should immediately be taken out of service and that large-scale renewal was the only viable option left. the formal procedures prescribe that train dispatchers should be notified immediately in the event that rail infrastructure is no longer safe to be used. this allows them to immediately take the required safety measures to prevent trains from running over the designated switches and tracks. the smc should provide the train dispatchers with a specific reference number (rvo number), which is also used by the contractor. in view of the huge impact of their decision on the train service, the track inspector and track manager decided that the switches and tracks could be used until 18:00 to give themselves and others some preparation time. hence, they decided to issue an early warning to the occr, to let them coordinate the whole process. as the track inspector explained: “in my opinion when you call the occr, which is our big institute, the coordination centre, they will manage things. they will make sure that the loop is closed and the train dispatchers are informed.” http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 37 university of bamberg press however, the train dispatchers were not officially informed about the switches and tracks until just half an hour before the 18:00 deadline. minutes before the deadline, the train dispatcher in area a deemed it necessary to suspend all rail traffic in the middle of rush hour. we will analyze the events leading up to this decision. 5. identifying key moments and actors in the process we start our analysis by making a reconstruction of the flows of information between the different teams and organizations leading up to the decision. the quantitative network analysis acts as a first stage in our research to identify key moments and actors in the process, which serve as important starting points for a more in-depth qualitative analysis of how the information was processed. we identified a total of 156 instances of information sharing among 40 actors. to grasp the dynamics in the spreading of information we created six time slices of half an hour each (from 15:15 until 18:15). each time slice shows all the information exchanges between actors or nodes in that period. network graphs of each time slice, showing the development of the communication network over time, are presented in figures 4 through 9. in the figures each node represents an individual performing a specific role in the process. the round nodes are actors of traffic management, the square nodes are asset management, triangular nodes are dutch railways and the diamond-shaped nodes are contractors. the arrows show who provided whom with information during that specific time period. we then used several network metrics to quantitatively assess the development of the network over time. as figure 1 shows, the number of actors involved and interactions fluctuated throughout the process, with a peak in the number of interactions at time slice 6 (17:45-18:14). the spike in the number of actors involved and information exchange between time slices 2 and 3 (15:45-16:44), as well as the sudden drop at t4 (16:45-17:14), are especially remarkable and need further investigation. http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 38 university of bamberg press figure 1: number of actors and interactions throughout the process. figure 2 shows the density and betweenness centralization scores of the networks. density describes the number of links between nodes as opposed to the maximum number of linkages possible. a dense communication structure enables a free flow of information between actors and can therefore facilitate coordination. furthermore, a dense network also gives actors more opportunities to engage in sensemaking dialogue with others. as figure 2 shows, density increased slightly during t1 – t2, but then fell to five percent at t5. overall, the density is very low, which shows that many actors were not directly communicating with each other, i.e. the network was rather sparse, resulting in long communication lines. betweenness centrality measures the extent to which a particular node lies in-between the other nodes in the network. the more central an actor is, the more control he or she has over information flows through the network and the more able to coordinate group processes (hossain, wu, & chung, 2006). networks with a high betweenness centralization thus have one node or a small group of nodes that have more potential to control the flows of information. figure 2 shows that the network features high centralization at t1 and t2, before becoming more decentralized in the following time periods. the overall low percentages as opposed to the theoretical maximum indicate that there was little potential for a single actor or a small group to control the information flows. this was confirmed by one of the actors. national coordinator rail: “it is true that everyone had received some information, but they were just bits and pieces of information. we all knew that something was going on, but there was no one in control of the process.” 0 5 10 15 20 25 30 35 40 45 t1 t2 t3 t4 t5 t6 nodes interactions http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 39 university of bamberg press figure 2: density and betweenness centralization of the network over time. it is also important to identify the most central nodes or actors in each time slice. table 2 shows the three most central actors based on their degree and betweenness centralities. although there are differences in the rankings between the time slices, some actors show a high level of consistency in terms of centrality. these are the national asset manager coordinator (riib 2), national traffic controller (lvl 1), and the national traffic coordinator (rlvl) in the occr and the regional traffic controller of area a. their consistency can be seen as an indication of their ability to digest and distribute information in different time periods (wolbers et al., 2013). as such their central position in the network makes them important in terms of sensedemanding and sensegiving and we should therefore follow up on the role of these actors in the sensemaking process. table 2. the most central actors per time slice in terms of degree and betweenness centrality time slice degree betweenness t1 (15:15-15:44) 1. riib 2 2. rlvl 3. wd 2 am 1. riib 2 2. rlvl 3. lbc t2 (15:45-16:14) 1. riib 2 2. 3. 1. riib 2 2. 3. t3 (16:15-16:44) 1. traffic control a 2. rlvl 3. lvl 1 1. traffic control a 2. lvl 1 3. riib 2 t4 (16:45-17:14) 1. riib 2 2. traffic control a 1. traffic control a 2. lvl 1 0,00% 5,00% 10,00% 15,00% 20,00% 25,00% t1 t2 t3 t4 t5 t6 density centralization betweenness http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 40 university of bamberg press 3. lvl 1 3. rlvl t5 (17:15-17:44) 1. wd rbi vl 2. riib 2 3. rlvl 1. wd rbi vl 2. rlvl 3. riib 2 t6 (17:45-18:14) 1. rlvl 2. lvl 1 3. riib 2 1. rlvl 2. train dispatcher a1 3. riib 2 finally, we created a two-mode network on the basis of time slices of fifteen minutes to identify the critical moments of coordination in more detail. critical periods are the ones when many actors are connected together and information can be shared to create a common understanding. figure 3 shows the time slices ordered according to their normalized degree and betweenness centrality. degree centrality shows the number of direct ties that a node has. in this case nodes are the time slices and direct ties indicate the number of actors sharing information during that time period. for betweenness centrality calculations, we follow wolbers et al., (2013), by understanding time slices with a high betweenness centrality as critical periods in which actors could relay information to others since they were only linked to each other at that time period. figure 3: time slices degree and betweenness centrality in a two-mode network figure 3 shows that time slice 5 (16:15-16:29) features the highest degree centrality, with half of the actors involved during this time period. this means that the spike in the number of actors communicating occurred quite early on in the process. the large gap between the degree centrality of t4 and t5, represents a sudden increase in the number of actors involved during t5, as could also be seen in figure 1. during time slices t6 and t8 (16:30-17:14) there is a drop in the number of actors communicating, with the time slices t9-t12 (17:15-18:14) showing once again a large number of actors present in these time periods. the betweenness scores show that time slices 5 and 9-12 were critical periods of information sharing and collective sensemaking. we 17% 24% 7% 10% 50% 31% 29% 26% 45% 40% 36% 48% 8% 6% 0% 0% 10% 3% 5% 3% 12% 10% 12% 14% 0% 10% 20% 30% 40% 50% 60% t1 t2 t3 t4 t5 t6 t7 t8 t9 t10 t11 t12 degree betweenness http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 41 university of bamberg press used these critical periods to divide the whole process into specific episodes of collective sensemaking to study how actors make sense of a situation, the actions they subsequently undertake, and the revisions that may be made to these interpretations. figure 4: time slice 1 (15:15-15:44) figure 5: time slice 2 (15:45-16:14) http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 42 university of bamberg press figure 6: time slice 3 (16:15-16:44) figure 7: time slice 4 (16:45-17:14) http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 43 university of bamberg press figure 8: time slice 5 (17:15-17:44) figure 9: time slice 6 (17:45-18:14) 6. making sense of the decision to stop the train service having covered the structural and quantitative features, it is now time to turn to the content of the communication, i.e. the sensemaking processes. table 3 describes the process leading to the train dispatchers’ decision to stop the train service. this table is based on the one cornelissen et al. (2014) developed for their analysis of sensemaking in the stockwell shooting. using the findings from the dna, we divided the process into three episodes. the first episode started with the first phone call from the track manager to the asset management coordinator (riib) in the occr http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 44 university of bamberg press and ended at time slice four (figures 3, time 16:14) when the schematics with all the details were sent to the occr. episode 2 started at t5 (16:15), when there was a sudden peak in the number of actors and interactions, marking the start of the official procedure. the second episode ended at t8 (17:14), when it was discovered that no one had taken responsibility to inform the train dispatchers. the third episode consists of time slice t9-12 (17:15-18:14) during which there was considerable communication. it started with the track inspector informing the train dispatchers and ended with the decision by the train dispatchers in area a to suspend the rail traffic during rush hour. for each episode we looked at how the key actors (as identified in table 2) tried to make sense of the disrupting events or how they shaped the meaning construction of others (sensegiving). the last column describes the actions that followed from the initial sense made by the key actors, from which a new cycle of sensemaking commenced. table 3. overview of the episodes of sensemaking among key actors following the decision to take several rail tracks and switches out of service, and outcomes. time event key actor(s) sensemaking actions/ outcome 08:30 the track inspector receives the information that two rail tracks do not meet the safety standards, while reading the monthly maintenance reports on four switches which have deviated from the standard for quite some time. track inspector the track inspector feels that he can no longer look away and that concrete measures have to be taken, given the recent audits by the inspectorate and the worsened measurement results for the switches. the track inspector decides to consult with his superior, the track manager. 10:30 13:00 the track inspector discusses the measurement results with the track manager. track inspector and track manager after a long debate the track inspector and track manager conclude that it is too great a risk to continue using the switches and tracks. hence, they should be taken out of service, but not immediately given the impact this would have on the train service. the track inspector and track manager start giving the other parties in the rail system an early warning that the switches and tracks will be taken out of service at 18:00. episode 1: an early warning to the occr 15:19 the track manager (wd 2 am) gives an early warning to the asset management coordinator in the occr (riib). track manager and riib the riib was told that there was a very serious situation which would have a huge impact on the train service for which he used the term red flag often: “while you are seated, i want to share something with you (…). we have four red flags at this moment (in area a) and it’s pretty serious. there have been inspections last night, which showed major deviations from standard. in terms of responsibility we have to plant a red flag now.” the riib hands this information over to his colleague working the next shift (riib 2). 15:27 the riib 2 asks for track the riib does not hesitate to the riib warns his http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 45 university of bamberg press more information on the situation from the track manager and schematics showing which switches and tracks should be taken out of service. manager and riib 2 abide by the track team’s decision and wants to stay in control of the situation: “it’s just a fact that this red flag will be planted, but when we have more information we will be able to prepare our logistics.” colleagues in the occr of traffic control and dutch railways (rlvl and rlbc). 15:30 the rlvl calls the team leader of traffic control area a. rlvl and team leader the rlvl feels that there is still not enough information to justify alarming the train dispatchers: “we don’t know exactly what is going on and why, so we first want to gather more information. because imagine if it (red flag authors) isn’t necessary, then we would be making a lot of fuss about nothing, and i would regret that.” the rlvl tells the team leader to keep his eyes and ears open, without letting the train dispatchers know anything. 15:41 nevertheless, the information that a red flag will be planted in the rail tracks reaches the regional control centre of area a. regional traffic controller and rlvl both agree that since this is a safety issue, the train dispatcher should have been called first and that the occr shouldn’t be the first in the line of communication. they also find it strange that trains are still running over the switches and tracks despite the apparent safety risks. both the regional traffic controller and the rlvl still believe it to be a regular situation in which a mechanic has judged the switches and rail track as no longer safe and that the train dispatcher will be called anytime soon. 15:57 the riib receives schematics with details on the four switches and two tracks and discusses them with the track manager. riib and track manager the track manager once more warns that there are safety risks and that they actually should have acted immediately, but that they have given a deferment until 6 pm: “you know they have been running (over the switches and tracks) all morning, so i expect them to hold for another couple of hours. however, if the inspectorate saw these values, we would get a big fine.” the riib appreciates the postponement of the red flags in order to keep the train service in control. he shares the latest information with his colleagues in the occr during the standard meeting at the beginning of their shifts. episode 2: the official procedure is started 16:1816:43 the track inspector calls the smc to start the official procedure. track inspector and smc the track inspector enumerates which switches and tracks should no longer be used after 6 pm and that the surrounding switches should be locked to prevent trains from passing over the switches. when the smc asks who will inform the train dispatchers, the track inspector states that they have informed the occr and thus the train dispatchers should be aware of the situation and preparations should be ongoing to adjust the train service. based on advice from the smc it is decided to follow the simpler rvo procedure, instead of the more timeconsuming buta procedure, the latter being the normal procedure in the case of deferred maintenance work. in contrast to what the rvo procedure prescribes, the smc believes they have no role in informing the train dispatchers about the situation. http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 46 university of bamberg press 16:23 16:29 once again information on the red flags reaches the traffic control centres of areas a and b from the train operating companies. both regional traffic control centres are surprised to receive this information indirectly and to hear that the red flag has been deferred until 6 pm. regional traffic controllers of areas a & b and the national traffic controller (lvl). the national traffic controller assures the regional traffic controllers that the normal rvo procedure will be followed: “we expect a phone call from you later on that the train dispatchers have been informed on what will be taken out of service at 6 pm. i just received a whole bunch of information, but said sorry let them speak with the train dispatcher.” the rlvl and lvl feel that they have no role in the process until the train dispatchers have been informed. they therefore decide not to communicate with the regional traffic controllers until the train dispatchers have been informed and it is clear which switches should be taken out of service. 16:29 by now the train dispatchers in area a are starting to look for confirmation of the ‘troubling messages’ they have received about several red flags with the smc train dispatcher and smc the smc can’t give the train dispatcher more information and instead blames the train dispatcher for acting on rumours. he assures the train dispatcher that official procedures and lines of communication will be followed. the train dispatchers have the feeling that, given the many rumours, other actors have more information on the switches and tracks and that they are intentionally being excluded from the process: “i got the impression that everyone knew what was going on and we, the train dispatchers, knew nothing! who is responsible for safety? and how is it possible that we can use the switches until 6 pm?” 16:30 the riib asks the back office of prorail to make a logging in the communication system with the exact numbers of the switches and tracks that have to be taken out of service, which can be read by the regional traffic controllers. regional traffic controllers, train dispatchers, national traffic controller the regional traffic control rooms are still in doubt as to whether the information in the communication system is correct and if so, in which position the switches will be locked. the latter is of great importance in terms of rail capacity. hence, they find it difficult to formulate a contingency plan. the train dispatchers in area a even show reluctance to seek solutions, as long as they are not informed. given the high amount of uncertainty it is decided to prepare firm measures in order to stay in control of the train service. 17:08 after waiting and inquiring with the regional control rooms if they have received more details on the red flags, the rlvl urges the riib to ask the smc when they will inform the train dispatchers. riib and smc here there is a clash between the two different frames of the situation held by the smc and the riib. while the riib believes that the smc will inform the train dispatchers in accordance with formal procedures, the smc is under the impression that the occr has already informed the train dispatchers. both parties believe they are not authorized to impose restrictions on the usage of the switches and tracks. they conclude that the track inspector or a contractor does have the authority and therefore should do the job. the riib calls the track manager to let the track inspector inform the train dispatchers. episode 3: the train service is stopped 17:27 the track inspector calls a train dispatcher in area a to inform her train dispatcher area a and since the train dispatchers have been using the switches and tracks for the last couple of hours the train dispatchers in area a decide not to cooperate and continue to operate the http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 47 university of bamberg press of the situation. the track inspector mentions the switches and tracks that should be taken out of service at 18:00 and indicates that the rvos will be provided shortly. he also emphasizes that there are safety issues concerning the switches and tracks. track inspector they believe that there is no immediate risk and therefore do not feel obliged to cooperate. cooperating is also seen as an invitation to the track team to let them deviate from the procedures more often. moreover, the track inspector is not seen as an expert who can tell them to take the rail infrastructure out of service: “only experts can tell me if there is something wrong with the infrastructure (…). we need to hear it from an expert, which is a mechanic or the smc.” switches and tracks. the track inspector is told to get in contact with the team leader of area a. 17:30 the track inspector calls a train dispatcher in area b to inform him of the situation. once more, the track inspector says the switches that should be taken out of service at 18:00 and that the rvos will be provided shortly. he also mentions that there are safety issues with the switches and that all approach switches need to be locked. train dispatcher area b and track inspector for the train dispatcher it is inconceivable that the red flags have been deferred until 18:00 on the basis of its impact on the train service and workload of the operators, thereby giving less priority to the role of safety. the train dispatcher feels it is too great a risk to let another train pass over one of the switches, given the information provided by the track inspector. immediately after the phone call from the track inspector, the train dispatcher in area b decides to take the switches out of service. in other words: 30 minutes before the deadline. 17:40 a regional manager (wd vl rbi) calls the team leader of area a, but gets put through to one of the train dispatchers (a1). regional manager and train dispatcher the regional manager asks the train dispatcher what they need to take the switches and tracks out of service at 18:00. they are, however, unable to create a shared understanding of the situation. the train dispatcher expresses a lot of negative emotions, as she feels they are being left on their own: “the whole course of events is flawed and everyone is just actively cooperating. it (taking the switches and tracks out of service) should happen no matter what. this is just ridiculous!” the train dispatcher tells the regional manager to get in contact with the team leader as she refuses to cooperate 17:47 a mechanic calls a train dispatcher in area a to tell him that two switches have to be taken out of service because they don´t meet safety standards. however, the mechanic and train dispatcher in area a the train dispatcher repeatedly asks whether he can use the switch since a train is approaching it, but the mechanic can’t give any absolute certainty. given the high amount of uncertainty the train dispatcher is no longer certain whether the the train dispatchers see the mechanic as an expert on whose judgement they should act. however, since the mechanic also can’t provide exact details, the train dispatcher decides to let the train make an emergency stop as he believes it is no http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 48 university of bamberg press information he received from the smc is very sketchy and he is still unsure in what position the switches should be locked at 6 pm. at that moment a train approaches one of the switches mentioned by the mechanic. switches in his area of control are safe to be operated. longer safe. 17:54 the train dispatcher in area a calls the smc to check the information he received from the mechanic. train dispatchers area a and smc the smc assures the train dispatcher that the message on the switches and tracks comes from an expert, the track inspector. nevertheless, the train dispatchers feel that there is too much uncertainty regarding which switches are safe and which ones are not, and that they are thus unable to guarantee safe rail traffic. train dispatcher: “look, a situation is safe or unsafe! how is it possible that the people responsible for safety know nothing, at least not officially?” the dispatchers once more reject the track inspector as an expert, along with his information. they decide to suspend all rail traffic around the two major train stations in their area of control until they receive the correct information from a contractor, along with the official reference numbers, according to formal procedures. consequently, there are no services between various main cities during rush hour. 7. discussion as the summaries in table 3 show, even though it was rooted in good intentions, the track team’s decision to give an early warning to the occr created an ambiguous situation for the other actors in the system. the early warning violated expectations in several ways. firstly, an early warning is not a regular practice in the dutch railway sector. secondly, although the track team designated the situation as a ‘red flag’, they also allowed a six-hour delay. this sent a contradictory message. thirdly, informing the occr created a top-down flow of information, which deviated from the formalized bottom-up approach for maintenance work. in such situations people have to ask themselves and others: “what is going on?” as the case has shown, the term ‘red flag’ played a very important role in the sensemaking process. a red flag is jargon for a situation in which the safe usage of a railway track or switch can no longer be guaranteed and so trains are forbidden to run over the track or switch. by framing their actions as placing several red flags the track team thus created the impression of an immediate safety risk. as one of the managers explained: “a red flag can be like a red rag to a bull. a red flag means an unsafe situation”. labels, like the term red flag, carry their own implications for action. they focus attention and shrink the number of possibilities as to what is occurring (weick, 2001). in this case the term red flag triggered a routine procedure in which the train dispatchers should take the lead according to formal procedures and be informed by a mechanic or the smc. this frame was dominant among the actors in the occr throughout the entire process and was reinforced through their communication and actions. for instance, there was a strong commitment to restoring action to familiar practices, i.e. to make sure that the train dispatchers were officially http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 49 university of bamberg press notified by the smc or a mechanic in order to start the official procedure. this frame was also shared with the regional traffic control centres. even when the regional traffic control centres confronted the national traffic controllers with the absence of an official notification and the contradictory signals they had received, the national traffic controllers repeatedly reinforced the frame of a routine procedure. in fact, they decided to reduce communication with the regional control centres (as could be seen in figure 1, fourth time slice) when confusion started to increase, agitation grew among the regional operators, and communication became mainly focused on blaming instead of problem solving. the dominance of the frame also explains why the national asset management coordinator and the national traffic control coordinator did not use their central position in the network (table 2) to provide others with this crucial information, despite them having full details on the switches and tracks. they simply did not believe that it was their role, or that they had the authority to do so. instead, communication and actions were aimed at restoring standard procedures, which conflicted with the intentions of the track team for an improvised course of action with the occr coordinating the process and informing the regional control centres. the latter framing of the situation was also shared with the smc by the track inspector. as a result of these different interpretations of the situation, actors started to make wrong assumptions about what others knew and which actions they would take. consequently, the task of informing the train dispatchers was not assigned to anyone. in addition, the strategy of waiting until the train dispatchers had been officially notified was severely undermined by the many rumours that were circulating, because other actors in the system were checking, updating and revising their sense of events. the time slices show that the train dispatchers and regional traffic controllers in areas a and b were approached several times by train operating companies seeking confirmation of the ‘rumours’ they had heard about the red flags (figures 4 through 9). however, the train dispatchers and regional traffic controllers could not officially confirm any information to the train operating companies, as they had still not been officially notified of the situation. the sensedemanding of the train operating companies caused a chain of reactions, which explains the sudden increase in the number of interactions between time slices 2 and 3 (figure 1). more and more actors became involved and information on the red flags spread through the network uncoordinatedly. with information being dispersed among people and locations, sensemaking became fragmented, i.e. diverse accounts of the situation existed among the actors in the railway system. as a result, the train dispatchers felt isolated and lost grasp of what was happening. train dispatchers rely on a strong dichotomy between safe or unsafe, as they are held responsible for safe operation. hence, for them it is very difficult to understand that they are running trains over a piece of infrastructure, the safety of which cannot be guaranteed. although they knew that there were issues with the safety of some switches and tracks, they had not received any official information or a reference number, and therefore they could not fulfil their role. what explains the difference in response between the train dispatchers in areas a and b? the telephone conversations revealed that there were considerable negative emotions among the train dispatchers in area a, and this negativity increased during the day. these emotions were fuelled by the fact that critical information was not shared with them. this was not only seen as a threat to their social identity, but also created a state of anxiety which was widely shared among the train dispatchers in the control room. the literature on sensemaking points to the importance of emotions (maitlis & sonenshein, 2010; maitlis, vogus, & lawrence, 2013; maitlis & christianson, 2014). these studies show that emotions are an important factor in individual and collective sensemaking. negative emotions, in particular, are contagious and can easily spread http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 50 university of bamberg press among group members (bartunek, rousseau, rudolph, & depalma, 2006). in this case, a collective belief that emerged among the train dispatchers of area a was that non-compliance and strictly following procedures was the best course of action. hence, they continued to operate the switches and tracks and rejected the track inspector as an expert. such negative emotions were less prominent in area b. an important explanatory difference is that area b had a team leader onsite, while in area a the team leader was on call. as the team leader of area b explained: “after receiving the phone call from the rlvl i was busy tempering emotions, saying ‘guys keep on going, don’t get carried away by emotion because of this uncertainty about what we can and can’t do. make sure that the trains keep running!’” the lack of contextual information made it difficult for the team leader of area a, who was on call, to identify the specific coordination issues and to recognize the negative emotions emerging among his train dispatchers. in area a, the regional traffic controller played an important role in the line of communication between the train dispatchers and the occr, as can be seen in the graphs and table 2. the regional traffic controller, however, was very busy with his preparations to adjust the train service and showed resistance in his communication. therefore, national traffic control did not want to antagonize him. instead, new actors who were able to circumvent the formal lines of communication, such as the regional manager, stepped in to help create some common ground between the track team and the train dispatchers and to mediate in their conflict. however, the regional manager failed to develop a congruent understanding of the situation with the train dispatcher and actually contributed to the growing negative emotions. in the end the lack of a common ground between the track team and the train dispatchers resulted in the decision by the train dispatchers in both areas a and b not to comply (each in their own manner) with the track inspector’s six o’clock deadline. the unexpected split-second decision taken by the train dispatchers of area a to suspend the rail traffic, also cascaded into area b, where the dispatchers and regional traffic controllers were struggling to keep the traffic flowing. large service cuts had to be made in order to cope with the reduced capacity in an orderly fashion. as a result, many passengers were stranded. altogether, it took more than an hour for the train dispatchers in area a to receive the correct information and reference numbers so that they could gradually restart the train service. a nightly inspection of the switches by prorail revealed that three out of the four switches could be put back into service by applying new broadened safety standards that were scheduled to enter into effect two months later. 8. conclusion we have demonstrated how and why the actors involved were unable to create a shared understanding on which to base coordinated action. the occr’s framing of the situation as a normal procedure on the basis of the term ‘red flag’ and their desire to restore action to familiar practices was in conflict with the track team’s intention to find an improvised way of managing the process. we have shown how these different understandings of the situation accumulated over time, leading to inconsistent actions, incorrect assumptions and a lack of effective communication. our dna results show how information spread among the actors in the system, rapidly and uncoordinatedly. consequently, actors held different pieces of information and created fragmented accounts of the situation. we found that informal and indirect communication (sensedemanding) negatively influenced the process, as it increased uncertainty among the train dispatchers about the course of action to be followed and their role in the http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 51 university of bamberg press process. we observed two different responses to this uncertainty and the time pressure: one in which procedures were strictly followed and the track inspector was excluded as an authority, which eventually resulted in the decision to stop the train service (area a), and one in which safety concerns triggered improvisation (area b). the findings in this research confirm earlier work on networks of teams, in that there are additional challenges for effective coordinated action between multiple teams with a variety of skills, functions, and knowledge (shuffler, rico, & salas, 2014; zaccaro et al., 2012). in order for these teams to work together effectively, it is important to develop shared mental models on expected behaviour patterns concerning task procedures, team and team member behaviours and needs, and patterns of communication (rentsch & staniewicz, 2012). shared mental models thus help individuals to choose actions that are coordinated with other team members. as we show in our study, failure to develop a common set of assumptions and expectations may lead to role violations, communication failures, and even conflicts between teams. building shared mental models in a large system with many diverse teams can, however, be a challenge. it is therefore important to ensure that common understanding is established around a congruent framing of the situation through collective sensemaking (cornelissen et al., 2014). however, strict adherence to a framing can also have a negative outcome. commitment to frames reduces the number of cues that are considered and so it ties actors to a certain repertoire of actions and assumptions regarding the behaviour of others (cornelissen et al., 2014). in our case, adherence to the initial framing of the situation by the actors in the occr, and their attempts to restore procedures, along with the long and indirect lines of communication, created blind spots that led people to miss the signs that they were not dealing with a routine situation. hence, it is important to be able to update frames when dealing with ambiguous events (maitlis & sonenshein, 2010). this particularly applies to networks of teams. as teams are geographically separated it is often difficult to quickly identify misunderstandings and prevent escalation. hence, it is important that actors feel free to doubt and question the information they receive from their partners and to take the time to deliberate with them on the framing, instead of blaming each other for not following procedures or diminishing communication (weick, 1993; weick, 2005). this is not something that is easily achieved, but in the long run it can help to improve the adaptive capacity of the system. acknowledgement this research was carried out as part of the explorail research programme, within the managing complex system disruptions (macsyd) project. this project is co-funded by nwo and prorail. funding number 438-12-308 http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 52 university of bamberg press references abbasi, a., & kapucu, n. (2012). structural dynamics of organizations during the evolution of interorganizational networks in disaster response. journal of homeland security and emergency management, 9(1), article 23. bartunek, j. m., rousseau, d. m., rudolph, j. w., & depalma, j. a. (2006). on the receiving end sensemaking, emotion, and assessments of an organizational change initiated by others. the journal of applied behavioral science, 42(2), 182206. bechky, b. a. (2003). sharing meaning across occupational communities: the transformation of understanding on a production floor. organization science, 14(3), 312-330. bigley, g. a., & roberts, k. h. (2001). the incident command system: high-reliability organizing for complex and volatile task environments. academy of management journal, 44(6), 1281-1299. borgatti, s. p., & halgin, d. (2011). analyzing affiliation networks. in p. carrington, & j. scott (eds.), the sage handbook of social network analysis (pp. 417-433). london: sage publications. brown, a. d., stacey, p., & nandhakumar, j. (2008). making sense of sensemaking narratives. human relations, 61(8), 1035-1062. cornelissen, j. p., mantere, s., & vaara, e. (2014). the contraction of meaning: the combined effect of communication, emotions, and materiality on sensemaking in the stockwell shooting. journal of management studies, 51 (5): 699-736. de bruijne, m., & van eeten, m. (2007). systems that should have failed: critical infrastructure protection in an institutionally fragmented environment. journal of contingencies and crisis management, 15(1), 18-29. de bruijne, m. (2006). networked reliability: institutional fragmentation and the reliability of service provision in critical infrastructures (phd). dekker, s. (2006). the field guide to understanding human error. aldershot: ashgate. fiore, s. m., salas, e., cuevas, h. m., & bowers, c. a. (2003). distributed coordination space: toward a theory of distributed team process and performance. theoretical issues in ergonomics science, 4(3-4), 340-364. gioia, d. a., & chittipeddi, k. (1991). sensemaking and sensegiving in strategic change initiation. strategic management journal, 12(6), 433-448. gittell, j. h., & weiss, l. (2004). coordination networks within and across organizations: a multi‐level framework*. journal of management studies, 41(1), 127-153. hinds, p., & mcgrath, c. (2006). structures that work: social structure, work structure and coordination ease in geographically distributed teams. proceedings of the 2006 20th anniversary conference on computer supported cooperative work, 343-352. hossain, l., wu, a., & chung, k. k. (2006). actor centrality correlates to project based coordination. proceedings of the 2006 20th anniversary conference on computer supported cooperative work, 363-372. johansson, b., & hollnagel, e. (2007). pre-requisites for large scale coordination. cognition, technology & work, 9(1), 513. kontogiannis, t., & malakis, s. (2013). strategies in controlling, coordinating and adapting performance in air traffic control: modelling ‘loss of control’ events. cognition, technology & work, 15(2), 153-169. maitlis, s. (2005). the social processes of organizational sensemaking. academy of management journal, 48(1), 21-49. maitlis, s., & christianson, m. (2014). sensemaking in organizations: taking stock and moving forward. the academy of management annals, 8(1), 57-125. maitlis, s., & sonenshein, s. (2010). sensemaking in crisis and change: inspiration and insights from weick (1988). journal of management studies, 47(3), 551-580. maitlis, s., vogus, t. j., & lawrence, t. b. (2013). sensemaking and emotion in organizations. organizational psychology review, 3(2), 222-247. muhren, w., eede, g. v. d., & van de walle, b. (2008). sensemaking and implications for information systems design: findings from the democratic republic of congo’s ongoing crisis. information technology for development, 14(3), 197-212. powell, a., piccoli, g., & ives, b. (2004). virtual teams: a review of current literature and directions for future research. acm sigmis database, 35(1), 6-36. rentsch, j. r., & staniewicz, m. j. (2012). cognitive similarity configurations in multiteam systems. in s. j. zaccaro, m. a. marks & l. a. dechurch (eds.), multiteam systems: an organization form for dynamic and complex environments. (pp. 225-253). new york: routledge. roe, e., & schulman, p. (2008). high reliability management. stanford, ca: stanford university press. salas, e., burke, c. s., & samman, s. n. (2001). understanding command and control teams operating in complex environments. information knowledge systems management, 2(4), 311-323. sandberg, j., & tsoukas, h. (2014). making sense of the sensemaking perspective: its constituents, limitations, and opportunities for further development. journal of organizational behavior, 36(s1), s6-s32. http://dx.doi.org/10.20377/cgn-57 complexity, governance & networks – vol. 3, no. 2 (2017), p. 31-53 danny schipper, lasse gerrits: communication and sensemaking in the dutch railway system: explaining coordination failure between teams using a mixed methods approach doi: http://dx.doi.org/10.20377/cgn-57 53 university of bamberg press schipper, d., gerrits, l., & koppenjan, j. f. (2015). a dynamic network analysis of the information flows during the management of a railway disruption. ejtir, 15(4), 442-464. schulman, p., roe, e., eeten, m. v., & bruijne, m. d. (2004). high reliability and the management of critical infrastructures. journal of contingencies and crisis management, 12(1), 14-28. shuffler, m. l., rico, r., & salas, e. (2014). pushing the boundaries of multiteam systems in research and practice: an introduction. in m. l. shuffler, r. rico & e. salas (eds.), pushing the boundaries: multiteam systems in research and practice (pp. 3-16). bingley: emerald group publishing limited. vlaar, p. w., van fenema, p. c., & tiwari, v. (2008). cocreating understanding and value in distributed work: how members of onsite and offshore vendor teams give, make, demand, and break sense. mis quarterly, 32(2), 227-255. weick, k. e. (1993). the collapse of sensemaking in organizations: the mann gulch disaster. administrative science quarterly, 38, 628-652. weick, k. e. (2001). making sense of the organization. oxford: blackwell publishing. weick, k. e. (2005). 5 managing the unexpected: complexity as distributed sensemaking. in r.r. macdaniel jr. & d.j. driebe (eds), uncertainty and surprise in complex systems (pp. 51-65). berlin: springer. weick, k. e., sutcliffe, k. m., & obstfeld, d. (2005). organizing and the process of sensemaking. organization science, 16(4), 409-421. wolbers, j., & boersma, k. (2013). the common operational picture as collective sensemaking. journal of contingencies and crisis management, 21(4), 186-199. wolbers, j., groenewegen, p., mollee, j., & bím, j. (2013). incorporating time dynamics in the analysis of social networks in emergency management. journal of homeland security and emergency management, 10(2), 555-585. zaccaro, s. j., marks, m. a., & dechurch, l. a. (2012). multiteam systems: an introduction. in s. j. zaccaro, m. a. marks & l. a. dechurch (eds.), multiteam systems: an organization form for dynamic and complex environments (pp. 332). new york: routledge. endnotes 1 see schipper, gerrits & koppenjan (2015) for more details on the use of dna and ora in analyzing railway disruptions. 2 see schipper, gerrits & koppenjan (2015) for more details on traffic control in the dutch railway system http://dx.doi.org/10.20377/cgn-57 strengthening the institution-behaviour link in the ses framework to faciliate analysis of environmental public goods dilemmas complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 5 university of bamberg press strengthening the institution-behaviour link in the ses framework to faciliate analysis of environmental public goods dilemmas author: courtney r. hammond wagner rubenstein school of environment and natural resources, university of vermont, u.s.a., e-mail: chamwag@stanford.edu many environmental challenges take the form of environmental public goods dilemmas, including climate change, water quality deterioration and biodiversity loss. there is a great need for analysis of these challenges to better inform the design of governance institutions for adaptive and sustainable resource management. the social-ecological systems (ses) framework provides a foundational structure for analyzing the sustainability of complex and multi-scale environmental challenges. however, in its applications, the ses framework has struggled to facilitate analysis of environmental challenges beyond common-pool resource (cpr) regimes and the emergence of community-based governance institutions. in this paper, i propose that one way to facilitate application of the ses framework to environmental public goods dilemmas is to incorporate a greater focus on the link between institutions and behavior. first, i examine the attributes of environmental public goods dilemmas that differentiate them from cpr regimes. specifically, these are the absence of behavioral feedback, multi-actor and multi-resource system dynamics, higher levels of uncertainty and complexity, and lack of built-in social capital. i suggest that these same attributes also increase the need to study a broader suite of potential governance institutions. finally, i propose incorporating psychological and behavioral theory into the ses framework to facilitate an increased focus on the institution-behavior link. ultimately, a better understanding of which institutions promote behavior change within and across environmental public goods regimes can improve the sustainability of these systems. keywords: social-ecological systems; public goods,; environmental governance; institutions; decision making introduction many of the regional and global environmental challenges we face are environmental public goods dilemmas, including climate change, ocean acidification, declining water quality, and biodiversity loss (rockström et al., 2009; shortle & uetake, 2015). because of the scale and persistence of many environmental public good dilemmas, there is a critical need to improve our understanding of how institutions can support sustainable resource regimes in environmental public goods dilemmas. to improve our understanding of these systems we need analysis of environmental public good dilemmas to identify generalizable trends for robust and adaptive management regimes (ostrom, 2005). the social-ecological systems (ses) framework provides a foundational structure for this type of analysis, specifically for analyzing and improving the sustainability of complex environmental dilemmas (ostrom, 2007, 2009). the framework acknowledges the context-specific nature of human decisions and behavior within environmental dilemmas, and the existence of influences and feedbacks between the ecological and the social (ostrom, 2011). however, the ses framework has yet to fill this analytical need for environmental public goods dilemmas. i propose that a critical gap in applications of the framework lies in relying http://dx.doi.org/ mailto:chamwag@stanford.edu complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 6 university of bamberg press on behavioral assumptions about actor decision-making, in particular the lack of attention paid to how institutions shape decision-making and behavior. in an effort to inspire transdisciplinary analysis of environmental public goods dilemmas using the ses framework, i propose to improve the link between governance institutions and actor behavior within the framework. thus far, applications of the ses framework have been limited to the study of common pool resource (cpr) regimes and the conditions that lead to the emergence of community-based natural resource management (thiel, adamseged, & baake, 2015). in this paper, i will suggest that the attributes of environmental public goods dilemmas, in particular those that differentiate them from cpr dilemmas, increase the relevance of actor behavior and decisionmaking for social-ecological outcomes compared to the role of actor behavior in cpr regimes. i identify these attributes as the absence of behavioral feedback, multi-actor and multi-resource system dynamics, higher levels of uncertainty and complexity, and lack of built-in social capital. an expanded focus on the link between governance institutions and behavior within the ses framework will allow analysts to examine how institutions shape social-ecological outcomes in environmental public goods dilemmas, in light of their unique attributes. methodologically, the expanded ses framework that i am proposing for environmental public goods dilemmas includes (1) designing research questions around the institution-behavior link, (2) incorporating new variables into the ses framework on drivers, influences and psychological components of actor decision-making, and (3) utilizing actor mental models to identify the salient components of the social-ecological dilemma. the expanded ses framework allows the analyst to draw from the literature of social psychology, cognitive psychology and behavioral economics to investigate behavior under different institutions rather than rely on behavioral assumptions. this approach allows for the examination of diverse types of institutional arrangements in a broader range of environmental resource dilemmas, including environmental public goods. in section one, i begin by reviewing the vision behind the ses framework and outline its current constraints. in the second section, i explore the specific challenges in the management of environmental public goods for implementation of the ses framework. in section three, i build on the exploration of environmental public goods to suggest the need for examining diverse institutional arrangements with the ses framework to motivate collective action. in section four, i establish the importance of the institutional-behavioral link to address these challenges. finally, in section five, i propose revisions to the ses framework to strengthen researchers’ ability to examine the institutional-behavioral link in public goods dilemmas using the framework. ses framework vision & constraints the seemingly simple question of which institutions promote sustainability under which social-ecological contexts is hugely complex. variation in scale, scope, resource attributes, community attributes, market forces, and governance regimes, among other factors, makes drawing concrete conclusions and proposing solutions challenging. furthermore, researchers analyzing these systems from different disciplinary perspectives use different terminology, use different scales of analysis and focus on different variables, which make drawing system-wide transdisciplinary conclusions difficult (agrawal, 2003). the ses framework was proposed as a solution to this problem, following the success of a research program on cpr regimes (poteete, janssen, & ostrom, 2010). the framework provides a theoretical basis from which to examine interactions between ecological resource dynamics, underlying biophysical systems, governance regimes and human behavior (see figure 1 below) (ostrom, 2009, 2011). these first-order variables, and the secondand thirdorder variables nested below them, are organized to guide research http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 7 university of bamberg press design and data collection so that analysts can communicate across cases and begin to form theories about how ses work (for more detailed introduction to the ses framework see ostrom (2007, 2009)). the long-term goal of the ses framework is to enable research that can recognize “which combination of variables tends to lead to relatively sustainable and productive use of particular resources systems operating at specific spatial and temporal scales and which combination tends to lead to resource collapses and high costs for humanity” (ostrom, 2007, p. 15183). figure 1. the ses framework with example elements of watershed agricultural nonpoint source pollution dilemma adapted from mcginnis and ostrom (2014) thus far, the ses framework has struggled to live up to this initial vision. applications of the ses framework still lie mostly within the realm of cpr regimes (thiel et al., 2015), despite the intention for the framework to branch out to different types of resource regimes (mcginnis & ostrom, 2014). additionally, the framework has been applied primarily in community-based natural resource management contexts, such as when and under what conditions will resource users self-organize to address resource degradation or overharvesting (thiel et al., 2015). yet, the framework has the potential to examine a much wider breadth of governance questions. theoretically the ses framework centers around individual behavior and the choices made by individuals or collaborative groups (mcginnis & ostrom, 2014). the ses framework “does not dictate a particular model of decision-making; instead it prompts the analyst to explicitly identify what participants value; what resources, information, and beliefs they have, what their informationprocessing capabilities are, and what internal mechanisms they use to decide on strategies” (ostrom, cox, & schlager, 2014, p. 274). however, in its applications, very few studies explicitly identify actor perceptions that drive behavior and how actors perceive costs and benefits (thiel et al., 2015). this may be in part because, as ostrom (2009) observes, “accurate and reliable measures of users’ perceived benefits and costs are difficult and costly to obtain, making it hard to test theories based on users’ expected net benefits” (ostrom, 2009, p. 420). therefore, in applications of the ses framework, analysts tend to overlook individual perceptions and values that drive decision-making, despite the fact that these lie theoretically at the core of the framework. this relates to a broader http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 8 university of bamberg press challenge for the ses framework: the lack of a common understanding of what it means to apply the ses framework (ban & cox, 2017). in the following section, drawing from the vision and constraints laid out here, i explore the attributes of environmental public goods dilemmas that present challenges for applications of the ses framework. environmental public goods dilemmas in the disciplines of economics and political science, public goods are often defined as nonsubtractable (e.g. one person’s use of the good does not subtract from another person’s) and nonexcludable (e.g. it is difficult or impossible to exclude others from accessing the resource) (ostrom, 2005). another related way to consider environmental public goods dilemmas is that they are, at their core, environmental externalities. environmental externalities occur when a behavior in a specific domain results in an output outside or external to the domain in question. this external output results in the deterioration or degradation of an environmental good shared by all. one example of an environmental public good dilemma, which i will draw on for illustrative purposes throughout this paper, is water quality deterioration from agricultural nonpoint source (nps) pollution, as illustrated in figure 1. in this dilemma, farmers spread nutrient fertilizers on their fields to increase agricultural yield, but as a result, these added nutrients may runoff into nearby waterbodies and decrease water quality of rivers and lakes. environmental public goods, such as water quality, differ from the traditionally studied cpr regimes, such as irrigation networks, fisheries and forests, on a couple of important characteristics: they lack a clear behavioral feedback and they typically feature larger geographic scales, greater complexity and more uncertainty. public goods dilemmas lack of behavioral feedback cpr dilemmas also feature externalities at their core, but with an important distinction. within cpr dilemmas, overuse of resource results in degradation of that same resource for all. therefore, an individual who overharvests, say overfishes in a vulnerable fishery, will ultimately see reductions in their own ability to fish because of aggregate overfishing. as shown in figure 2 below, this can be conceptualized as a negative behavioral reinforcement mechanism. hardin described this situation in his classic tragedy of the commons paper: “each pursuing his own best interest” will bring “ruin to all” (hardin, 1968, p. 1244). http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 9 university of bamberg press figure 2. resource system and resource use relationships in common-pool resource and environmental public good. note the lack of behavioral feedback between undesirable outcomes in public good example. this feedback link, in which the behavior in question ultimately affects the ability to engage in that same behavior in the future, is not present with environmental public goods dilemmas, as show in figure 2. rather, environmental public goods dilemmas often feature many different types of actors using or exploiting the same resources for different purposes (young, 2002). in the water quality example, farmers use nutrients to produce agricultural products, and citizens more broadly enjoy water quality for recreation, aesthetics or drinking water, as shown below in table 1. importantly, the decline in water quality, in and of itself, will not limit a farmer’s ability to apply fertilizer on the farm. this is true even if a farmer is a part of the population that also enjoys water quality. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 10 university of bamberg press table 1. characterization of appropriation and provisioning action situations for agricultural watersheds. structure adapted from hinkel et al. (2015). actors benefits activity stock of resource units (ru) subtractability resource system (rs) action situation (s) farmers sale of crops/ products extracting nutrients from soil and adding nutrients through fertilizer/feed nitrogen and phosphorus no farm system addition of nutrients at the farm scale creates a water quality provisioning problem at the watershed scale water quality aesthetics, recreation, drinking water, local economic nutrient assimilative capacity of the waterbody yes watershed (biogeochemical processes) rural and urban citizens this lack of behavioral feedback in environmental public goods dilemmas creates challenges for characterizing the resource system within the ses framework because it is unclear which resource system dynamics drive behavior and at which scales. environmental public goods dilemmas generally do not feature the one-to-one resource system-resource unit link that is characteristic of most cpr regimes (e.g., resource unit: fish; resource system: fishery). hinkel et al.’s (2015) diagnostic approach for identifying the resource units, resource system, and provisioning and/or appropriation problem using the ses framework is not as clear cut for a public goods dilemma without this one-to-one link. environmental public goods dilemmas typically feature at least two nested resource systems, and often many more. table 1 shows the relevant action situation for the example case of declining water quality in agricultural watersheds. connected to this action situation are two distinct levels of resource system, resource units, activities, benefits and actors. the first resource system is that of the focal behavior: the farm in which a farmer is applying fertilizer to produce agricultural products. the second resource system is the higher-level system that absorbs the negative externality of the behavior in the first system: the watershed, which receives nutrients from the land into local waterbodies. similarly, the resource units are related, but have distinct characteristics. the units of nitrogen and phosphorus at the farm system scale are units of flow on and off the farm, whereas the nutrient capacity of the waterbody at the watershed scale is a stock that can be used up or exceeded. additional units of nitrogen and phosphorus added to farms will overwhelm the nutrient capacity of the waterbody but will not prevent farmers from adding additional nutrients to their farms in the future. importantly, what i am characterizing here as an environmental public goods dilemma could also be described as a number of other types of environmental dilemmas depending on perspective and context. as young (2002) suggests, “environmental problems are socially constructed in the sense that there are almost always a number of plausible ways to think about them, and the choice http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 11 university of bamberg press of conceptualizations is likely to have significant consequences for the interests of one or more members of the relevant group” (p. 142). for example, the issue of declining water quality in agricultural watersheds also bears elements of a cpr regime: the nutrient capacity of the waterbody can be viewed as a cpr that farmers and other landowners use (and overuse) for disposing of wastes (e.g., excess nutrients applied to fields or sewage). the key point here is that there can be multiple “correct” ways of viewing the system and that the study of these systems need to take into account these various perspectives. complexity and uncertainty in public goods dilemmas environmental public goods dilemmas tend to exist at larger geographic scales and feature greater levels of complexity and uncertainty than small-scale cpr regimes. all ses share elements of complex systems, including feedbacks, nonlinear dynamics and emergent system properties (meadows & wright, 2008). however, environmental public goods dilemmas tend to include multiple interacting resource systems and multiple user groups. accounting for multiple resources systems and user groups increases the complexity of interactions throughout the system. this increased complexity can further complicate the search for sustainable resource management regimes for environmental public goods dilemmas and confound our ability to understand system dynamics and measure system outcomes. to illustrate the complexity and uncertainty in environmental public goods dilemmas, i return to the example case of water quality deterioration in agricultural watersheds. the movement of nps nutrient pollution throughout a watershed is dependent on both deterministic factors, such as land use and soil composition, and random, factors such as weather-related processes. phosphorus and nitrogen molecules, whether originating from agricultural, urban storm water, or streambank erosion, are extremely difficult to differentiate from each other at the watershed scale (moss, 2008). therefore, it is both challenging and costly to attribute nutrient pollution to any one source (horan & ribaudo, 1999). the measurement of pollution is generally done at the watershed scale as farm scale models and measurements have high degrees of uncertainty (moss, 2008). yet, even at the watershed scale, modeling of diffuse nutrient pollution involves combining hydrological, geological, meteorological, land cover classification and other data sources across a complex, heterogeneous landscape, with trade-offs in specificity and uncertainty. typically, at the watershed scale, government agencies and researchers use a variety of modeling techniques to estimate and attribute nps pollution contributions to individual sectors, such as agriculture. management strategies within or across polluting sectors then rely on proxy measurements such as ambient water quality, or production-related measures, such as input-use or practice implementation (horan & ribaudo, 1999). furthermore, at the watershed scale, the transport of nutrients from farms to waterbodies may face significant time lags in the system, again resulting in a high degree of uncertainty and unpredictability (meals, dressing, & davenport, 2010). these attributes of the agricultural nps pollution, while unique to this specific problem, are representative of the broader challenges of uncertainty and complexity that make many environmental public good dilemmas more difficult to address than small-scale cpr regimes. the challenges outlined here, namely the multi-resource, multi-actor, highly complex and uncertain nature of environmental public good dilemmas, make them difficult to characterize within the ses framework. in particular, it is difficult to identify which elements of the system are important to examine for motivating collective action, and how to portray these systems given multiple potential ways to conceive of them. here i suggest, and will further describe in the fourth section below, that it is important to focus on the actors within the system, and particularly, to draw on their perceptions of the system to define the relevant elements driving behavior and collective http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 12 university of bamberg press action. before describing this proposal, i first need to consider another important and related gap within the applications of the ses framework: the need to apply the ses framework to examine a broader range of institutional interventions for ses. examining diverse institutions to motivate collective action ses issues are collective action problems. in collective action problems individuals’ pursuit of their own welfare leads to collective suffering. in other words, all individuals would be better off if they worked together. the collective action problem in agricultural nonpoint source pollution can be defined as: each farmer’s individual pursuit of a maximum yield, realized through excessive nutrient inputs, results in poor water quality that is shared by all farmers (and other residents of the watershed). overcoming collective action problems requires motivating individuals to pursue their collective welfare (olson, 1971; ostrom, 1990). in the literature on community-based natural resource management, collective action, or the need to collectively change actor behavior, is often used synonymously with useror self-organized community management. this body of work, led by ostrom and others, aimed to suggest that community-based, self-organized governance systems are a viable, potentially sustainable form of natural resource management. more broadly, community-based natural resource management, or self-organized management of resource systems, is just one type of institutional arrangement that can motivate collective action. collective action can be motivated in a number of different ways, including market-based strategies, statebased strategies, or any combination of these options with community-based strategies. according to ostrom, institutional arrangements should be context specific, hence the formation of the ses framework to guide analysis of which institutions promote sustainable system outcomes under specific conditions (ostrom, janssen, & anderies, 2007). the focus thus far on the emergence of community-based management using the ses framework is just a narrow slice of the potential institutional arrangements that could effectively manage resource systems. more broadly, there is a need to examine a diversity of potential institutional arrangements in environmental dilemmas. this is especially the case within the study of environmental public goods dilemmas, where community-based management strategies may not be the most effective or the most efficient means to address the issue. indeed, the traditional economics perspective dictates that the state should intervene to provision public goods because individuals face incentives to freeride off the contributions of others (singh et al., 2013). characterizing environmental public goods as dualor multi-resource system dilemmas, as i do above, illustrates an additional reason for this need: when compared to small-scale cpr regimes, the structure of environmental public good dilemmas does not lend itself to the emergence of social capital. small-scale cpr regimes typically consist of many opportunities for fostering social capital, which are not present in environmental public goods dilemmas. social capital is the concept that norms and bonds between people and communities are important for the functioning of society and can strongly shape interactions (pretty, 2003, p. 1913). in small-scale cpr regimes, actors typically engage with each other on a daily basis, inherently observing other actors’ behavior. furthermore, actors utilizing a resource are typically the very same individuals that are providing the public infrastructure for the resource (e.g. the head of an irrigation network is often an irrigator themselves) (anderies, janssen, & ostrom, 2004). in environmental public goods dilemmas, by contrast, these conditions that lead naturally to the building of social capital, such as norms and trust, are not necessarily present. in public goods dilemmas, resource users are not often the same as the public infrastructure providers. resource users may not be observing each other’s behavior on a daily basis. finally, it may be impossible to see the impact of each other’s behavior on the http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 13 university of bamberg press broader resource system. therefore, reciprocity, trust, norm creation, and repeated interactions may not be feasible mechanisms to create sustainable resource management regimes. take, once again, the example of water quality declines in agricultural watersheds: a farmer’s fertilizer use and nutrient management behavior takes place on her farm in isolation. it is difficult for other farmers to observe her nutrient management actions. additionally, public infrastructure providers, such as employees of government water resources agencies, are typically not farmers and do not interact with farmers on a regular basis. it may be difficult for farmers to make the connection between regular nutrient applications on their farm and broader water quality issues at the watershed-scale because of the complexity challenges outlined in the previous section. furthermore, water quality issues may be less salient to those farmers who farm higher up in a watershed when they don’t experience or witness water quality issues firsthand. therefore, the rule and norm creation, trust, reciprocity, and shared understanding may not exist amongst the farming community, making self-organized collective action on water quality very difficult. this suggests that there may be a place for other types of institutional interventions, such as regulatory or marketbased policies, to encourage or require collective farmer behavior change to improve water quality. i propose that to investigate diverse institutional arrangements using the ses framework, it is important to more closely examine this institution-behavior link. this focus on how and why institutions shape behavior and decision-making is what is needed to improve our understanding and adaptive management of environmental public goods. expanding the institution-behavior link an increased focus on actor decision-making within the ses framework can facilitate its application to both environmental public good dilemmas and a broader diversity of institutional arrangements. as described in the previous two sections, environmental public goods dilemmas differ from cpr regimes, the archetypal application of the ses framework, by a number of factors, including the lack of a behavior-reinforcing feedback, multiple resource systems and actor groups, complex and uncertain biophysical dynamics, and a lack of key interactions that build social capital. these elements of environmental public goods dilemmas increase the importance of understanding actor decision-making and behavior. in this section, i will first describe how an increased focus on decision-making can improve the applicability of the ses framework to environmental public goods dilemmas and diverse institutional arrangements. then i will describe a sampling of social-psychological and behavioral theories of decision-making to demonstrate how pulling from theory and models in these fields can improve our understanding of ses system outcomes through incorporation into the ses framework. in the last section of this paper, i outline how to incorporate these theories and models into the ses framework. the critical role of actor decision-making in ses the ses framework is designed around actor interaction and behavior, but as thiel at al. (2015) finds in their review of applications of the ses framework, actor decision-making processes, or at least actors perceptions of costs and benefits, themselves are typically not empirically analyzed. actor decision-making and behavior are important components of any ses because ultimately, actor behavior is a key driver of both social and ecological outcomes, and moreover, one of the major leverage points that humans have to intervene in ses. i propose that the attributes of environmental http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 14 university of bamberg press public goods dilemmas, as compared to cpr dilemmas, make actor decision-making and behavior even more relevant for ses system outcomes. as described above, public goods dilemmas lack the behavioral feedback present in cpr dilemmas as illustrated in figure 2. this link in cpr dilemmas serves as leverage to change selfinterested actors’ behaviors. it is possible to design incentives around this self-interest to motivate a behavior shift towards long-term individual (and social) payoffs. this, of course, is not a simple shift, and much research has been dedicated to designing incentives to solve this difficult problem (ostrom, 1990; ostrom et al., 2002; poteete et al., 2010). however, the shift to sustainable resource management for public goods dilemmas, which lack this behavioral feedback, can be even more difficult. without this self-interested mechanism to motivate behavior change, there must be another value-based motivator. a few possibilities include altruism, stewardship, and deterrence from fees or penalties. drawing again from the example of water quality and agriculture, when water quality declines, there is nothing physically preventing farmers from applying nutrients to their farms. it is possible that informing a farmer of the impact of nutrient applications on the lake could inspire behavior change due to a farmer identifying with stewardship values. whether or not this is the case would have implications for designing effective policy. identifying and understanding what motivates collective behavior in context-specific environmental public goods dilemmas is important for the design of institutional interventions to change behavior. the difficulty in defining behavioral motivation and incentives is further exacerbated by the multi-resource, multi-actor nature of many environmental public goods dilemmas. with different actors, using a resource, or multiple-related resources systems for different purposes, it is hard to decipher what the breadth of motivators are for distinct actor groups/resource uses. moreover, individual actors understanding of the ses is likely to vary actor to actor. the way in which an actor perceives the ses and their role in it is likely to influence their decision-making process and behavior. ostrom highlighted the importance of actors’ mental models and knowledge of the ses on system outcomes by designating these as variables in the framework (mcginnis & ostrom, 2014). these variables are likely to be of increased importance in studying environmental public goods dilemmas. here again, understanding the drivers, perspectives and values that comprise actors’ mental models and knowledge of the ses that underlie the decision-making process can help with identifying behavioral interventions. in environmental public goods dilemmas, ses are often larger scale, more complex, and more uncertain. this poses a challenge for designing sustainable governance solutions because often it is impossible to accurately measure ecological system outcomes. with water quality decline in agricultural watersheds, due to time lags in the movement of nutrients from farms to waterbodies, it may take decades for collective behavior change to result in water quality improvement (meals et al., 2010). in these cases, we often use models to project future ecological outcomes based on land use behavior. therefore, behavior change itself becomes the proxy for ecological outcomes. the central focus on behavior change in these systems suggests that understanding the drivers of behavior to then change behavior is the most direct pathway to improve ecological outcomes. finally, in environmental public goods dilemmas the lack of built-in processes for building social capital amongst actors calls for greater attention to how specific institutional interventions influence decision-making. the lack of built-in processes for building social capital suggest that there is less likelihood for the emergence of community-based collective action (pretty, 2003). this is not to say that trust, reciprocity, norms, and shared rules do not play a role in shaping behavior, but it does suggest that greater emphasis is required to understand where and how they play a role. in making the case for the relevance of decision-making and behavior in the analysis of environmental public goods dilemmas, it is important to note that the ses framework is compatible with a wide range of decision-making and behavioral theories (mcginnis & ostrom, 2014). this flexibility means that we can use a variety of decision-making and behavioral theories to better http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 15 university of bamberg press understand ses actor behavior and outcomes. however, so far, this flexibility has been underutilized. ostrom and her colleagues were aware of the importance of incorporating a broader understanding of human behavior into the study of ses. poteete et al. (2010), in their book on methods for studying collective action, identify this as one of the key next steps for the field. to do this, building on poteete et al. (2010), requires drawing on a broad range of psychological and behavioral theories to empirically examine the conditions and contexts that drive decision-making and behavior with ses. next, i will highlight a few psychological and behavioral concepts that may be particularly relevant for the study of environmental public goods dilemmas. this is not meant to be an exhaustive list, but rather a sampling to suggest the utility of this approach. for additional reviews of decision-making theories of relevance to understanding environmental dilemmas, i refer the reader to singh et al. (2013) and schlüter et al. (2017). promising decision-making concepts and theories for environmental public goods dilemmas a number of decision-making and behavioral models exist that could prove useful for incorporation within the ses framework for the study of environmental public goods and beyond. the body of research on collective action and common pool resource dilemmas, including ses framework applications, has typically relied upon the behavioral assumptions of bounded rationality to explain individual behavior, as suggested in section 2 above and as stated by ostrom (2005). this approach models decision-making as dependent upon limited information, cognitive processing, and attention in shifting individual cost-benefit analysis of potential actions (poteete et al., 2010). bounded rationality offers insight into the heuristics and biases that shape individual behavior, such as a greater aversion to losses than gains, anchoring on a given value rather than intrinsic values, or habit-formation (gsottbauer & van den bergh, 2011). within the ses framework, these aspects of decision-making may have important implications for institutional design, such as whether to design incentives to motivate behavior or sanctions, or what level to set a baseline incentive offer. social psychology also offers a number of decision-making models and theories, which are not mutually exclusive with bounded rationality. theories such as the theory of planned behavior (fishbein & ajzen, 2011), the value-belief-norm theory (stern, dietz, abel, guagnano, & kalof, 1999), and the norm-activation theory (schwartz, 1977) model individual behavior as embedded in individual’s beliefs and perceptions of the world. as a complement to bounded rationality, individuals’ beliefs and perceptions are grounded in an individual’s worldview and experience, as opposed to full information about any given decision-making situation. these social psychological theories suggest that constructs such as an individual’s attitudes toward a behavior, subjective and personal norms surrounding the behavior, and perceived behavioral control, or self-efficacy in engaging in a behavior are important predictors of how an individual will behave (bandura, 2000; fishbein & ajzen, 2011; stern et al., 1999). klöckner’s (2013) comprehensive action determination model combines elements of the theory of planned behavior, norm-activation theory and bounded rationality in an integrated decision-making model and has shown strong predictive power across a number of domains of environmental behavior. in environmental public goods dilemmas, where individuals are faced with high levels of uncertainty and ambiguity, social psychological decision-making theories can help identify the way in which individuals are making decisions in these highly variable conditions (e.g. based on other’s actions, their own level of understanding, what they think is right, etc.). furthermore, these theories can help to identify types or typologies of actors that value different types of information and http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 16 university of bamberg press assistance. this stands in stark contrast to an institutional rational choice model which assumes individuals to be self-interested and motivated by utility maximization (in practice mainly profit maximization). as a proof of concept, i will suggest a few psychological and decision-making concepts from the theory of planned behavior (fishbein & ajzen, 2011) and value-belief-norm theory (stern et al., 1999) that may be particularly helpful in examining actor behavior in environmental public goods dilemmas and describe their potential contribution. these are: self-efficacy and perceived behavioral control, experience and personal norms, and attitudes. self-efficacy and perceived behavioral control when considering actor behaviors in collective action dilemmas, an individual’s autonomy, their capacity to make their own decisions, is an important concept. autonomy is closely aligned with the concept of self-efficacy, for motivating behavior change. according to bandura, (1986, 2000) an individual’s incentive to act is dependent upon their perceptions that their behavior will have the desired impact. another closely related psychological concept is perceived behavioral control, which is an element of the psychological theory of planned behavior (ajzen, 1985). perceived behavioral control refers to “a general sense of personal competence or perceived ability to influence events” (fishbein & ajzen, 2011, p. 153). beliefs of self-efficacy strongly inform an individual’s confidence in their ability to perform a behavior and it is predicted that those with higher levels of perceived behavioral control with respect to a specific behavior or action would be more likely to adopt the behavior (ajzen, 1991). there is a strong link between individual selfefficacy, individual action and collective action. individual’s hold beliefs about collective efficacy within themselves, as opposed to some external representative. therefore examining individual’s beliefs about perceived behavioral control may be an important component of actor decisionmaking and behavior in ses (bandura, 2000). furthermore, different institutional arrangements intended to motivate collective behavior change in ses may impact actor self-efficacy and perceived behavioral control in different ways. this could have important consequences for overall ses system outcomes. experience and personal norms exposure or experience with a policy can be through participation in town hall meetings, planning committees, or public hearings, as well as measurement, monitoring or enforcement exercises can shape individual decision-making. edward-jones (2006) highlighted this as an important area for future research, in the light that engaging in a behavior due to policy requirements could continue to feedback on the policy goals after the end of policy. this is further in line with krosnick et al.’s (2006) attitude, certainty and existence model which includes personal experience and informant’s messages as predictors of general public support for a policy agenda. throughout a policy process and through engaging in target behaviors, individuals may be exposed to information that updates belief sets and norms to reinforce the target behavior. it is possible that exposure to a policy process activates personal pro-environmental norms. personal norms are a component of the norm-activation theory (schwartz, 1977) and value-belief-norm theory (stern, 2000) which suggest that given awareness of a behaviors consequences and personal ascription of responsibility for a given outcome, personal norms for a behavior will be activated and increase the likelihood that an individual will engage in the behavior. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 17 university of bamberg press attitudes attitudes are a central concept to many psychological models of decision-making, including the theory of planned behavior (ajzen, 1991). according to fishbein and ajzen, (2011) attitudes are the “latent disposition or tendency to respond with some degree of favorableness or unfavorableness to a psychological object” (p. 76). attitudes can be seen as an evaluation, on a scale from negative to positive, of a given action or behavior. in the context of actor behavior in environmental public goods dilemmas, actors may hold attitudes toward specific behaviors of interest that may influence whether or not they engage in the behavior. according to the theory of planned behavior, one would expect an individual with positive attitudes towards an action to be more likely to engage in that action (fishbein & ajzen, 2011). i propose that incorporating the study of actors’ attitudes, experience and personal norms, and self-efficacy and perceived behavioral control, along with other psychological theories, into the study of environmental public goods dilemmas using the ses can improve our ability to design institutions to promote sustainability in these systems. incorporating the institutional-behavior link into the ses what would it look like to expand the institutional-behavior link in the ses framework to apply it to environmental public goods dilemmas and to examine diverse institutional arrangements? given the flexibility of the ses framework towards decision-making and behavioral theories, the incorporation can take many different forms. here i recommend two potential approaches to strengthen the institution-behavior link in the ses framework. the first approach is to add social-psychological variables to the suite of actor attribute variables to test and explore the role of relevant psychological and behavioral theories in driving outcomes in ses. the second is to examine actor mental models within environmental public goods dilemmas to redefine ses framework categories to capture the relevant actor motivations and drivers of behavior. these recommendations complement each other and ultimately, a mixed methods approach combining both recommendations would be the most beneficial to improving our ability to identify elements of institutional design that lead to robust environmental public good regimes. the first recommendation, to add social-psychological variables to the ses framework, methodologically begins with framing research questions around the connection between institutions and actor behavior. for example, in the context of declining water quality due to agricultural activities, a potential set of institution-behavior questions could be: are farmers more likely to adopt water quality best management practices in mandatory or voluntary policy regimes? do farmer feelings of self-efficacy explain the difference in behavioral response to these policies? these questions would then drive the application of the ses framework to a series of cases to be compared. in table 2, i demonstrate the application of the ses framework to these example questions by listing the variables that could be used in defining and testing the relationship of interest. first, one would define the independent variables of interest, the bold, underlined variables in table 2. the first independent variable is the institutional variable: gs6 rules-in-use, which is defined in this example as being either mandatory or voluntary. then the second independent variable, “a10 actor(s) values and motivations,” is a new second tier variable that i am proposing to add to the framework under the first tier actor category. this new variable is where theories from psychology and behavioral economics, such as those reviewed above, can be incorporated into the framework as shown in table 2 below. drawing from the sampling of theories reviewed in the previous section, some potential third tier variables under “a10 actor(s) values and motivations” http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 18 university of bamberg press are personal norms, attitudes and self-efficacy. in the example case, self-efficacy can be included as an actor attribute, falling on a spectrum from high to low. table 2. example application of a revised ses framework to the case of declining water quality due to agriculture. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 19 university of bamberg press social, economic and political settings (s): s1 economic development, s2 demographic trends, s3 political stability, s4 government resource policies, s5 market incentives, s6 media organization resource systems (rs) governance systems(gs) rs1 sector: watershed gs1 policy area rs2 clarity of system boundaries gs2 geographic scale of governance system: state/region rs3 size of resource system: large gs3 population rs4 human-constructed facilities gs4 regime type rs5 productivity of system gs5 rule-making organizations: state/regional rs6 equilibrium properties: declining water quality gs6 rules-in-use: voluntary or mandatory rs7 predictability of system dynamics gs7 property-rights systems: private rs8 storage characteristics gs8 repertoire of norms and strategies rs9 location gs9 network structure gs10 historical continuity resource units (ru) actors (a) ru1 resource unit mobility a1 number of relevant actors: few to many farmers ru2 growth or replacement rate: continued nutrient applications a2 socioeconomic attributes: small to large farms ru3 interaction among resource units a3 history or past experiences ru4 economic value a4 location ru5 number of units a5 leadership/entrepreneurship ru6 distinctive markings a6 norms (trust-reciprocity)/social capital: existence of farmer ru7 spatial and temporal distribution: history of nutrient a7 knowledge of ses/mental models a8 importance of resource (dependence) a9 technologies available a10* actor(s) values and motivations: high or low self-efficacy interactions (i) -> outcomes (o) i1* resource use levels of diverse users: adoption of practices o1 social performance measures i2 information sharing among users o2 ecological performance measures i3 deliberation processes o3 externalities to other sess i4 conflicts among users i5 investment activities i6 lobbying activities i7 self-organizing activities i8 networking activities i9* public good provisioning levels: ambient water quality related ecosystems (eco): eco1 climate patterns; eco3 flows into and out of focal ses; eco2 pollution patterns http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 20 university of bamberg press notes: new additions to the framework, or revised categories are marked with an asterisk (*). the bold, underlined variables are the example independent variables. variables bold, italicized are example dependent variables. variables that are only bolded are covariates and variables are only italicized are held constant. all other variables would be excluded from the example analysis. ses framework adapted from mcginnis & ostrom (2014) the relationship of interest is the interaction of these two variables, and their effect on the dependent variable “i1* resource use levels of diverse users,” defined here as adoption of water quality best management practices. in order to better accommodate environmental public goods dilemmas, i also propose including this revised label category for i1 (“resource use”) as opposed to the previous label for the category “harvesting levels of diverse users”, since not all resource use behavior within an ses is harvesting behavior. furthermore, i suggest explicitly adding a variable to examine public good provisioning behavior: “i9* public good provisioning levels.” this variable could be another dependent variable of interest in the example case, defined as ambient water quality levels. with the independent and dependent variables of interest defined, the framework can then be used to guide case selection, by selecting cases that have similar resource systems, resource units, governance systems and actors. this is demonstrated by the resource system, resource unit and governance system variables in italicized text in table 2. these variables would define the criteria that all cases must meet to be included in the analysis. alternatively, or within the same analysis, some variables could be allowed to vary to examine covariates of the institution-behavior link of interest. an example for these is given by the bolded variables, including farm size for socioeconomic attributes (a2) and existence of a farmer group for social capital (a6). this approach, as defined here, lends itself to a quantitative analysis in which the relationship between the dependent variable (e.g. behavior or system outcomes) and the independent variables (e.g. policy type and covariates) is measured using a regression model, system dynamic model, network model or agent-based model (schlüter et al., 2017). the hope is that analyzing environmental public goods dilemmas with this revised version of the ses framework will ultimately inform adaptive governance. for example, the analysis can identify which institutions are adaptive for specific ses problems and their behavioral responses with the goal of producing the desired social and ecological outcomes. the key element to incorporating this new social-psychological second tier variable (a10) is data collection, as well as agreement by the research community on standardized constructs and validated questions. data will need to be collected from actors within the system of interest, either from surveys, interviews, experimental games or any other number of methods (see poteete, janssen and ostrom (2010) for a review of methods for studying collective action problems). in doing so, it is important that a core set of psychological constructs are include and questions are asked in the same way to provide internally valid comparisons. one potential example that can be informative for this is the new ecological paradigm, a standardized, broadly used, internally valid questionnaire to measure environmental concern (dunlap & van liere, 1978; dunlap, van liere, mertig, & jones, 2000; stern, dietz, & guagnano, 1995). this quantitative approach is one way that we can begin to better understand trends in how institutional design shapes decision-making, behavior and ecological outcomes in environmental public goods dilemmas. furthermore, experimentation with institutional designs can also inform adaptive governance of environmental public goods dilemmas. as further governance experimentation takes place, we can use this quantitative approach to study the institution-behavior link in these systems to continue to inform sustainable governance. the second recommendation for strengthening the institution-behavior link within the ses framework is to draw upon actor mental models to define the relevant system attributes in environmental public goods dilemmas. mental models of actors are acknowledged to play an important role in ses outcomes and are included in the framework (see variable a7 in figure 2) http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 21 university of bamberg press and in many applications of the framework (thiel et al., 2015). mental models are individuals’ internal representations of the world and are made up of concepts linked together, and it is these relationships between concepts that are used to make meaning of the world (carley & palmquist, 1992, p. 602). in environmental public goods dilemmas it is important to look to actors to define the concepts/variables that they perceive to be important influences on their behavior. again, actor behavior is important because ultimately it is actor behavior that policy-makers and society in general are looking to change to improve social-ecological outcomes. examining actor mental models lends itself to a qualitative methodologies, much like the indepth case studies ostrom and others pursued in the early work on cpr regimes (ostrom, 1990). interviews, focus group, and other ethnographic approaches can be used to elicit actors’ perceived motivations, interactions and system outcomes within an environmental public goods dilemma. the analyst can then using qualitative coding techniques such as grounded theory (strauss & corbin, 1994) or other forms of content analysis (see saldaña, 2015) to identify the salient or relevant aspects of the system according to actors. these elements can then be merged with the ses framework, either falling under existing second tier variables, or adding new ones as needed to better define the environmental public goods dilemma context. working up from actor perceptions of their own behavior and experience to the system level will allow analysts to identify institutions and contextual variables shaping system outcomes in environmental public goods dilemmas. ultimately these two recommended approaches to strengthening the institution-behavior link can be used in conjunction, or iteratively to improve our understanding of institutional design in environmental public goods dilemmas. conclusions the study of environmental public goods dilemmas and of diverse institutional arrangements can help us to identify design principles to improve the sustainability of these regimes. the ses framework was designed with the ambition to fill this need and facilitate the study of all types of ses. however, in application, the ses framework has struggled to facilitate the study of ses beyond cpr regimes and community-based natural resource management institutions. the same attributes of environmental public goods dilemmas that differentiate them from cpr regimes, namely the lack of a behavioral feedback, the multi-actor and multi-resource system dynamics, higher levels of uncertainty and complexity, and lack of built-in social capital, also increase the need to understand how a broader suite of institutions govern these systems. i have proposed that one way to address these linked challenges within the ses framework and facilitate the application of the ses to these types of systems is to expand the framework’s focus on the institution-behavior link. i suggest that this can be done through incorporating decision-making and behavioral models from psychology and behavioral economics into the ses framework and through examining actor mental models to define relevant system attributes. both of these recommendations will improve the ability of the ses framework to accommodate the analysis of more diverse resource regimes and facilitate the design of context-specific, adaptive institutional interventions to support sustainable resource management. acknowledgements the author would like to acknowledge funding from the social sciences and humanities research council of canada and the gund institute for environment. additionally, the author http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 22 university of bamberg press would like to thank aaron schwartz and hilary byerly for their valuable feedback on early versions of this manuscript. the funders were not involved in research design, data collection, data analysis, interpretation, manuscript preparation, or the decision to submit the article for publication. references agrawal, a. (2003). sustainable governance of common-pool resources: context, methods, and politics. annual review of anthropology, 243–262. ajzen, i. (1985). from intentions to actions: a theory of planned behavior. in p. d. j. kuhl & d. j. beckmann (eds.), action control (pp. 11–39). https://doi.org/10.1007/978-3-642-69746-3_2 ajzen, i. (1991). the theory of planned behavior. organizational behavior and human decision processes, 50(2), 179– 211. https://doi.org/10.1016/0749-5978(91)90020-t anderies, j. m., janssen, m. a., & ostrom, e. (2004). a framework to analyze the robustness of social-ecological systems from an institutional perspective. ecology and society, 9(1), 18. ban, n. c., & cox, m. (2017). advancing social-ecological research through teaching: summary, observations, and challenges. ecology and society, 22(1). retrieved from http://www.jstor.org/stable/26270044 bandura, a. (1986). social foundations of thought and action: a social cognitive theory. retrieved from http://psycnet.apa.org/psycinfo/1985-98423-000/ bandura, a. (2000). exercise of human agency through collective efficacy. current directions in psychological science, 9(3), 75–78. https://doi.org/10.1111/1467-8721.00064 carley, k., & palmquist, m. (1992). extracting, representing, and analyzing mental models. social forces, 70(3), 601– 636. https://doi.org/10.2307/2579746 dunlap, r. e., & van liere, k. d. (1978). the “new environmental paradigm.” the journal of environmental education, 9(4), 10–19. dunlap, r. e., van liere, k. d., mertig, a. g., & jones, r. e. (2000). new trends in measuring environmental attitudes: measuring endorsement of the new ecological paradigm: a revised nep scale. journal of social issues, 56(3), 425–442. edwards-jones, g. (2006). modelling farmer decision-making: concepts, progress and challenges. animal science, 82(06), 783–790. fishbein, m., & ajzen, i. (2011). predicting and changing behavior: the reasoned action approach. taylor & francis. gsottbauer, e., & van den bergh, j. c. (2011). environmental policy theory given bounded rationality and other-regarding preferences. environmental and resource economics, 49(2), 263–304. hardin, g. (1968). the tragedy of the commons. science, 162(3859), 1243–1248. https://doi.org/10.1126/science.162.3859.1243 hinkel, j., cox, m. e., schlüter, m., binder, c. r., & falk, t. (2015). a diagnostic procedure for applying the socialecological systems framework in diverse cases. ecology and society, 20(1), 32. horan, r. d., & ribaudo, m. o. (1999). policy objectives and economic incentwes for controlling agricultural sources of nonpoint pollution. jawra journal of the american water resources association, 35(5), 1023–1035. https://doi.org/10.1111/j.1752-1688.1999.tb04191.x klöckner, c. a. (2013). a comprehensive model of the psychology of environmental behaviour—a meta-analysis. global environmental change, 23(5), 1028–1038. krosnick, j. a., holbrook, a. l., lowe, l., & visser, p. s. (2006). the origins and consequences of democratic citizens’ policy agendas: a study of popular concern about global warming. climatic change, 77(1–2), 7–43. https://doi.org/10.1007/s10584-006-9068-8 mcginnis, m. d., & ostrom, e. (2014). social-ecological system framework: initial changes and continuing challenges. ecology and society, 19(2), 30. meadows, d. h., & wright, d. (2008). thinking in systems: a primer. chelsea green publishing. meals, d. w., dressing, s. a., & davenport, t. e. (2010). lag time in water quality response to best management practices: a review. journal of environment quality, 39(1), 85. https://doi.org/10.2134/jeq2009.0108 moss, b. (2008). water pollution by agriculture. philosophical transactions of the royal society of london b: biological sciences, 363(1491), 659–666. olson, m. (1971). the logic of collective action: public goods and the theory of groups, rev. ed. new york: schocken. ostrom, e. (1990). governing the commons: the evolution of institutions for collective action. cambridge university press cambridge: ostrom, e. (2005). understanding institutional diversity. princeton university press. ostrom, e. (2007). a diagnostic approach for going beyond panaceas. proceedings of the national academy of sciences, 104(39), 15181–15187. https://doi.org/10.1073/pnas.0702288104 ostrom, e. (2009). a general framework for analyzing sustainability of social-ecological systems. science, 325(5939), 419– 422. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 5-23 doi: http://dx.doi.org/10.20377/cgn-75 23 university of bamberg press ostrom, e. (2011). background on the institutional analysis and development framework. policy studies journal, 39(1), 7–27. ostrom, e., cox, m., & schlager, e. (2014). an assessment of the institutional analysis and development framework and introduction of the social-ecological systems framework. theories of the policy process, 267. ostrom, e., dietz, t., dolšak, n., stern, p. c., stonich, s., & weber, e. u. (2002). the drama of the commons. national academy press. ostrom, e., janssen, m. a., & anderies, j. m. (2007). going beyond panaceas. proceedings of the national academy of sciences, 104(39), 15176–15178. https://doi.org/10.1073/pnas.0701886104 poteete, a. r., janssen, m. a., & ostrom, e. (2010). working together: collective action, the commons, and multiple methods in practice. princeton university press. pretty, j. (2003). social capital and the collective management of resources. science, 302(5652), 1912–1914. https://doi.org/10.1126/science.1090847 rockström, j., steffen, w., noone, k., persson, å., chapin, f. s. i., lambin, e., lenton, t. m., scheffer, m., folke, c., schellnhuber, h. j., nykvist, b., de wit, c. a., hughes, t., van der leeuw, s., rodhe, h., sörlin, s., snyder, p. k., costanza, r., svedin, u., falkenmark, m., karlberg, l., corell, r. w., fabry, v. j., hansen, j., walker, b., liverman, d., richardson, k., crutzen, p., and j. foley. (2009). planetary boundaries: exploring the safe operating space for humanity. ecology and society, 14(2). https://doi.org/10.5751/es-03180-140232 saldaña, j. (2015). the coding manual for qualitative researchers. sage. schlüter, m., baeza, a., dressler, g., frank, k., groeneveld, j., jager, w., janssen, m. a., mcallister, r. r. j., müller, b., orach, k., schwarz, n., and n. wijermans. (2017). a framework for mapping and comparing behavioural theories in models of social-ecological systems. ecological economics, 131, 21–35. https://doi.org/10.1016/j.ecolecon.2016.08.008 schwartz, s. h. (1977). normative influences on altruism1. in l. berkowitz (ed.), advances in experimental social psychology (vol. 10, pp. 221–279). retrieved from http://www.sciencedirect.com/science/article/pii/s0065260108603585 shortle, j. s., & uetake, t. (2015). public goods and externalities: agri-environmental policy measures in the united states [oecd food, agriculture and fisheries papers]. singh, a., wilson, r., bruskotter, j., brooks, j., zwickle, a., & toman, e. (2013). the human dimensions of environmental degradation and ecosystem services: understanding and solving the commons dilemma. in ecosystem services and carbon sequestration in the biosphere (pp. 383–405). springer. stern, p. c. (2000). towards a coherent theory of environmentally significant behavior, journal of social issues, 56. stern, p. c., dietz, t., abel, t. d., guagnano, g. a., & kalof, l. (1999). a value-belief-norm theory of support for social movements: the case of environmentalism. human ecology review, 6(2), 81–97. stern, p. c., dietz, t., & guagnano, g. a. (1995). the new ecological paradigm in social-psychological context. environment and behavior, 27(6), 723–743. strauss, a., & corbin, j. (1994). grounded theory methodology. handbook of qualitative research, 17, 273–285. thiel, a., adamseged, m. e., & baake, c. (2015). evaluating an instrument for institutional crafting: how ostrom’s social– ecological systems framework is applied. environmental science & policy, 53, 152–164. young, o. r. (2002). the institutional dimensions of environmental change: fit, interplay, and scale. mit press. http://dx.doi.org/ strengthening the institution-behaviour link in the ses framework to faciliate analysis of environmental public goods dilemmas rubenstein school of environment and natural resources, university of vermont, u.s.a., e-mail: chamwag@stanford.edu how effective public managers tranform separation of powers into “inseparable powers“ in foster care administration networks complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 121 university of bamberg press how effective public managers tranform separation of powers into “inseparable powers“ in foster care administration networks author: rebecca padot1 1 assistant professor, department of history and government, misericordia university, pennsylvania, usa e-mail: rkowal@sas.upenn.edu results from a four state foster care administration field research study with over 55 key player interviews produced data on what particular networking practices public managers perform that contributes to foster care administration effectiveness. one of these practices was the concept of inseparable powers, whereby the traditional checks and balances roles of the judicial, executive, and legislative branches are slightly altered in eras of state-level foster care administration effectiveness. during a period of inseparable powers, effective public managers work across the state branch boundaries with partners from other branches to produce better foster care outcomes. keywords: networks: collaboration; governance; performance measurement; public management; foster care researchers have sought better understanding of whether public management networks work, how they work, and whether these networks ultimately improve government. (mcguire & agranoff, 2011). referring to the growth of network research in the public administration field, kapucu (2015) argues that we need more methodological rigor and empirical evidence. this case study on the foster care administration in the united states aims to answer that call of whether utilizing an external network, if at all, can lead public managers to producing better outcomes. researchers recognized that public management networks provide value for problem identification, information exchange, and mutual capacity-building (agranoff, 2007); advanced the understanding that governing by network is the new norm in the public sector (goldsmith & eggers, 2004); and connected managerial network to performance (o’ toole & meier, 2011). in o’toole (1997) wondered how the increasing role of networking in the public administrative world might affect managerial success. utilizing four foster care state cases in the united states, the goal in this paper is to better identify the networking practices public managers perform that leads to effectiveness. one way to measure networking practices is via analyzing external networks, defined here as identifying network formation across the three branches at the state level. this leads to the following research question: does an external network lead public managers to produce better foster care outcomes? in these u.s. foster care cases, the author tabulated data from the child welfare outcomes report published by the u.s. department of health and human services to identify states that produced better foster care outcomes regarding increasing permanency for children in foster care, reducing abuse and/or neglect, and increasing placement stability (u.s. dept. of hhs, child welfare outcomes: 2003). since outcomes studies of this sort are expensive to conduct, they are not conducted commonly. furthermore, political scientists have not paid much attention to the united states foster care system, particularly as it relates to outcomes studies. yet foster care remains a major federal, state, and local program which effects 700,000 persons per year in the care of the u.s. foster care agencies (u.s. dept. of hhs, trends, 1). foster care utilizes approximately $5 billion in federal spending via the u.s. department of health and human services (u.s. dept., fiscal year 2016, 7) and at least http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 122 university of bamberg press $4.1 billion in annual state and local spending (radel, personal communication, 2009).i this issue area is extensive enough to understand how public management practices with regards to networking relate to foster care outcomes. in conjunction with conducting field research in four case states in the united states (delaware, new york, michigan, and rhode island), this author obtained 57 interviews with key players connected to the foster care issue. there were at least twelve interviews per case state plus others that represented the federal government, foundations, and other foster care generalists. all interviewee data were reviewed to assess external networks in each of the four case states. although external networks could be identified in a variety of ways, external networks are identified here as network formation across the three branches in each case state. although these foster care administration cases concern the united states, the findings regarding networking, institutions, inseparable powers, the nexus, and public management are useful when studying other nationstates and their sub-governments. in the second section of this article, an overview of the theoretical background is provided, followed by the data collection and methods in the third section. the fourth section offers findings concerning these public manager practices and the fifth section offers a conclusion regarding public managers and external network formation. theoretical background the role of public management networks in government performance better public management can play a role in increasing the effectiveness of government (kaufman, 1967; moore, 1995; diiulio, 1994; kettl and milward, 1996). additional research then uncovered ideas about the role of the public manager and the variety of tasks they perform. there is a need for public administration to uncover the ways in which public officials (including public managers) interact with values, structures, regulations, and administration (mazouz & rousseau, 2016). defined as public value pragmatism, public managers were viewed by the value they produced, the context or the nature of the task (alford & hughes, 2008). additional tasks for a public manager included creating and guiding networks of deliberation and delivery as well as maintaining and enhancing the effectiveness of the system. (bryson, crosby, & bloomberg, 2014). one of the practices that these effective public managers performed is networking and collaboration. kettl (2009) identified public managers as rocket scientists as those who were able to obtain results through interrelated partnerships and those who crafted relationships on trust prior to the need for that trust. further, these networks were affecting the way government organizations operated and a public official now has less of a monopoly over a public domain (agranoff, 2004). in addition to the areas of problem identification, identification of extant technologies, enhancement/development of merging technologies, improving knowledge infrastructures, reciprocal strategies and programming, and joint policymaking/programming, public management networks also performed information exchange and mutual capacity building (agranoff, 2007). goldsmith and eggers (2004) recognize how these networks even affected the rise of third-party governance, streamlining processes, technology breakthroughs, and demand for greater choices. further, in order to establish these collaborative networks, public managers must view government systems from a multijurisdictional strategic perspective, in addition to, transforming their systems into collaborative systems (abels, 2012). there is precedent for public manager collaboration across agencies when the public managers can overcome impediments to practice. turf is often cited as a barrier to interagency collaboration (bardach, 1996; dryfoos, 1994). directly connecting managerial network to performance, public managers need to treat networking as a mandatory tool rather than an optional one if time permits (o’toole & meier, 2011). recognizing that there is complexity involved with collaboration, the http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 123 university of bamberg press factors that can affect a public manager’s outcomes include purpose or mission of the collaboration, trust, and perceived legitimacy (o’leary & vij, 2012). since the literature has acknowledged the value of public manager networks and collaboration, the research question posed here seeks to understand whether the public managers who utilized collaborative external networks produced better foster care outcomes. effective public managers as principled agents the research question looks at both effective and ineffective public managers and their use of external networking in the foster care cases. this warrants a definition for what sets apart effective public managers from ineffective ones based on existing literature. moore (1995) argues that a successful public manager works to increase the public value output of government organizations both short term, in addition to, long term. effective public employees, including superior public managers, have also been described as principled agents. in response to the principal agent rational choice theory (dunleavy, 1991; o’toole, 1996), diiulio (1994) created the term principled agent. he described principled agents as public employees who “do not shirk, subvert and steal on the job” even when consequences are lacking (p. 281). by also holding a range of positive characteristics as well, these principled agents “strive [work hard and go ‘by the book’], support [put public and organizational goals ahead of private goals], and sacrifice [go ‘above and beyond the call of duty’] on the job” (p. 281). in essence, a principled agent is a public manager who creates public value (moore, 1995). although every public manager is prone to error, a public manager is defined as a principled agent when they have a pattern of operating as such and of creating public value. this definition allows one to identify principled agents operating as effective public managers separately from ineffective public managers, and therefore, evaluating their use of external networks. state-to-state effectiveness studies there is precedent in the literature for the type of research design utilized in this study. however, due to cost and time, there are a limited number of these public administration and political science state-to-state variance studies that mine variance in government performance among state or city agencies. examples of existing studies include james q wilson’s varieties of police behavior (1969) which focused on 8 local police agencies, john j. diiulio jr.’s governing prisons (1987), which focused on 4 state prison administration systems, lawrence mead’s from welfare to work (1997), and stephen kelman’s procurement and public management (1990). this study is the first political science state-to-state variance study on the united states foster care. although the use of networks by public managers has research precedent as outlined above, this type of study to understand the role of external networks by public managers in foster care has no precedent. http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 124 university of bamberg press data collection and methods: the case of foster care administration case selection the four case states in this study (delaware, new york, rhode island, and michigan) were selected utilizing united states federal foster care data. delaware and michigan represented topperforming foster care states and new york and rhode island represented bottom-performing foster care states based on this outcomes data. yin (2012) notes that although multiple case research is harder to pursue, the findings are less prone to error. one type of multiple case design, according to yin, is intentionally contrasting cases. this foster care administration multiple-case study was crafted with contrasting cases of top and bottom performing states. there is a united states federal government database that offers foster care performance outcomes on each state within the united states. this dataset, child welfare outcomes 2003: annual report, was produced by the u.s. department of health and human services (hhs), children’s bureau. this report (u.s. dept. of hhs, child welfare outcomes: 2003), which was the latest published data at the time of initial research design,i utilized data from the afcars foster care file and the national child abuse and neglect data system (afcars) (u.s. dept. of hhs, adoption, n.d.) and featured seven outcome indicators with submeasurements on each state for the years 2000 to 2003. outcomes focused on reducing recurrence of child abuse and/or neglect, reducing time in foster care to adoption, reducing institutional placements, reducing time to reunification without increasing reentry, and increasing placement stability (u.s. dept. of hhs, child welfare outcomes: 2003). there was no state that regularly performed in the top five across all outcome measures. therefore, utilizing this database, the author was able to tabulate several foster care outcomes for all states to determine top and bottom performing states. because hhs reported the median on each state’s individual outcome, the author was able to look for states that consistently performed above or below the median by tabulating the outcomes that specifically related to foster care performance. top performing states had an overall lower ranking of lower rates of abuse, reduced time in foster care to adoption, reduced institutional placements, etc. bottom performing states had higher collective overall rates of abuse, increased time in foster care to adoption, and increased institutional placements, etc. the author also reviewed the outcomes data to determine whether foster care caseload size and foster care per capita were related to success. the finding was that this was not related and so this was not a determination in case state selection. once outcomes were tabulated across all states, the author ensured that state geographical size was not a predictor of success. after the top-performing and bottom-performing cases were identified, the author then selected one large population state from the top-performing outcomes group (michigan); one small population state from the top-performing outcomes group (delaware); one large population state from the bottom-performing outcomes group (new york); and one small population state from the bottom-performing outcomes group (rhode island). table 1: case states large population small population top performing michigan delaware bottom performing new york rhode island http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 125 university of bamberg press strong states, also known as top-performing states, were those that consistently performed above the median across a variety of foster care outcomes, whereas weak states, or bottomperforming states, were those that consistently performed below the median across a variety of foster care outcomes. the case selection and data for this article is based on data from the original research design of a state-to-state variance study utilized in a four-state foster care public administration study documented in padot (2015). for full quantitative outcomes data details, please refer to this book. data i conducted field research in 2007-2008 in each of the four case states by visiting the following locations: detroit; bronx; wilmington, de; milford, de; harlem; lower manhattan; providence, ri; east providence, ri: pascoag, ri; and rensselaer, ny. the purpose of the field research was to conduct interviews and to visit foster care agencies, state level executive, legislative, and judicial branches, and other organizations linked to foster care administration. the field research allowed the author to view child welfare documents, visit institutions, and obtain leads for additional interviews. i obtained a total of 57 interviews with key players connected to the foster care issue. the goal was to speak with public managers in foster care administration, such as the state foster care program manager and foster care agencies, and then to interview other major key players working on foster care in each case state. there were at least twelve interviews per case state plus others that represented the federal government, foundations, and other foster care generalists. most of these interviews were elite level interviews, including a governor; the director of a state health and human services agency; a state cabinet secretary; an hhs federal regional director for foster care; and executive directors of foster care agencies. interviews reflected a cross-sector approach so that information could be obtained from multiple vantage points. in other words, multiple accounts were utilized in order to build an assessment of a state’s foster care system. fenno (1978) practiced this type of in-the-field research when he interviewed united states house of representatives members while they were in their districts. he called his field research interviews exploratory, open-ended, and one in which he integrated the soaking and poking method. this latter method was referred to as simply hanging around the interviewees and institutions in the field. additionally, according to yin (1994), these interviews are generally openended in cases for which interviewers may seek facts and interviewee’s opinions about events. the interviews conducted for this study followed that rationale in that each interview was open ended and centered on the interviewee’s area of expertise in relation to their knowledge and involvement with foster care. rather than having a fixed set of questions for every interview, the author allowed the interviewees to speak to their specific involvement with foster care or foster care administration. when relevant to the type of interviewee, they were asked about whether and if they had any partners and networks in foster care. after interview requests were made via e-mail and phone, the author conducted most interviews in person. the length of these in person interviews was 15 minutes to 2 hours, with an average interview length of 1 hour. furthermore, telephone interviews were conducted based largely on leads identified from field research. in addition to personally conducting all interviews, the author typed or hand-wrote all interview questions and answers. when notes were hand-written, the author later transcribed them into a computer. for phone interviews, interview notes were transcribed at the time of interview. sixty-one interviewees were initially invited for a conversation, however, the author declined to interview two people as they were only available outside of this case study timeline. one interviewee did not respond to the repeated interview request and one interviewee declined. therefore, 57 interviews were completed suggesting an above-average key player response rate. some http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 126 university of bamberg press interviewees were willing to disclose additional quotes in exchange for confidentiality or for background data only. see list of interviewees referred to in this article in appendix a, with a full list of interviewees in (padot, 2015). previous findings one of the variables tested in the padot (2015) study was whether there were increased levels of public management in the two stronger foster care outcome performing states (delaware and michigan), and lower levels of public management in the two weaker foster care outcome performing states (new york and rhode island). in that study, increased levels of public management were defined based on public managers who operated as principled agents with lower levels of public management lacking the presence of public manager principled agents.ii the results in the study regarding this public management variable found that increased public management was present in stronger foster care outcome producing states (delaware and michigan), with decreased public management in weaker foster care outcome producing states (new york and rhode island). new data analysis there are several ways networks could be defined with regards to these four foster care administration cases. internal networks are those that are intra-branch, such as the networks that develop inside the foster care administration executive branch. external networks could be measured by the networked relationship between public managers and external non-profits (institutional networks), or between public managers and collaborators in other branches. this article specifically focuses on that latter type of external network, defined as identifying network formation across the three branches. this occurs when a state-level public manager in the executive branch forms networks with personnel in the state-level legislative or judicial branches. specifically, the author reviewed interviewee data regarding external networks that related to public managers who developed or did not develop networks across the three branches. findings: external networks by public managers this section provides interviewee data from the 57 interviews conducted by the author that relate to the external network patterns of public managers in the four case states. this section offers an overview of how the strong foster care outcome performing states (delaware and michigan) are utilizing external networks differently than the weak foster care outcome performing states (rhode island and new york). delaware as concluded in the initial published study (padot, 2015), there were increased levels of public management in the stronger foster care outcome performing state of delaware. at the delaware division of family services (dfs), field research uncovered the following beliefs held by workers towards their public managers in this top-performing foster care outcomes state: the director of the division of family services “legitimately cares about kids” (v. giampeitro, personal communication, august 24, 2007). “a lot of people move up in [in government] and tend to lose sight of the kids, but that’s not true for [john bates, the division of family services program http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 127 university of bamberg press manager]” (v. giampeitro, personal communication, august 24, 2007). delaware public managers utilized inter-branch networking to achieve better outcomes from foster children. there was an effort by the judiciary and the executive to have a “very improved” relationship with “natural tension [over cases] — and to agree to disagree” (confidential, personal communication, 2007-2008). both dfs and the courts debated which of them was responsible for decisions for the child. these two branches of government, the executive and the judicial were wrestling over how to define the best interest of the child. foster care, then becomes just another issue that can get caught in the gears of a federalist, separation of powers, checks and balances focused u.s. governmental system. the executive branch led by their self-created motto, “think of the child first.” whereas, the judicial branch assessed cases by a combination of “the best interest of public safety along with the best interest of the child” (confidential, personal communication, 2007-2008). the judicial branch and executive branch started out distrusting each other and disagreeing about how to handle cases, but, at some point, the two branches decided to work together. a trip funded by the pew charitable trusts, became the turning point when leaders in both branches began to build trust. before the trip, dfs would make decisions about placements unilaterally and ignore the court’s viewpoint. the court, on the other hand, had judge’s taking an activist role. some estimated that “approximately half the judges [would] push the envelope” (confidential, personal communication, 2007-2008). this judicial activism led to an increased focus on the best outcome for the child. over time the relationship between the two branches improved, with one employee of one branch stating, “i trust [the other] with my life.” (confidential, personal communication, 2007-2008). the court is now seen as “better than it was 25 years ago” (v. giampeitro, personal communication, august 24, 2007). furthermore, the “family court judges are coming off the bench and working to support kids. they worked with a community service organization [global youth help] to provide luggage for kids” (c. giddens, personal communication, august 15, 2007). both branches have also launched nongovernmental foster care websites,4 in order to work with more partners, and a showcase of how these public managers went above and beyond akin to principled agents. furthermore, the child protection accountability commission (cpac) was created by the legislature in the office of the child advocate (delaware, n.d.) with the goal to “monitor delaware’s child protection system to ensure the health, safety, and well-being of delaware’s abused, neglected, and dependent children” (delaware, 2007). rather than having a contentious relationship with other branches, cpac works with community partners and executive branch partners to ensure children are protected. in the previous study, i concluded that there were increased levels of public management in the stronger foster care outcome performing state of delaware, but the data here offers an example of the practices these public managers utilized regarding external networking. one of these practices was the concept of inseparable powers, whereby the traditional checks and balances roles of the judicial, executive, and legislative branches are slightly altered in eras of state-level foster care administration effectiveness. during a period of what the author terms here as inseparable powers, effective public managers work across the state branch boundaries with partners from other branches to produce better foster care outcomes. the results suggest for the case of delaware that when public managers utilize inseparable powers at the state level, better foster care outcomes are produced. http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 128 university of bamberg press michigan as concluded in the initial study (padot, 2015), there were increased levels of public management in the stronger foster care outcome performing state of michigan. marianne udow, director of department of health services (dhs), was viewed as making many improvements and “held on longer than we expected [udow] to – she only wanted to be there for a short time. she was more responsive than had been the case with previous directors.” (j. zehnder-merrell, march 18, 2008). udow pushed michigan to be a leader in the area of kinship care. she also led a skillman foundation project that looked at why there were uneven numbers of minority children in foster care and what dhs practices led to racial inequality. the state foster care public manager, mary chaliman, created the structured decision-making model, which leads caseworkers through a series of decisions for foster care children. this model became a 2003 finalist for the harvard university innovations in american government awards (government innovators network, 2003) further, michigan public managers utilized inseparable powers to achieve better outcomes for foster children. the michigan judicial system created the foster care review board (fcrb) to build a better relationship with the community while operating 30 review boards evaluating state foster care case performance. the fcrb is located in the judicial department but is state staffed and operated. hundreds of citizens serve as volunteers and some are even placed on the board. the fcrb frequently seats seven members surpassing the statute’s requirement of five. james novell, fcrb program manager, said: “we get amazing volunteers. kids are a sell. we’re part of pushing a boulder up a hill. we’ve made a difference in cases, i’m certain if we weren’t involved a case would go differently” (j. novell, personal communication, march 17, 2008). although the fcrb recommendations are “advisory only,” novell added that “we’ve got a lot of action on the last report” (j. novell, personal communication, march 17, 2008). novell is also active with creating regional judicial training and serving on an adoption placement committee that meets with both dhs and the courts (j. novell, personal communication, march 17, 2008). the fcrb is an example of an inseparable powers partnership in which the courts, the department of human services (dhs), and citizens have come together to improve foster care performance. the agency-judiciary relationship started out problematic, but then improved over time. there was an initial child and family services review during which dhs did not communicate with the judicial branch that it was going to review performance measurements by going inside the courts. justice corrigan said, “the first cfsr took me by surprise” (m. d. corrigan, personal communication, april 7, 2008). the communication improved over time and dhs began to reach out to the judicial branch. corrigan stated, “i think now we do have a coordinated strategy—so now we know what’s coming. we’re better prepared and more organized” (m. d. corrigan, personal communication, april 7, 2008). after being elected to the michigan supreme court in 1998, chief justice maura corrigan served as chief justice from 2001 to 2004 and then again beginning in 2006. it was in her role as chief justice to which she was appointed to the area of children and families. she was the most active chief judge of all the case states studied. akin to delaware, michigan’s executive branch public managers reached across the aisle to the judicial branch. dhs director udow said, “michigan was fortunate because [supreme court chief justice] maura corrigan was in office when i came in. she was really engaged in these issues and wanted to make things better” (m. udow, personal communication, march 18, 2008). the service on the pew commission was what corrigan said, “stirred my involvement [on foster care]” (m. d. corrigan, personal communication, april 7, 2008). she also weathered a big media crisis hit related to a number of children in foster care who were missing (m. d. corrigan, personal communication, april 7, 2008). this crisis motivated corrigan and her colleagues to create lists in all of michigan’s jurisdictions to track down runaways. serving on the pew commission allowed corrigan to meet with executive and judicial branch executives from many states to solve the problem of court delays. she found the main lesson from http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 129 university of bamberg press the pew commission to be the “silo effect,” in that these branches operated in silos in child welfare instead of partnering across branches (m. udow, personal communication, march 18, 2008). these branch leaders also learned that the leadership at the top of each branch was responsible for the partnerships. in 2005, corrigan was part of the leadership for a national center for state courts conference, which worked to improve child welfare in the states. each state organized a group to focus on crossbranch collaboration and then a federal bill was passed so that funding was tied to collaboration. united states senator charles grassley from iowa was a champion for this legislation. in 2007, there was a follow-up meeting to review the performance of the state’s success (m. d. corrigan, personal communication, april 7, 2008). corrigan’s leadership resulted in national impact. as she watched the judicial and executive branch relationships develop, michigan’s state foster care manager chaliman (m. chaliman, personal communication, march 18, 2008) said: udow really strengthened our relationship with maura corrigan and really ran with youth issues, kids exiting the system. we had legislation to have a task force and she was a very nononsense director. [udow] even brought in the national governors association. corrigan and udow led an effort to present before the national governors association (m. d. corrigan, personal communication, april 7, 2008). corrigan noted, “getting into the organization is greater than getting an individual governor to buy in” (m. d. corrigan, personal communication, april 7, 2008). they also were members of a task force dealing with children aging out of foster care. this resulted in them both changing their belief that children should age out at a later age, and their care be extended until age 22. corrigan (m. d. corrigan, personal communication, april 7, 2008) said: a lot of work in the area of adoption [is about how] to take kids to permanency—we have organized teams into 13 large jurisdictions to crash the dockets there. we’re taking identified adoptive parents and marrying them to the children. this is a collaboration with dhs—we’ve called it the adoption forum. we’re moving more children to adoption and older children to adoption quickly. when the author asked about whether dhs and the courts ever disagreed on case placements, the response was unusual. corrigan appeared surprised by the question. unlike the rhode island courts, michigan courts worked alongside dhs on foster care placements. corrigan said, “i frankly don’t want judges to jeopardize iv-e5 dollars—don’t touch a placement, that’s dhs’s call” (m. d. corrigan, personal communication, april 7, 2008). corrigan noted that her philosophy stemmed from both the family to family model and the jim casey youth opportunities initiative. these michigan courts were aware that their placement decisions for these children would directly affect how much money the state would receive from the title iv-e federal dollars. one interviewee said, “we had some courts send some kids home and then re-remove them so that they could make the proper iv-e budget. some courts push their weight around” (confidential, personal communication, 2007-2008). the michigan courts partnered with the executive branch to do what was best for the children, including the funding of their care. another initiative justice corrigan launched was to ensure that there was an in-person meeting for all foster children and their lawyers. previously, corrigan found that when she asked children if they had met their lawyers, they would say no. so, she reorganized the mechanism by which lawyers were assigned and then held lawyers to a higher standard. gayle robbert of the lansing foster care review board commented, “lawyers seeing their children didn’t happen before. that has improved. one of the supreme court justices is very interested, that’s why this has improved. she wanted [the attorneys] to come into compliance” (g. robbert, personal communication, august http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 130 university of bamberg press 8, 2008). corrigan (m. d. corrigan, personal communication, april 7, 2008) noted, “i ask kids now if they know their lawyer, [and] they say yes. i want to stand up and cheer.” once a new chief judge was hired, they had the authority to reassign the children and youth area to a new judge, however, the chief judge decided to keep corrigan in charge of that area. justice corrigan said, “if i left, do i think things would be different? i sure do” (m. d. corrigan, personal communication, april 7, 2008). there were also numerous examples in michigan of inseparable powers between the executive branch and the legislative branch, and in some cases even integrating the judicial branch. public managers utilized inseparable powers to achieve better outcomes for foster children. when former access nonprofit creator and now michigan department of human services director ishmael ahmed took office in september 2007, he made it a priority to build a relationship with the legislature. ahmed (i. ahmed, personal communication, march 18, 2008) said: “i came with a legislative relationship because i used to have to lobby with both [the] federal and state [governments]. i had a relationship with legislature on both sides of the aisle.” ahmed was focused on continuing his relationship with the legislature and believed that the legislature was taking the children’s rights lawsuit against their state seriously. the michigan state foster care manager, mary chaliman, said, “i think foster care has [received] attention no matter the budget cuts” (m. chaliman, personal communication, march 18, 2008). both a justice of the michigan supreme court maura corrigan (judicial branch) and dhs director marianne udow (executive branch) were heavily involved in the jim casey youth opportunities initiative that resulted in a partnership with the legislative branch. corrigan even recruited foster children to testify in support of the bill before the legislature. corrigan (m. d. corrigan, personal communication, april 7, 2008) said: we had a five-bill package pass in the senate 48–0, [meaning] it has five different reforms in foster care. it’s now in the house, now waiting for a vote on the floor. kids believe they have a future. this is getting through to kids. [a] marvelous partnership. the bills focused on subsidized guardianship, concurrent planning, and lawyer ad litem notifications (j. novell, personal communication, march 17, 2008). another bill “would shift the burden back on the state to terminate rights. it gives the state more time to think about it and expands the options available for the child.” (m. d. corrigan, personal communication, april 7, 2008). in this michigan case, public managers again utilized what the author terms inseparable powers to achieve better outcomes for foster children. like, delaware, the state of michigan had higher foster care outcomes and also utilized inseparable powers via state-level cross-branch collaboration. rhode island as hypothesized and concluded in the initial study, there were decreased levels of public management in the weaker foster care outcome performing state of rhode island. field research uncovered negative attitudes toward public managers. when the author asked one rhode island foster care agency … if she had brought up some of the problems stemming from various stipend rates across the state, she said “i have gotten nowhere on legitimate issues” (h. mulligan, personal communication, february 11, 2008). a chief casework supervisor for the department of children, youth, and families, maureen robbins, reported that the dcyf leadership proposed giving each caseworker a “$4,000 bonus per year in return for no limits on caseloads” (m. robbins, personal communication, january 24, 2008). the deal was struck which resulted in poor care for foster children as caseworkers could barely supervise their existing caseloads, let alone the new case overloads. http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 131 university of bamberg press further, public managers in this case did not utilize inseparable powers to achieve better outcomes for foster children. in delaware, the judicial and executive branches are in natural tension and work in tandem to determine the best action plan for the foster child. in rhode island, the courts and the agency work against each other by wrestling over control of case outcomes. according to the rhode island department of children, youth, and families (dcyf) policy officer in central management, there is a “lot of animosity between the family courts and the state. there is a fight going on” (d. hultine, personal communication, january 23, 2008). dana mullen is the permanency program manager for children’s friend and service who believed improving the court system should be the number one priority. the author asked her about the family court and the state’s relationship, and she replied, “or the lack . . . of a relationship?” (d. mullen, personal communication, february 5, 2008). she added, “we need to have a more collaborative relationship, better communication with [the court system]. they need to [better] understand what we’re up against” (d. mullen, personal communication, february 5, 2008). in rhode island, the courts and dcyf ignore each other’s rulings. the two cannot agree on whether 19-year olds are under the jurisdiction of dcyf or the judicial branch. during field research, in order to save money, the legislature ended state jurisdiction of children at age 18, and then determined that dcyf would provide services to existing children in care up until age 21. the dcyf public policy officer said, “the court is upset they no longer have jurisdiction, so they’re saying it’s prescriptive, which only applies to kids going forward not 19-year-olds currently under their jurisdiction” (d. hultine, personal communication, january 23, 2008). although the legislature objected, the court took jurisdiction over any older children already in state care. during january 2008 field research, the issue remained unsettled as to what branch was responsible for these 18 to 21-year olds. the current dcyf director at the time of field research, patricia martinez, noted that she has reached out to the family court. director martinez said, “we have a better relationship with the family courts. i meet with the chief judge once a month. i also pick up the phone and say, ‘your honor, i don’t think that’s the best placement for that particular child.’ ” (p. martinez, personal communication, february 5, 2008). however, tension lingers between the agency and the court. the dcyf senior casework supervisor philip steiner noted, “the family court still orders things that it thinks kids will need, regardless of whether they can be provided to them” (p. steiner, personal communication, january 24, 2008). dana mullen, the permanency program manager for children’s friend and service stated, “one of the things we’re trying to do is get buy-in from the court system. you can be advocating for one permanency plan and the judge out of nowhere rules otherwise and there’s no debating it once that’s made” (d. mullen, personal communication, february 5, 2008). lisa guillette of the rhode island foster parents association6 said, “we can’t count on rhode island family court to make appropriate decisions. foster parents have a right to be heard in court, but [they] can’t be. there’s no oversight by family court” (guillette, 2008). maureen robbins, chief casework supervisor stated, “we need to open up the cases to hold the judges accountable. child deaths are sometimes the court’s fault. judges order certain kids into certain foster care placements over the heads of the agency” (m. robbins, personal communication, january 24, 2008). the family court chief judge is jokingly referred to as “judge jeremiah for life” because judges receive lifetime appointments with no accountability (m. robbins, personal communication, january 24, 2008). interbranch conflict over the control of foster care policy further undermines any progress in this area generally, and the well-being of specific children. the state of rhode island had lower foster care outcomes and, also, utilized separated powers, whereby there is increased tension between the branches rather than state-level cross-branch collaboration as shown in delaware and michigan via inseparable powers. http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 132 university of bamberg press new york as concluded in the initial study, there were decreased levels of public management in the stronger foster care outcome performing state of new york. nick pirro, who was a former president of the new york state association of counties and a former president of the county executives association, said there was “very little involvement by the state” on foster care (n. pirro, personal communication, october 14, 2008). further, public managers did not utilize inseparable powers to achieve better outcomes for foster children. new york, identified as a weaker foster care outcome state showed some signs of improvement within new york city, the major hub of foster care activity in the state. this occurred during field research based on changing foster care administrations. with that being said, there were still zero examples from case state interviewees suggesting incorporation of inseparable powers as improvements began to take hold. secondly, the state operated in a vacuum of separated powers in that there were zero examples from the field research or case state interviewees suggesting the utilization of inseparable powers. unlike rhode island, there was no evidence of major inter-branch fighting, however, nor were there examples of the utilization of inseparable powers to achieve outcomes. conclusion the findings in the case studies show that public managers in the strong foster-care outcome producing states (michigan and delaware) utilized what the author terms inseparable powers, whereby the traditional checks and balances roles of the judicial, executive, and legislative branches are slightly altered in eras of state-level foster care administration effectiveness. when public managers utilize inseparable powers at the state level, better foster care outcomes are produced. whereas, the public managers in the weak foster-care outcome producing states (rhode island and new york) utilized separated powers, which occurs when there is fighting between the branches or no cross-branch collaboration at all. the public managers in all four cases had the same inter-branch institutions (executive, legislative, and judicial) in their states; however, in the strong states, the public managers made the choice to work across the institutional lines. the various branches could serve as natural institutional boundaries or could become boundaryless dependent on the public manager’s desire. the public managers in strong states choose inseparable powers. lynn (1981) notes that there are three blurry areas: where policy ends and administration starts, where the congressional role ends and the executive branch and its’ administration starts, or where the federal government ends and the state and local administration starts. this notion that branches are separate, that public managers operate within silos, that policy is solely administered by the executive, has long been challenged. lynn, along with his co-author bertelli, (2006) continued to wrestle with the role of a public administrator in relation to the separation of powers. although their premise reflects on the role of a public administrator, rather than more specifically a public manager, the concept that the separation of powers is an integral part of understanding public administration has precedent. rohr (1986) recognized that public administrators have multiple branches to interact with in order to obtain their outcomes. although he was speaking about the federal government, there are implications here for the three branches at the state level as well: rather than wait to be captured now by congress, now by presidents, now by the courts, statesmanlike administrators might consider delivering their agencies for a time to a constitutional master of their own choosing. which master the administrators would favor and for how long would depend on the administrators’ judgment of which branch of government needs to be http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 133 university of bamberg press strengthened to maintain the correct constitutional balance and to achieve the appointed ends stated so elegantly in the constitution’s preamble. (p. 89) in these four foster care cases, statesmanlike administrators (public managers) chose to function cross-branch to produce better foster care outcomes. at the state level in the case of foster care administration, separated powers operating in silos led to worse outcomes. the traditional checks and balances roles of the judicial, executive, and legislative branches are slightly altered in eras of state-level foster care administration effectiveness. although this study concentrates on 4 case states, there is some evidence to believe that the findings would be applicable to other states as well. during the author’s 2017 interview with the former chief judge of the city of richmond in virginia, anne holton, she noted that she “started out without a lot” of relationships with the executive branch child welfare services agency and then “we reached out to our city child welfare folks and they were a very regular part [of the conversation]” (a. holton, personal communication, june 7, 2017). she added “there was a lot of rock and roll to our relationship and social services had a lot of challenges” but “we worked together more and more closely over time” (a. holton, personal communication, june 7, 2017). holton’s judicial partnership with the executive branch’s child welfare agency resulted in a caseload drop of “800 something to 200 kids in foster care” (a. holton, personal communication, june 7, 2017). in other words, this trend of networking in the form of inseparable powers in foster care administration may reach beyond the four case states studied. further, the non-partisan united states national governors association, which serves as a conduit for the nation’s governors, has recognized and championed the utilization of some version of inseparable powers. the u.s. national governors association center for best practices, representing state governors, along with the united states national conference of state legislatures (national, 2019), representing state legislators, have co-hosted the three branch institute on child social and emotional well-being (nga center, 2014). the states selected for the program are given the task of improving foster care by integrating and aligning the work of all three branches of government (nga center, 2013). the u.s. national governors association and the u.s. national conference of state legislatures have recognized the value of working across branches at the intersection of the nexus. this external networks study focusing on collaboration across the branches offers several implications for future research in that: 1) these networks do not develop naturally, but are cultivated by those with often competing goals 2) complexity may arise as boundaries are movable, and 3) qualitative studies providing details on institutional boundaries can complement other forms of network research, such as social network analysis. the public management literature has alluded to a type of external networking by managers as upward networking. van den bekerom, schalk, and torenvlied (2017), recognize upward networking as one in which a manger conducts networking with superiors and politically necessary people. the challenge this presents for public managers is to determine the correct amount of effort to put into developing these external networks and to operate in an environment where the goals across branches may be in competition at times. further, these partnerships do not naturally develop but take effort on the part of public managers to cultivate relationships despite inherent tensions. this suggests that nexus research may benefit from assessing the efforts involved to create inseparable powers. more so, this challenge of working across the nexus on wicked problems, or what can be referred to as complex problems, extends well beyond these united states cases (rittel & webber, 1973). this research utilizes the term wickedness to encompass all the challenges associated with practicing across a nexus (kirschke, zhang, & meyer, 2018). in a global context, wicked problems often need thoughtful solutions that integrate a multitude of necessary partners across boundaries to implement the solution. wilson promulgated this concept of boundaries in government by recognizing that boundaries needed to be drawn in government (wilson, 1887). however, a http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 134 university of bamberg press continued research question remains as to how boundaries should be drawn in government, with or without external partners, and under what conditions to obtain optimal outcomes. kettl (2006) notes that the field of public administration has had boundaries on its’ mind. when boundaries are movable, complexity may arise. whereas, some systems behave in predictable ways, complex systems have many connections and interdependencies and behave in unpredictable ways (barbrook-johnson, 2018). this foster care administration case attests to that in that as the effective public managers chose to change the institutional boundaries, they were able to intersect effectively at the nexus. when public managers chose to maintain institutional boundaries, there was less complexity, but also less effectiveness. finally, this external networks study has implications for other forms of network research, such as the growing interdisciplinary field of social network analysis. the sna field has focused on new methods and approaches among actors in networks (butts, 2008) and the conversation specifically points to the need for more qualitative methods to balance the traditional quantitative-driven social network analysis literature (schipper & spekkink, 2015). this four-state foster care study suggests that qualitative case studies could offer insight regarding the role’s institutions play when crafting sna. ward, stovel, and sacks (2011) notes that network analysis is able to perform a needed task in the social sciences which is to assess the patterns of influence among people and institutions. institutional boundaries and boundaryless institutions are important when recognizing relationships among actors in that the structures can aid or hinder the development of networks. acknowledgements this research received financial support from the following sources: the earhart foundation, the mumford fund, the bradley foundation, the fox leadership program, prrucs at the university of pennsylvania, the faculty research grant at misericordia university and the summer research grant at misericordia university. the author previously worked at the pew charitable trusts but had no role in the administration, implementation, or funding of the trip cited herein. http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 135 university of bamberg press appendix a: list of interviews interviewees listed here may have been utilized for quotes, off-the-record data, and/or background data only. ahmed, i. (2008, march 18). director. department of human services. michigan. personal interview. lansing, michigan. chaliman, m. (2008, march 18). foster care program manager. department of human services. michigan. personal interview. lansing, michigan. corrigan, m. d. (2008, april 7). justice of the supreme court. michigan; chief justice of the supreme court. michigan. phone interview. giampeitro, v. (2007, august 24). program director. children and families first. personal interview. wilmington, delaware. giddens, c. (2007, august 15). director. division of family services. delaware. personal interview, wilmington, delaware. guillette, l. (2008, january 23). executive director. rhode island foster parents association. personal interview. east providence, rhode island. holton, a. (2017, june 7). former juvenile and domestic relations district court chief judge, city of richmond. former secretary of education, state of virginia. brief personal interview. philadelphia, pennsylvania. hultine, d. (2008, january 23). policy. central management. department of children, youth and families. personal interview. providence, rhode island. martinez, p. (2008, february 5). director. department of children, youth and families. rhode island. phone interview. mullen, d. (2008, february 5). permanency program manager. children’s friend and service. providence, rhode island. phone interview. mulligan, h. (2008, february 11). foster care recruiter. st. mary’s home for children. north providence, rhode island. phone interview. novell, j. (2008, march 17). program manager. foster care review board. courts. michigan. personal interview. detroit, michigan. pirro, n. (2008, october 14). former onondaga county executive. syracuse, new york area; former president of the new york state association of counties; former president of county executive association for new york; former chairman of the onondaga legislature. phone interview. radel, l. (2009, august 25). senior social scientist. u.s. department of health and human services. office of the assistant secretary for planning and evaluation. division of children and youth policy. e-mail correspondence. robbert, g. (2008, august 8). program representative. lansing foster care review board. courts. michigan. phone interview. robbins, m. (2008, january 24). chief casework supervisor. adoption and foster care preparation and support unit. department of children, youth and families. personal interview. providence, rhode island. steiner, p. (2008, january 24). senior casework supervisor. adoption and foster care preparation and support unit. department of children, youth and families. personal interview. providence, rhode island. udow, m. (2008, march 18). director. department of human services. michigan; director. center for healthcare quality and transformation. personal interview. ann arbor. michigan. zehnder-merrell, j. (2008, march 18). kids count in michigan. senior research associate. michigan league for human services. personal interview. lansing. michigan. http://dx.doi.org/10.20377/cgncomplexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 136 university of bamberg press notes 1. “title iv-e foster care fy 2007 expenditures as reported by states, may 2, 2008,” provided in august 25, 2009 e-mail to author by laura radel, u.s. department of health and human services. note: the $4.1 billion mentioned here only refers to the hhs reported amount of state spending (including local spending) on foster care. state and local spending is significantly higher when including foster care expenses for non-title iv-e eligible claims. hhs does not report overall data for state/local funded foster care, which is non-title iv-e eligible. 2. this remained the latest available comprehensive data until september 3, 2008, when the u.s. department of health and human services released the “child welfare outcomes 2002–2005: report to congress.” the most recent published report, “child welfare outcomes 2015: report to congress,” was released june 13, 2018. there is a significant delay from the year of the data collection to the report release date. 3. see earlier definition regarding principled agent theory. 4. during the period of field research, one of these websites was www.delawaregirlsinitiative.org. 5. corrigan is referring to title iv-e, an open-ended entitlement, which is the primary vehicle for federal reimbursement to states for a portion of their foster care costs. 6. after field research was completed, the rhode island foster parents association was renamed foster forward. http://dx.doi.org/10.20377/cgnhttp://www.delawaregirlsinitiative.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 137 university of bamberg press references abels, m. (2012). managing through collaborative networks: a twenty-first century mandate for local government. state and local government review, vol 44. issue 1. 29s – 43s. vol 44, issue 1_suppl, 29s 43s. doi:10.1177/0160323x12442517 agranoff, r. (2004). leveraging networks: a guide for public managers working across organizations. in j. m. kamensky and t. j. burlin (ed.) collaboration: using networks and partnerships, (pp. 61-102) lanham, md: rowman & littlefield. agranoff, r. (2007). managing within networks: adding value to public organizations. washington, dc: georgetown university press. alford, j. & o. hughes. (2008). public value pragmatism as the next phase of public management. the american review of public administration, vol 38. no 2. 130-148. doi:10.1177/0275074008314203 barbrook-johnson, p. & penn, a. (2018). a participatory systems map of the energy trilemma. center for the evaluation of complexity across the nexus, a cecan report for the department for business, energy, and industrial strategy, university of surrey. bardach, e. (1996). turf barriers to interagency collaboration. in d. f. kettl and h. b. milward (ed.), the state of public management (pp. 168-192). baltimore: johns hopkins university press. bertelli, a. & lynn, l. (2006). madison’s managers: public administration and the constitution. baltimore: johns hopkins university press. bryson, j. m., b. c. crosby, & l. bloomberg. (2014). public value governance: moving beyond traditional public administration and the new public management. public administration review, vol 74. iss. 4. 445-456. doi:10.1177/0275074008314203 butts, c. (2008). social network analysis: a methodological introduction. asian journal of social psychology, vol. 11. 13-41. doi:10.1111/j.1467-839x.2007.00241.x delaware state courts (2007, september 16). the office of the child advocate. history and statutory mission. retrieved from http://courts.delaware.gov/childadvocate/cpachistory.stm delaware state courts (n.d.). the office of the child advocate. welcome to the office of the child advocate of the state of delaware. retrieved from http://courts.delaware.gov/childadvocate/index.stm diiulio, jr., j. j. (1987). governing prisons: a comparative study of correctional management. new york, ny: the free press. diiulio, jr., j. j. (1994). principled agents: the cultural bases of behavior in a federal government bureaucracy. journal of public administration research and theory, vol. 4. no. 3. 277–318. doi:10.1093/oxfordjournals.jpart.a037210 dryfoos, j. g. (1994). full-service schools: a revolution in health and social sciences for children, youth, and families. san francisco: jossey-bass. dunleavy, p. (1991). democracy, bureaucracy and public choice: economic models in political science. london: prentice-hall. fenno, r. (1978). home style: house members in their districts. usa: harper-collins publishers. goldsmith, s., and eggers, w. d. (2004). governing by network: the new shape of the public sector. washington, dc: brookings institution. government innovators network. (2003). structured decision making. harvard kennedy school ash institute for democratic governance and innovation. retrieved from https://www.innovations.harvard.edu/awards programs/overview?id=53271 kapucu, n. (2015). introduction: social network analysis applications in complex governance networks. complexity, governance, & networks, vol. 02. iss. 01. 1-4. doi:10.7564/15-cgn29 kaufman, h. (1967). the forest ranger. baltimore, md: johns hopkins university press. kelman, s. (1990). procurement and public management: the fear of discretion and the quality of government performance. lanham, md: university press of america. kettl, d. f. (2006). managing boundaries in american administration: the collaboration imperative. public administration review, dec 2006: special issue. doi:10.1111/j.1540-6210.2006.00662.x kettl, d. f. (2009). the next government of the united states: why our institutions fail us and how to fix them. new york: w.w. norton and co. kettl, d. f. & milward, h. b. (1996). the state of public management. baltimore, md: johns hopkins press. kirschke, s., l. zhang, & k. meyer. (2018). decoding the wickedness of resource nexus problems-examples from watersoil nexus problems in china. resources, vol 7. 67. 1-15. 10.3390/resources7040067 lynn, jr., l. e. (1981). managing the public’s business: the job of the government executive. new york: basic books. mazouz, b., & a. rousseau. (2016). strategic management in public administrations: a results-based approach to strategic public management. international review of administrative sciences, vol. 82(3). 411-417. doi:10.1177/0020852316655522 mcguire, m., & r. agranoff. (2011). the limitations of public management networks. public administration, vol 89. no 2. 265-284. doi:10.1111/j.1467-9299.2011.01917.x mead, l. (1997). from welfare to work: lessons from america. london: institute of economic affairs. http://dx.doi.org/10.20377/cgnhttp://libproxy.misericordia.edu:2182/10.1177%2f0160323x12442517 http://dx.doi.org/10.1177/0275074008314203 http://dx.doi.org/10.1177/0275074008314203 https://doi.org/10.1111/j.1467-839x.2007.00241.x http://courts.delaware.gov/childadvocate/cpachistory.stm http://courts.delaware.gov/childadvocate/index.stm https://doi.org/10.1093/oxfordjournals.jpart.a037210 https://www.innovations.harvard.edu/awardshttps://www.innovations.harvard.edu/awardshttps://doi.org/10.1111/j.1540-6210.2006.00662.x http://dx.doi.org/10.3390/resources7040067 http://dx.doi.org/10.1177%2f0020852316655522 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 121138 doi: http://dx.doi.org/10.20377/cgn-89 138 university of bamberg press moore, m. (1995). creating public value: strategic management in government. cambridge, ma: harvard university press. national conference of state legislatures. (2019, july 19). about us. retrieved from http://www.ncsl.org/aboutus.aspx the nga center for best practices. (2013, may 28). states to improve social well-being of children in foster care. retrieved from http://www.nga.org/cms/home/news-room/news-releases/2013--news-releases/col2content/states-toimprove-social-well-be.html the nga center for best practices. (2014, july 2). three branch institute on child social and emotional well-being. retrieved from http://www.nga.org/cms/home/nga-center-for-best-practices/meeting--webcastmaterials/pageehsw-meetings-webcasts/col2-content/main-content-list/three-branch-institute-on-chi1.html o’leary, r., & n. vij. (2012). collaborative public management: where have we been and where are we going? the american review of public administration, vol. 42(5). 507-522. doi:10.1177/0275074012445780 o’toole jr., l. j. (1996). rational choice and the public management of interorganizational networks. in d. f. kettl and h. b. milward (ed.) the state of public management (pp. 241-263), baltimore, md: johns hopkins press. o’toole, jr., l. j. (1997). treating networks seriously: practical and research-based agendas in public administration. public administration review, vol 57. no. 1. 45-52. doi:10.2307/976691 o’toole, jr., l. j. & k. j. meier. (2011). public management: organizations, governance, and performance. cambridge: cambridge university press. padot, r. (2015). the politics of foster care administration in the united states. routledge research in public administration and public policy series. new york: routledge press. rittel, h. w., & m. webber. (1973). dilemmas in a general theory of planning. policy sciences vol 4. 155-169. doi:10.1007/bf01405730 rohr. j. (1986). to run a constitution: the legitimacy of the administrative state. lawrence, ks: university press of kansas. schipper, d. & w. spekkink. (2015). balancing the quantitative and qualitative aspects of social network analysis to study complex social systems. complexity, governance, & networks, vol 2. iss 1. 5-22. doi:10.7564/15-cgn23 u.s. department of health and human services. (n.d.) administration for children and families. administration on children, youth and families, children’s bureau. adoption and foster care analysis and reporting system (afcars). retrieved from http://www.acf.hhs.gov/programs/cb/systems/afcars/fedguid.htm u.s. department of health and human services. (n.d.) administration for children and families. administration on children, youth, and families, children’s bureau. child welfare outcomes 2003: annual report. safety, permanency, well-being. retrieved from http://www.acf.hhs.gov/programs/cb/pubs/cwo03/cwo03.pdf u.s. department of health and human services. (n.d.) administration for children and families. administration on children, youth, and families. children’s bureau. trends in foster care and adoption—fy 2008–fy 2017. retrieved from https://www.acf.hhs.gov/cb/resource/trends-in-foster-care-and-adoption u.s. department of health and human services. fiscal year 2016 budget in brief. administration for children and families: mandatory programs. retrieved from https://www.hhs.gov/about/budget/budget-in brief/acf/mandatory/index.html van den bekerom, p., schalk, j., & torenvlied, r. (2017, june 16). transforming input into output: how downward networking mediates the effect of external networking on organizational performance. public performance and management review, 40:4. 625-651. doi:10.1080/15309576.2017.1305910 ward, s. k. stovel, & a. sacks. (2011). network analysis and political science. annual review of political science, 14: 245-264. doi:10.1146/annurev.polisci.12.040907.115949 wilson, james q. (1969). varieties of police behavior: the management of law and order in eight communities. cambridge: harvard university press. wilson, woodrow. (1887). the study of administration. political science quarterly. vol. 2. no. 2. 197-222. yin, r. k. (1994). case study research: design and methods, 2nd ed. applied social research methods series, vol. 5, thousand oaks, ca: sage publications. yin, r. k. (2012). applications of case study research, 3rd ed. thousand oaks, ca: sage publications. http://dx.doi.org/10.20377/cgnhttp://www.ncsl.org/aboutus.aspx http://www.nga.org/cms/home/news-room/news-releases/2013--news-releases/col2-%09content/states-tohttp://www.nga.org/cms/home/news-room/news-releases/2013--news-releases/col2-%09content/states-tohttp://www.nga.org/cms/home/nga-center-for-best-practices/meeting--webcast-%09materials/page-ehswhttp://www.nga.org/cms/home/nga-center-for-best-practices/meeting--webcast-%09materials/page-ehswhttp://dx.doi.org/10.1177%2f0275074012445780 http://dx.doi.org/10.2307/976691 https://doi.org/10.1007/bf01405730 http://dx.doi.org/10.7564/15-cgn23 http://www.acf.hhs.gov/programs/cb/pubs/cwo03/cwo03.pdf https://www.acf.hhs.gov/cb/resource/trends-in-foster-care-and-adoption https://www.hhs.gov/about/budget/budget-inhttps://www.hhs.gov/about/budget/budget-in theoretical background the role of public management networks in government performance effective public managers as principled agents state-to-state effectiveness studies data collection and methods: the case of foster care administration case selection data previous findings new data analysis findings: external networks by public managers delaware michigan rhode island new york conclusion acknowledgements transboundary water governance in the kabul river basin: implementing environmental and public diplomacy between pakistan and afghanistan complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 101 university of bamberg press transboundary water governance in the kabul river basin: implementing environmental and public diplomacy between pakistan and afghanistan authors: bindu panikkar, asim zia, susan sgorbati, michael cohen, muhammand abid, faiz, muhammad fayyaz, nadia hashimi, shaukat ali, monir ahmad, zuhra aman, suleiman halasah, diana rice, gemma del rossi, benjamin ryan, kashif hameed, mujahid hussain, and naeem salimee rubinstein school of environment and natural resources, university of vermont, u.s.a. e-mail: bindu.panikkar@uvm.edu this research highlights the outcomes of the environmental diplomacy workshop held between members of civil society from afghanistan and pakistan on water cooperation in the kabul river basin, one of the most heavily conflicted transboundary river basins in the world. lack of trust among these upstream and downstream riparian partners, hydro hegemonic frameworks of governance, and persistent failures of track 1 diplomacy initiatives has led to an absence of governance mechanisms for mitigating the water security concerns in the region. the immediate problems identified to transboundary water rights, allocation, and security included a lack of focus on benefits to the local and rural population in the region, lack of access to water in rural areas, degrading water quality from multiple activities, and lack of scientific knowledge on the transboundary river systems to build better transboundary water relations, cooperation, and treaties. we argue that a collaborative civic-scientific social learning and engagement network can be an effective framework for building transboundary adaptive governance systems. such an effort on social learning is focused on building a bottom up process by strengthening local capacity, local knowledge systems, and transboundary engagement and exchange. initializing localized transboundary civic-scientific social learning capacities and exchanges can be a powerful avenue to building a localized adaptive governance framework. keywords: transboundary water governance; water cooperation; environmental diplomacy; kabul river basin; social learning introduction water is never managed for a single purpose. it supports multiple needs and interests— ecological, domestic, agricultural, hydroelectric, industrial, and recreational, which can often turn contentious under conditions of scarcity of potable water (wolf, kramer, carius, & dabelko, 2005). transboundary rivers make the issue of addressing multiple needs trickier. there are additional borders, actors, and networks to consider. inadequate representation of multi-level actors and institutions in such cases can worsen transboundary disputes and create deadlocks to effective management of water resources. at the heart of the transboundary water politics as the united nations convention on the law of the non-navigational uses of international watercourses (unwc) puts it is the question of water cooperation and equity on matters of allocation and utilization. established in 1997, unwc prescribes “equitable and reasonable” use, considering the relevant factors: geography, hydrology, socio-economic needs, dependent populations, impacts of use, conservation, efficiency of use, and availability of other sources (un general assembly, 2014). therefore, upstream and downstream states have to consider each other’s uses, both existing and potential reserves, when developing projects. while unwc has set the course of expectations in water sharing and cooperation, the http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 102 university of bamberg press implementation of these goals can be messy due to power differences between different user groups. riparian relations and transboundary water cooperation present complex dynamics, which are never static or easily classifiable as black and white situations; hence, cooperation and conflicts are deeply entangled as a process. this process may pose serious water allocation and utilization challenges and a linked social order of institutions, people, and processes may be required to manage it (mirumachi, 2015). yoffe, wolf, and giordano (2003) define trans-boundary water conflicts as: discordant actions between or among nations over shared water resources, expressed through verbal, economic, or military hostility, and with “violent conflict” as the subset of militarily hostile actions over shared water resources. water conflicts are also implicated in graver threats, such as the syrian civil war, which was fueled in part by a long drought that forced the migration of over a million rural syrians into urban areas (petersen-perlman, watson, & wolf, 2018). however, not all water-related conflicts are so visible. zeitoun and warner (2006) show that some conflicts lie somewhere between the much feared but nonexistent water wars and water cooperation. water cooperation efforts in the kabul river basin focusing on transboundary water conflicts and governance using traditional neo-realist/neoinstitutionalist paradigms is an insufficient approach to understanding the complex dynamics of transboundary water cooperation. this research offers a unique approach in understanding transboundary water cooperation by examining the participatory social learning processes of the localized non-state actors –community and scientific stakeholders – and their efforts at adaptive management to enhance trust building, information sharing, collaboration, and capacity building across the kabul river basin between afghanistan and pakistan (lee, 1999). the kabul river basin is an international river basin and a critical source for water, food and energy for both pakistan and afghanistan. the kabul river basin is one of the most conflict ridden areas in the world. as pakistan faces a looming water crisis and as afghanistan plans to increase its use of water resources to meet its increased demand for energy and agriculture, the tensions between these two countries are pronounced. currently, there are no bilateral treaties or agreements in place to guide negotiations over the kabul river basin water supply. the lack of legitimate scientific data on the tributaries bordering the two countries further complicates the tensions across the borders. in the absence of these mechanisms, the relations between afghanistan and pakistan on water security is heavily based on assumptions and the transboundary water politics is characterized by mistrust and suspicion, impeding any progress on negotiations for a successful transboundary cooperation between these countries (kakakhel, 2017). the need for engaging localized sub-state or non-state partners including local civic, scientific, community, environmental, and political actors to enhance information sharing, capacity building, and collaboration on interventions across the immediate borders has been recognized (saeed et al., 2016). the research presented is based on a week-long collaborative work-meeting among scientific and community leaders from both afghanistan and pakistan to explore the role and scope of scientific and public diplomacy (including community outreach, organizing, and engagement) in transboundary water cooperation, adaptive governance, and ecological preservation. we situate this study within the theoretical framework of hydro-hegemony and adaptive governance systems to develop a localized adaptive transboundary governance system based on social learning. the working sessions explored 1) the challenges to transboundary cooperation between afghanistan http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 103 university of bamberg press and pakistan; 2) ways to increase transparency by addressing the knowledge gaps and uncertainties related to the kabul river basin; and 3) the ways scientific diplomacy and public diplomacy can aid water cooperation in the transboundary kabul river basin. theoretical underpinnings: building adaptive governance systems for water cooperation as conditions of water scarcity, environmental degradation, and resource scarcity become pronounced due to developmental needs and climatic shifts, the risks for conflict and political tensions between states that may already have fragile relationships are intensifying (haddadin, 2001; jansky, pachova, & murakami, 2004; kliot, 1993; lowi, 1993; sosland, 2007). conventional governance models are inadequate in addressing these evolving transboundary tensions. instead proactive adaptive governance strategies are required to minimize risks and increase cooperation (haddadin, 2001; jansky, pachova, & murakami, 2004; kliot, 1993; lowi, 1993; sosland, 2007). two conventional models in governance in transboundary conflict resolution are the neorealist and neo-institutionalist (or neoliberal) paradigms (mirumachi, 2015). power is seen as the main factor shaping the outcome of discussions on bilateral issues in a neo-realist paradigm. neorealists expect states to maximize their relative power rather than choose cooperation, especially if the upstream states attempt to reduce their dependence (e.g. trade or power) on downstream countries. neo-realists may also challenge their neighbors’ sovereignty, territorial integrity and security. neo-institutionalists and neo-liberalists, in comparison, reject the notion that power is the only factor that determines relationships between states. neo-institutionalist focus on cooperation by mitigating the compliance problems and mistrust—with the help of institutions that provide information, lowered transaction costs, increased transparency, and reduced uncertainty. such cooperation is accomplished often by utilitarian compliance or by incentivizing cooperation and normative compliance by the use of norms, regimes, treaties, policies, and political will and international law, often to the advantage of the hegemonic state (zeitoun and warner, 2006; julien, 2012; slaughter, 2011). neo-institutionalists also believe that scarcity and degradation may also justify technological innovations and infrastructural fixes to manage the crisis, with also a certain control of the society (mirumachi, 2015; zeitoun and warner, 2006). according to the neoliberal paradigm, water management holds key to development, economic progress and social benefits. both these models still do not address inequities, and democratic and adaptive ways of addressing transboundary tensions and cooperation. zeitoun and warner (2006) identifies this paradigm as the hydro-hegemony framework developed by the use of state-centric hegemonic approaches to guarantee cooperation which may include exertion of supremacy in water management to determine “who gets how much water, how and why” (p.436), where resources are not equally endowed. the identification of these hydro-hegemony frameworks have helped differentiate asymmetries in power relations that make it difficult to change the status quo of water allocations and cooperative mechanisms by the implementation of political persuasion and consent within civil society (fontana, 1993). these studies show that the existing socio-political processes that attempt to determine the solutions and means to address water scarcity and environmental stewardship itself may be the process that brings about conflict and cooperation (sneddon, 2013; trottier, 2003; zeitoun and warner, 2006). water peace and cooperation therefore can be a manipulated concept through which politics occur, and can mediate state/self-interests, allowing it to become a hegemonic concept (mirumachi, 2015). folke et al. (2005) note that adaptive self organizations or governing systems are more creative at adaptive co-management efforts and it ” lowers the costs of collaboration, conflict resolution, enable legislation, and governmental policies” (p. 41). http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 104 university of bamberg press pahl-wostl (2007) defines adaptive management as “a systematic process for continually improving management policies and practices by learning from the outcomes of implemented management strategies” (p. 51). adaptive governance systems are multilevel, integrated governance modes that include social networks that represents multiple interests, actors, and community-based initiates. such systems may utilize integrative science that draws from various knowledge systems, experiences, and organizations through an open decision making process for the development of a common understanding and policies, in particular during periods of abrupt change or crisis in social-ecological systems (folke et al., 2005). adaptive systems hence highlight broader shifts from government to governance and the evolving, complex and dynamic nature of governance networks (koliba, meek, & zia, 2010; zia et al., 2014). such systems also emphasize social learning as an integral component of governance. strong institutions and social learning are essential for managing water scarcity and allocation issues (pahl-wostl, 2009). social learning in river basin management is catered to building the capacity of different authorities, experts, interest groups, and the public to manage their river basins in a sustainable way while balancing multiple and competing interests for the benefit of the social-ecological system as a whole (pahl-wostl, 2007; 2009). this model of social learning was utilized by the european project harmonicop to increase participatory river basin management in europe in order to generate practical and useful information to improve the scientific base of social learning in river basin management. within adaptive systems, it is not the lack of water that leads to conflict, but the inadequate way the resources are governed and managed that leads to conflict. carius et al. (2004) state that “the likelihood of conflict rises as the rate of change within the basin exceeds the institutional capacity to absorb that change” (p. 61). there is a growing body of evidence that both afghanistan and pakistan have struggled with hydro hegemonic frameworks of cooperation with transboundary water management (nagheeby & warner, 2018). oftentimes, access to water has been used as a geopolitical tool in the region to control territory for security and geopolitical interests. thus, water resource development has had profound implications in shaping the geopolitical dynamics in afghanistan (nagheeby & warner, 2018), and pakistan to a lesser extent. additionally, regional hydo-hegomonic tensions have flared between pakistan, afghanistan, and india with india supporting afghanistan’s efforts to develop dams along the kabul river and china supporting huge infrastructural projects in pakistan (kugelman et al., 2011; shroder, 2016). understanding the existing complex power dynamics and paradigms are helpful in redesigning cooperative civic-social strategies to shape adaptive transboundary water governance. in this research we examine the existing hegemonic frameworks that impede transboundary cooperation and identify avenues for building localized civic-scientific social learning efforts to enhance local adaptive transboundary governance systems. we argue that collaborative civic-scientific social learning processes can be an effective framework for building transboundary adaptive governance systems. such an effort on social learning is focused on building a bottom up process, by strengthening local capacity to understand and estimate the local challenges to transboundary water management and to build localized and contextual solutions for the adaptive management of transboundary water resources across borders. initiating localized transboundary social learning capacities can be a powerful avenue to lay the groundwork for peaceful cooperation and building a localized adaptive governance framework. methodology utilizing the concept of collaborative civic-scientific social learning, researchers from the university of vermont and bennington college hosted a week-long workshop in march 2018 for 11 participants (six from afghanistan and five from pakistan) to initiate dialogue and in-depth discussions on water cooperation between afghanistan and pakistan on the kabul river basin. over http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 105 university of bamberg press 20 participants were invited to attend the workshop at the peace centre at dawson college in montreal, a neutral place for people from both countries to gather, but only 11 were able to obtain visas to attend the workshop. holding such a workshop on building localized water cooperation strategies in either afghanistan or pakistan would not have been feasible. the participants included community leaders and professionals from both countries as well as representatives of international funding agencies working in the region on water issues. the network partners have been inspired and informed by successful environmental cooperation projects in palestine, israel, and jordan, as well as in peru, ecuador, and cyprus. the week-long work group was facilitated by trained peace and conflict resolution studies experts. the week-long workshop in a neutral setting among partners from afghanistan and pakistan, countries which have stressed relationship with each other, helped to maintain an open minded and impartial stance towards conversations on building transboundary water cooperation strategies. the in-depth discussions between civic and scientific community from afghanistan and pakistan included 1) the political, environmental, social, technical, scientific, and regulatory challenges to water cooperation; 2) sub-state needs and expectations for different uses of water in the region such as urban development, food and irrigation, hydroelectric development, and mining; and 3) hopes for the future of the region and strategies to improve border relations, trust, sharing of benefits, and building water cooperation between afghanistan and pakistan. the workshop was an attempt at creating a localized and grassroots-based or civic and scientific-based adaptive framework for transboundary water cooperation in the kabul basin. this vision aligns with all international conventions and treaties on equitable sharing of resources and the sustainable development goals of clean, accessible water, affordable and sustainable energy for all. the week-long discussions and mediations that transpired between the civic and scientific community was recorded, transcribed, and coded. we conducted a discursive analysis of the coded the interviews. our primary codes of analysis included 1) the hydro-hegemonic governance challenges to water cooperation, 2) knowledge gaps that generate animosity between transboundary partners to inform building social learning strategies, and 3) identified opportunities for localized social learning, and situated knowledge construction that could influence local transboundary water cooperation. we used zeitoun and warner’s (2006) hydro-hegemony framework which included examination of 1) coercive compliance producing mechanism which include military force, covert actions and threats; 2) hegemonic compliance producing mechanism which includes securitization of resources, coercive water sharing conditionalities, sanctioned discourses, and construction of popular knowledge and the mobilization of both international support and financial clout and the abuse of riparian position; 3) utilitarian compliance mechanism such as incentivizing cooperation; and 4) normative compliance mechanism such as use of norms, treaties, international law often to the advantage. we incorporated social learning concepts from folke et al. (2005) and pahl-wostl (2009), and identified social learning as opportunities for local capacity building, localized knowledge construction in partnership with the community-scientific collaboratives, and initiating transboundary exchange of social learning information gathered to further adaptive governance. in addition to the narrative analysis of discussions among the civic scientific partners from afghanistan and pakistan, we have substantiated their reports with valid sources from key journal articles, media reports, and policy briefs. below we provide a brief overview of the kabul river basin followed by the results of the workshop. http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 106 university of bamberg press brief overview of the study site—kabul river the kabul river lies between afghanistan and pakistan, creating a unique transboundary position and tension over water sharing, as both countries are upstream and downstream riparian partners. the kabul river originates from the sanglakh range in the northwestern hindu kush mountains of afghanistan about 72 km west of kabul. the kabul river and its tributaries – loghar, panjshir, ghorband, laghman-alingar, and kunar rivers – irrigate 72,000 km² or 11 percent of afghanistan’s land area (iucn, 2010; thomas, azizi, & behzad, 2016). its largest tributary is the kunar river located in the northeast part of afghanistan, and is also a transboundary river bordering pakistan. the kunar river originates from glacial fields in the hindu kush mountains in the chitral region in pakistan. as the kunar river flows down from the hindu kush mountains, the river is known by several names – mastuj river, latkoh river, and the chitral river in pakistan – before finally flowing into the kunar province in afghanistan with an annual flow of 10km3 (iucn, 2010). the kunar river runs downstream about 180km before joining the kabul river near jalalabad. the combined rivers then flows 80 km eastward along the khyber pass back into pakistan with renewable water potential of 55 km3/year, about half of which is used for agriculture in afghanistan (iucn, 2010). the kabul river is the only river system in afghanistan with an outlet to the sea as it joins the indus river in the city of attock in pakistan (iucn, 2010). the kabul river thus is also part of the upper indus basin. though pakistan’s dependency on kabul river is limited, tensions over water, especially in the transboundary kunar river, where both countries are upstream and downstream riparian partners is more pronounced as afghanistan looks to rebuild their country from being in war since the soviet invasion in the 1980s. results below we detail the hydro-hegemonic frameworks that challenge water cooperation, knowledge gaps that generate animosity between transboundary partners, and identify opportunities for building localized civic-scientific social learning, and situated knowledge construction practices to enhance transboundary adaptive governance systems and water cooperation. figure 1. kabul river basin and links to the indus river. source: thomas et al., 2016. http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 107 university of bamberg press hydro-hegemonic challenges to water cooperation in the kabul river basin hydro-hegemons expect states to maximize their relative power between their riparian partners. zeitoun and warner (2006) define hegemonic power systems as coercive compliance producing mechanisms, hegemonic compliance producing mechanism, utilitarian compliance producing mechanism, and normative compliance producing mechanism. we also find that these variations in hegemonic power structures may be influenced by historical aspects, geographical constraints, concerns about future water scarcity due to climate change in the kabul river basin contributing to the implementation of different hegemonic compliance mechanisms. historical aspects that fuel coercive and hegemonic compliance regimes: the historical relationship between pakistan and afghanistan is complicated. political animosity between afghanistan and pakistan date back five centuries since the founding of pakistan in 1947 and afghanistan’s continued refusal to accept the durand line as the formal border between afghanistan and pakistan as the ethnic, linguistic, and religious faiths of the communities settled on either side of the durand line are the same (micallef, 2015a; 2015b). as one participant remarked, “it was like the british drew the border line through the heart of the pashtun communities.” historically, the participant noted, “there were no borders between the largely pashtun and kutch communities in afghanistan and pakistan. kuchi tribes used to go to lowland pakistan in the winter and moved to the higher mountains in afghanistan in the summer time. they did not belong to any country, they had no citizenship.” though the coercive and enforced boundaries drawn by the hegemonic power divided tribes and communities, the perceived desire for pashtuns on both sides of the border to form “pashtunistan” was seen a worrisome movement to pakistan as well as afghanistan’s support of pashtun separatists in pakistan (micallef, 2015a; 2015b). in turn, pakistan military supported taliban terrorist camps in afghanistan (micallef, 2015b; saeed, hassani, & malyar, 2016; thomas et al., 2016). workshop participants from afghanistan felt that: “pakistan has been waging war against afghanistan for the last 40 years. you cannot deny that. but i try to be diplomatic and not call it a war.” the adjacent areas along the border between afghanistan and pakistan, and the kunar river have been the most vulnerable to violent conflict, terrorism, and explosive land mines since the us-nato invasion of afghanistan in 2001. between the soviet invasion (1980s), taliban uprising (1990s), the drought of 2000, and current conflicts, one in four afghans have been a refugee, mostly in pakistan and iran (ruiz, 2002). the participants noted that these long standing tensions over land disputes and borderlines have been quietly spilling over into disputes over water and kabul river, and the hegemonic control of water resources especially by iran and pakistan, and recently india (mashal, 2012; thomas et al., 2016, kugelman et al., 2011). by some estimates, one-third of afghanistan’s irrigation systems were directly affected by war (iucn, 2010). participants noted that as afghan refugees return home, there is a growing demand for water in the agriculture and energy sectors. diminished supplies of water flowing down from kabul river to pakistan has been a source of strain and conflict in severely water stressed pakistan, leading to subtle coercive acts towards afghanistan. lack of water treaties between afghanistan and pakistan further aggravate issues of abusive threats and riparian control. the only water-sharing treaty that exists between these countries is the 1921 treaty signed between afghanistan and great britain, which agrees to water withdrawal for residents in torkham, afghanistan, and for using the kabul river for navigation and irrigation rights in pakistan (hearns, 2015). between 2003 and 2011, pakistan and the world bank incentivized cooperation by initiating discussions on a water treaty similar to the indus basin treaty in order to build utilitarian and http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 108 university of bamberg press normative compliance and institutional capacity to promote greater cooperation to avoid dispute over water issues. these initiatives have not been fruitful. acute power asymmetries curtail dialogue and promote the view of unilateral resource capture and control within transboundary water resources development (thomas et al., 2016), leading to absence of mechanisms for relieving water security concerns within the kabul basin. geographic capacity and covert coercive actions: the kabul river basin is highly under-developed and its water resources are under-utilized. the participants discussion on water use in both countries concurred with the available metrics on water use. afghanistan utilizes less than 25% of its surface water resources. its groundwater use is less than a third of its total availability and shares 90% of its water resources with its neighbors (mudabber, 2016). the storage capacity of water in afghanistan is very poor and represents less than three percent of the average surface water availability (thomas et al., 2016). conflict and conflict-related deforestation further exhausts afghanistan’s capability to store water (price et al., 2014). due to geography, remoteness of the region, and conflict such as the persistence of landmines, many areas are unsafe to access, which was one of the major concerns noted by the participants. in contrast, the overall level of water use in pakistan is very high, compared to available resources. pakistan uses 70 percent of the available surface water and extracts more groundwater than the average annual recharge (thomas et al., 2016). pakistan also has lower water storage capacity (10%) and can barely store 30 days of water in the indus basin, constraining irrigation development (thomas et al., 2016). the total renewable water available in pakistan (1378-1000 m3) is far lower than the world average (kakakhel, 2017). pakistan gets 13 percent of its water supply to the indus basin from afghanistan’s portions of the kabul and kunar rivers (thomas et al., 2016). however, participants from pakistan claimed that the kabul river contributes 16 to 17 percent of pakistan’s water supply, as stated in some unsubstantiated media sources (kiani, 2013). any increase in water use in afghanistan (a projected 17 billion cubic meters (bcm) to 30 bcm) threaten the dimensions of water use in pakistan and its corresponding water security (mudabber, 2016). afghanistan also suffers from severe shortage of electricity, where only 28% of afghan households have electricity (kakakhel, 2017). water scarcity is already acutely felt in pakistan with increase in droughts and low storage capacity, according to the participants, an observation that has been noted by thomas et al. (2016) and saeed et al. (2016). declining flows in the kabul river has been previously observed in attock, pakistan (iucn, 2010), potentially due to climatic variability and upstream uses in both afghanistan and in the peshawar region in pakistan. the government of afghanistan’s attempt at water infrastructure plans for irrigation and power generation in the kunar river (iucn, 2010; thomas et al., 2016) have been cautiously monitored by pakistan, who have also been accused of savage executions (mashal, 2012) to deter the plans of afghanistan to build hydroelectric dams on kunar river. these speculations over water, created momentary deadlocks within our afghan-pakistani participants as well: if afghanistan implements all the dams on the kabul river, it would only have a 3% of effect on pakistan. building of dams is perfectly legitimate for our utilization. in asmar district (suritoq location) in kunar province, afghanistan has been trying to build a dam, it has been sabotaged a number of times. this is the reason the people in the kunar province are in the dark. because we do not have electricity from kunar river. dams in kunar is an important issue. we have terrorism, water issue, poverty, refugees, climate is another issue of conflict. much of this can be addressed by putting those dams on. i do not think we can come up with a treaty. we cannot have the community die while we come to a consensus. http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 109 university of bamberg press however, participants from pakistan highlighted the country’s severe water shortage, sharing that any construction of dams without consultation a threat with the possibility of 17-20% reduction in available water flowing to pakistan from such development. a study funded by the world bank under the award project found that the overall reduction in water flow towards pakistan would only be three percent, especially between april to june, based on the construction of six dams (world bank, 2013). however, the pakistani government-funded islamabad institute of policy studies claims a 15-20 percent reduction in the available water from the foreseeable construction of 13 dams on kabul river (rizvi, 2003). afghanistan believes that pakistan has been sabotaging afghan initiatives to increase hydropower development on the kunar river and considers the presence of pakistani militia on the border as a serious threat to any development in the region (mashal, 2012). the primary issue that pakistani participants objected to is afghanistan building its water storage capacity: “if we can show with data that this is not going to impact pakistan, given the climate variability, and access, then there can be a consensus.” hydroelectric dams are often large, internationally funded projects, and are often the result of state-building. if the source is a transboundary river, the potential for contentions between the two countries are far greater. while hydro hegemonic projects are under state jurisdiction, they are still dependent on the mobilization of international support and financial clout and donor conditionalities which may also enforce a set of coercive compliance producing mechanisms. neoliberal governance approaches and challenges to local water security: both afghanistan and pakistan have a lot to gain from compromises and implementing cooperative strategies. neoliberal paradigms are cooperative and focused on reducing conflicts by mitigating the compliance problems and mistrust, while seeing water management as the key to development, economic progress, and social benefits. big infrastructural projects for development are governance projects of great state importance and projects of nation-building and modernization. water is a key enabler for economic progress and social, and public good (mirumachi, 2015). in a growing country, there are many water demands, not only to serve the growing urban population, but also for irrigation, agriculture, energy production, and other industrial applications. this paired with similar needs of a neighboring state that shares river systems, the calculations of water use and sharing can get easily tricky. irrigation is the world’s largest use of water (70%), and this overall trend is reflected in both pakistan and afghanistan (un, 2014). in afghanistan over 85% of the irrigated area is river-fed and only 15% is groundwater-fed. in the absence of more water storage capacity, most of the country’s surface water runs to neighboring states, including the kabul river from afghanistan to pakistan. in pakistan water demand has been steadily increasing. from the 1970s to the 1990s, water demand increased by 39 percent, largely due to intensifying irrigation and the increase in cropped agricultural area. the water from the kabul river is indispensable to major cities such as peshawar city, where over two million residents depend on the river for drinking water, sanitation, and irrigation (iucn, 2010; thomas et al., 2016). hence any development in afghanistan raises grave concerns across the border. deforestation was highlighted as another huge challenge in both countries, and the lack of watershed management contributes to erosion downstream where the kunar and kabul rivers intersect, as well as after the kabul river crosses the afghan-pakistan border and joins the indus river. considerable potential exists for implementing soil conservation measures. participants also http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 110 university of bamberg press shared the observation that many water-related problems were connected to a larger set of environmental and social problems such as development and climate change. baig & al-subaiee (2009) note that factors like deforestation, overgrazing, soil erosion, salinity, and waterlogging are posing major threats to biodiversity in the region, and any attempts to reverse losses are ineffective without local participation. mining was also considered as an issue of concern among the participants to water security and cooperation. both pakistani and afghan governments have been touting mining development; they have invited the mining industry to come in. in many places, chinese companies are one example starting to develop mining operations as part of the new “silk road” project (awan, 2018). one participant describes the threats that mining poses to water security and water quality: “so they want to put in a lot of these mining technologies to mine metals like iron and copper and other kinds of ores, but the mining can have a huge effect on the water pollution, it can also destroy the habitat. mining it is an emergent threat.” new research articles show that there have been conflicts between insurgents over mineral resources. taliban and islamic state khorasan province (iskp) militants are making millions from exporting minerals like talc, marble, and lapis lazuli in afghanistan, generating revenue for insurgent groups and fueling a deadly cycle of violence in the country (zia & hameed, 2014; hussain, 2018). such illegal mining costs the afghan government massive losses in revenue from precious metal deposits, which u.s. officials’ estimate contains up to $1 trillion in value (2010 usd) (hussain, 2018). additionally, pakistan is the single biggest source of talc sold in the america. the effluents from these industries, marble and stone work, and pharmaceutical companies on pakistan’s side, said the participants, are heavily polluting the river. a major concern among the participants, and an impact of the neoliberalizing efforts previous identified, are surrounding water quality. the kabul river’s upper upstream rivers and tributaries are fast moving and filled with sediment and pollutants. in some provinces, especially sindh and punjab, the participants stated that the water quality is so bad the government has banned some companies from bottling the water (asian news international (ani), 2018). one of the participants remarked that “the trout is almost extinct – can you imagine the quality of water?” water pollution from human and agricultural waste and other pollutants can spread pathogens and diseases, which can also be transported from one country to another. waterborne diseases are high in both afghanistan and pakistan (gulab, 2016; gupta, 2017). neoliberalization of water and nature assign economic than social value where natural resources are seen as instrumental to economic growth, progress, and techno-scientific development, which are essential projects of nation building, statehood, and modernization. hence such activities of building hydroelectric dams, mining, and agriculture are also often subsidized activities for the cooperation and social benefit it provides. such appropriation of resources to support causes of social progress and economic profit do impact upstream and downstream communities, but there are no effective ways to address these differences and cooperation across the borders are strained due to lack of coordination and treaties among national actors. rural access to water and challenges to water security: among conversations on transboundary water conflicts and cooperation, the participants felt that there was little focus on rural issues of concern related to water and assessing realities on the ground regarding local access to water. afghanistan is among the poorest underdeveloped countries in the world (iucn, 2010). people in the border areas in both countries are largely illiterate, have not attended school, and have low levels of education. the participants commented that women are especially affected by water security. women often have to walk long distances (up to 5 kilometers) http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 111 university of bamberg press to get water from the tributaries, carrying buckets of water, and at times with an infant on their back. and the water fetched is often untreated, with possible contaminants in it. the re-use of water is often not possible because these regions do not have the appropriate facilities for water treatment. the participants felt that “if women are educated and integrated into the economic streamline of the country, it is the cheapest investment in any foreign policy” regarding water resource management. participants noted that access to water is especially poor in rural afghanistan. access to water for irrigation and agriculture in rural areas still remain a major problem as much of the infrastructure in the border areas in afghanistan were destroyed and need to be rebuilt. over 80% of the population do not have access to safe water in the rural areas (gulab, 2016; gupta, 2017). communities directly on the river benefit, while those living far have to dig channels to gain access to water in some rural areas where possible. however, these rural social practices have been practiced for centuries. the participants noted that the water for subsistence irrigation is drawn from the small tributaries more so than the main branch of the kabul river in rural pakistan and afghanistan. the right of water is also attached to right of land. land tenure rights blocks access, as sections of land are often owned by few feudal lords. so the users have to get the consent of other tributaries’ users to dig such channels. localized norms and practices thus still hold strong sway in building water security and cooperation. the participants recognized that the strains of water security were felt by all, even though there were differences in the emphasis of certain problems (e.g. scarcity vs. access) within and among the two countries. participants recognized that addressing these issues require that people and the states work together to reallocate power by focusing on the needs of those who live along the rural border areas who are underrepresented within these transboundary water negotiations: “the way to get people involved is through water. if we talk about water, we are reframing the political tensions and collaborating around water.” this lays in contrast to the current reality in which conversations focus on nation building and economic progress, rather than catering to the needs of those living along the river corridors that are most disadvantaged and underprivileged. knowledge gaps to water cooperation there is a poor understanding of the available water resources in the kabul river basin to support good governance. the knowledge deficits in the water sector in the kabul basin have also been reflected in the report by iucn (2010), which notes that the most reliable data for irrigated areas in afghanistan dates back to 1967. the information deficit is greater in afghanistan, but region-specific information in northeast pakistan in the chitral region is also limited (iucn, 2010; saleem & zia, 2016). the existing lack of reliable data also promotes political apathy, weak governance, lack of democratic institutions, red tapism, corruption, and a trust deficit. the participants strongly felt that data is essential for any discussion and negotiation to address water security concerns and conflicts, and for community mobilization. one participant explained the fundamental importance of good data: if the data is not available, how can we say that afghanistan and pakistan need this much water annually? why do afghanistan/pakistan claim these resources which do not belong to them? we don't know where the glaciers are, how fast they are melting, or the flow in different months of the year, the increase in temperature, change in relative humidity. the changes in the frequency and intensity of extreme events. would glacier lake outbursts be an extreme event? http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 112 university of bamberg press another participant commented that there is no accurate count of the demographic variables in this region as well. lack of data on water access in the region due to terrorism and land mines prevents portraying precise estimation of the burdens that the women and families in this region bear. the participants identified a need for proper research, which includes monitoring systems and a complete needs assessment on both sides of the border. gathering data in the region is not a straightforward process due to mutual mistrust. the participants reiterated the importance of doing these studies in mutual agreement and collaboration–water use in both pakistan and afghanistan need to be cooperatively monitored and regulated to address water allocation issues and to generate trust and avoid conflicts. another participated reiterated that state consent might be necessary so the data is not questioned. this, the participants suggested, might provide access to certain inaccessible areas due to conflict and government prohibition. conducting a thorough study in the region will eventually require track 1 diplomacy involvement to gain access to appropriate regions, especially vulnerable regions. government approval of the data is integral as it has authority over river discharge and management. independent research, the participants noted, conducted by international consultants are seen as more credible. other concerns they had were where the data should be gathered in the transboundary river. participants suggested utilizing models such as community-based monitoring and citizen science may also help in gaining trust with data collection as well as mobilizing communities through local control of evidence based research data. the participants found that a lack of a treaty on the kabul river is an opportunity to support a future treaty that results from both public consultation and scientific assessments, taking into consideration the local developmental needs in the region as well as the impacts of a changing climate. building localized civic-scientific social learning and engagement networks for water cooperation and adaptive governance while state relations are fraught with complexity, the participants felt that the current water security issues pakistan and afghanistan face cannot be solved without transboundary cooperation—without the acknowledgement that pakistan as well as afghanistan has rights to its share of the water from the kabul river and the right to develop its resources. the immediate problems identified from the workshop on transboundary water rights, allocation, and security were a lack of focus on benefits to the local and rural population in the region, lack of access to water in rural areas (80% in rural areas have no access to safe water), degrading water quality from multiple activities, and lack of scientific knowledge on the transboundary river systems to build better transboundary water relations, cooperation, and treaties. the participants agreed that the first step to building border relations is to bridge the gap between people and communities across borders. solutions discussed to improve transboundary water rights and cooperation include improving 1) community building and engagement, 2) local capacity and civic-scientific social learning and local knowledge construction, and 3) transboundary engagement, exchange and advocacy with a specific focus on improving bilateral communication and trust across borders on managing water resources. community building and engagement for trust building: trust deficit is one of the biggest problems between afghanistan and pakistan, and without addressing and building trust, the participants felt it would be difficult to build transboundary water cooperation. while the political animosity runs deep, the pashtuns share historical, religious, and http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 113 university of bamberg press cultural ethnolinguistic links that runs back thousands of years and the countries still share strong trade relations. the workshop participants agreed: “there is no conflict at the tribal level. they are brothers. the conflict is at the state level. we have to mobilize the civil society on both sides to compel our governments to sit and negotiate.” the working group agreed that starting with community building could be better in increasing trust: “there is a perception of conflict and insecurity… when you bring the community together, it is to create empathy, understanding, and trust. how can we find a way to influence and increase awareness?” participants felt that the best way to engage and build the community in the region is by working with the exiting community groups on the ground in afghanistan and pakistan. already in participation at the workshop were community motivation and development and organization (cmdo), a network of 1000 communities in pakistan, and coordination of afghan relief (coar), a network that works with 33 communities in afghanistan on water and sanitation issues. one of the participants remarked on the power of community groups: the community themselves can take charge. in my opinion when we are talking about public diplomacy, it has to be coming from ordinary people living on both sides of the border. it should reflect their concerns, needs, demands and their hopes for the future of the society and transboundary relationships. the suggestions by the community leaders included doing a public awareness campaign to build river-watching communities to clean, improve access, demine, demilitarize, and to increase cooperation across borders on the kabul river. river basin organizations are commonly used management structures internationally to reduce conflicts (molle & hoanh, 2011; mukhtarov & gerlak, 2013; petersen-perlman et al., 2017; yoffe et al., 2003). the structure for building regional cooperative river basin organizations already exists through the medium of coar in afghanistan and cmdo in pakistan, rather than through national diplomatic means which alone may not foster cooperation and trust-building locally on the ground. other avenues for community building discussed include targeting women, tribal leaders, religious leaders, setting up traditional community councils called jirgas, and engaging youth activists, all of whom can also be part of the river based organization in creating multi-level governance systems. governance in these regions is primarily mediated through traditional councils. the constitution of jirgas are prominent in the tribal areas on both sides of the border. matters related to conflicts and community decisions are often settled through convening jirgas (e.g. see zia & hameed, 2014). “it is a well-established and lubricated machine, a mechanism that works,” said one of the participants. another participant suggested holding trans-border jirgas to talk about water issues. the authority of a jirga is so strong that no one questions them; furthermore, they do not need funding, and they do not have many of the pitfalls of bureaucracy (e.g., corruption, power differentials). it is not a bureaucratic but democratic system, as anyone can call a jirga. people also respect the elder of the village and the information they give. dissemination of information in mosques and masjid through religious points of view can also be an effective way to convey messages by connecting water practices to religious beliefs. youth activism is growing in these countries as well and was observed as a powerful tool to educate people and families: “once you start mobilizing the young people, children can be real ambassadors of change.” the awareness campaign can also be done through mediums including, but not limited to, radio messages, television, mobile messages, and door-to-door awareness program. once communities are mobilized, specific projects and issues of local concern can be discussed and debated. however it was noted that, equitable, open, and flexible cooperation on water allocation might be difficult by focusing only on stakeholder interests without the necessary data to back the interests for decision-making. http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 114 university of bamberg press local capacity and civic-scientific social learning and local knowledge construction: the need for data was obvious among participants to settle any misconceptions, distrust, differences, and conflict across borders as well as to promote peace: “data can bring accountability and transparency.” it was realized that both sides need multiple levels of data to make informed decisions and to negotiate collaborative water management plans (also see ali & zia, 2016). the participants pointed out that there has not been systematic data collection on the region; even the existing demographic information in the region is not reliable as it is politically volatile. additionally, the existing technical data is not collected in one place. the science diplomacy goals that the participants agreed to were 1) build local capacity by setting up local social learning and knowledge construction initiatives that would also include citizen science or community-based monitoring stations that can be administered by farmers, fishermen, students, and community members with scientific partners; 2) synthesize existing baseline data, including satellite data on both countries as well as other relevant social, environmental, technical, health, and policy data into a central repository; 3) produce localized knowledge that can inform projects on the ground, conduct studies to understand the risks and benefits to both countries, as well as upstream and downstream communities under specific scenarios of development of hydroelectric dams, impacts of climate change, deforestation, urbanization, and mining or other local scenarios of concern; produce database that could help with early warning systems (ews)—as both countries are facing extreme threats of flash flooding, landslides, drought. presently there is no mechanism of sharing data between pakistan and afghanistan, however, both countries realize that there should be a mechanism of sharing data that can save lives, infrastructure, and farmlands at both sides; 4) establish a knowledge sharing network, promote data exchange, transparency, and sharing to mitigate risks to reduce adverse outcomes and promote transboundary water sharing and basin management practices. ways to address uncertainty, unexpected events, and emergencies emerged as a key challenges to address as well. this is where clearly defined obligations, procedures for cooperation and consensus are needed to make decisions under uncertainty and emergency situations (wwap, 2012). directing the focus on dispute prevention, resolution mechanisms, and adopting preventive measures and mutually beneficial negotiations might be essential in these cases to maintain peace and enhance cooperation. additionally project participants suggested conducting evaluations of power relations between the tribal areas and countries, as well as conducting stakeholder and actor analyses. the goal is to build effective water cooperative strategies even without a track 1 treaty. however, ministry level cooperation might be necessary to collect the data in the region, as the data must acceptable and legitimate to both the country officials to initiate peaceful negotiations. reliable data about the transboundary watershed is essential for stakeholders to buy into cooperative frameworks, and cooperative transboundary problem solving (petersen-perlman et al., 2017). as a community member remarked: then they are in a better position to talk about what is the demand and what is the use [of water resources]. then the goods and bads can be shared, the floods and droughts are shared based on the allocation formula. we can understand the baselines versus the variability. this will be negotiated by the critical parties as opposed to fighting with each other. the participants emphasized continual learning as one of the key features of adaptive management in building transboundary water cooperation and peace between afghanistan and pakistan. http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 115 university of bamberg press transboundary engagement, exchange and advocacy on managing water resources: social learning, knowledge construction, and advocacy go hand in hand. stakeholder engagement, public participation, and opportunities for local social learning at the local and river basin level is essential in jointly managing, protecting, and developing water effectively and sustainably. research demonstrates that where institutional capacity for dialogue and the management of disputes is present, conflict is less likely (jägerskog, 2012; subramanian et al., 2012; unep, 2008). stakeholder platforms, river basin organizations (rbos), and jirgas can be spaces where different stakeholders can articulate their concerns, manage differences, and come to settlements between countries, cities or sectors as well as play a crucial role in local water management. these steps of public diplomacy and science diplomacy are essential for pakistan and afghanistan to address transboundary water issues and development issues in the kabul river basin. a growing number of treaties, protocols, conventions, and institutional frameworks increasingly help crystallize mechanisms on the use, development and protection of shared freshwaters and peaceful resolution of disputes over water resources (bigas, 2012; gerlak, 2007; subramanian, brown, & wolf, 2012; un, 2013). around 40% of the world’s transboundary basins are covered by some form of cooperative institutional arrangement (gerlak, 2007). studies also show that international water relations among riparian states with treaties are generally more cooperative than basins without treaties or other cooperative management mechanisms (oregon state university department of geosciences, 2011). treaties and cooperative institutional agreements though often suffer setbacks from lack of political will, inadequate water management structures, and lack of enforcement at national levels (salman, 2007). formal networks with regional and international third party involvement are also essential for the resiliency and success of transboundary water cooperation, in order to lessen the animosity and improve negotiations between the two countries. third party involvement traditionally includes stakeholders who do not live in a shared basin but have interests in or influence on how shared waters are managed and allocated, such as development banks, experts, and environmental nonprofits. third party groups could also be resident and/or refugee groups, such as afghans residing abroad, who have the best interests in mind of their countries and have influence over its development (petersen-perlman, veilleux, & wolf, 2017). international multilateral engagements are increasingly employed in sustainable basin management. for example, the eu-funded switch (sustainable water management improves tomorrow’s cities’ health) research program brought stakeholders from a consortium of 33 partners from 15 countries, who jointly discussed problems and developed and tested ideas for sustainable integrated urban water management (butterworth, mcintyre, & da silva wells, 2011). the participants in the workshop felt that a global network was essential to address transboundary challenges, to be taken seriously by the ministry, and to later build the base for goal three – the advocacy and lobbying campaign. the global network would also build bridges with track 1 governmental diplomacy for broader transboundary cooperation and treaty building. this led to the conception and establishment of the transboundary water in-cooperation network (twin), a network of networks that connects transboundary water groups globally. twin’s mission is to strengthen and mobilize organizational networks that work locally, nationally, and globally towards equitable and reasonable sharing of water resources across borders for food, energy, livelihood, and ecosystem services. twin’s vision is to build peaceful resolutions for water cooperation among stakeholders and nations, to improve border relations, water use practices, water quality, ecosystem services, and public health on both sides of the border, and build http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 116 university of bamberg press resilience to climate change. twin abides by the un law of international watercourses and transboundary aquifers and the un sustainable development goal 6 framework which includes 1) the principle of equitable and reasonable utilization, 2) obligation to not cause significant harm, 3) obligation of cooperation in jointly managing, protecting and developing water resources, 4) mechanism of prior notification of and consultations on planned measures, 5) reduce and control pollution of shared aquifers and 6) peaceful settlement of disputes (un general assembly, 2014). in particular, twin focuses on facilitating negotiations between civic and scientific diplomacy by 1) enhancing community engagement and capacity building to promote water cooperation by helping to build localized social learning and knowledge construction practices to improve access to water and protect the quality of the water along transboundary rivers 2) identify the necessary information such as data collection, research, reports and mutually beneficial technical solutions to improve the quality of water resources, 3) convening meetings and negotiations to promote bilateral/multilateral communication across borders on managing water resources. building multilateral relationships with network of organizations, professionals, and individuals are not easy when there is animosity between the countries. twin is committed to mediate cooperative discussions and negotiations in neutral grounds to build relationships and cooperative solutions to transboundary issues. successful water negotiation, or reciprocal levels of communication frameworks, is essential to address interests, risks, and identity. as one program participant observed: “we have plans for the future, but it will only be possible by initiating transboundary water cooperation. we have to work together.” the participants identified that any governance and environmental diplomacy frameworks will have to address public interests foremost, and addressing risks that threatens identity, respect, and shared humanity. put in the words of a participant: “we are concerned about the people. that's the first step. and the second step is trust-building. and the third step is to mobilize the communities from both sides to force their governments as a civil society to come to an agreement.” conflict and cooperation are two sides of a coin. lack of cooperation inevitably leads to conflict: civic participation and local-scientific knowledge construction are essential to build effective protocols for cooperation and evidence-based decision-making for lasting trust and peaceful relations among these neighbors in the kabul basins. conclusion the management of water resources is essential to reduce poverty, increase sustainable economic and social development, improve quality of life for all, as well as to ensure an adequate supply of water for future generations of afghans and pakistanis. this research summarized the outcomes of the environmental diplomacy workshop held between afghanistan and pakistan on water cooperation in the kabul river basin, one of the most heavily conflicted transboundary river basins in the world. as afghanistan and pakistan are looking to modernize and build the nation, dependent upon neoliberal ideologies that increase the need for techno-scientific developments to increase productions and economic growth in agriculture, mining, energy production, and other manufacturing and social enterprises, the demand for water is increasing. water scarcity is already beginning to be felt in pakistan and its future further made indeterminate due to the uncertainties that climate change brings with its unpredictable shifts in glacial melts, flash floods, heat waves, and droughts with severe consequences to water security. strategies for dealing with water security are still dealt with state-centric coercive, hegemonic compliance producing mechanism where relative power determines the outcomes of discussions on bilateral water issues. the immediate problems identified to transboundary water rights, allocation, and security within this workshop included lack of focus on benefits to the local and rural population in the region, lack of access to http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 117 university of bamberg press water in rural areas, degrading water quality from multiple activities, and lack of scientific knowledge on the transboundary river systems to build better transboundary water relations, cooperation, and treaties. new adaptive governance strategies are needed to identify local concerns with water rights, allocation, and bilateral water cooperation among riparian partners to generate effective solutions for both sustainable use and conservation of water resources. the workshop also highlighted that transboundary water cooperation can be a powerful avenue for peace-building, to ease the strained border relationships and political tensions. the workshop stressed that one does not have to wait decades to formulate an official kabul river basin treaty to enforce water cooperation. rather, cooperation starts with people on the ground and across the borders, who are of the same tribe and culture, who were separated by colonial drawing of boundaries. we show that track 2 and track 3 science and public diplomacy, or collaborative civicscientific social learning and engagement network can be an effective framework for building transboundary adaptive governance systems. such an effort on social learning and adaptive governance is focused on building a bottom-up process, by strengthening 1) community building and engagement, 2) local capacity and civic-scientific social learning and local knowledge construction capacity, and 3) transboundary engagement, exchange and advocacy with a specific focus on improving bilateral communication and trust across borders on managing water resources. the workshop also laid the foundation to the conception and establishment of the transboundary water in-cooperation network (twin), a network of networks that connects transboundary water groups globally. twin’s mission is to strengthen and mobilize both local, national, and global organizational networks that work towards equitable and reasonable sharing of water resources across borders for food, energy, livelihood, and ecosystem services. twin’s vision is to build peaceful resolutions for water cooperation among stakeholders and nations, to improve border relations, water use practices, water quality, ecosystem services, resilience to climate change, and public health on both sides of the border. twin has also incorporated the solutions identified by the workshop participants on facilitating negotiations between civic and scientific diplomacy by 1) enhancing community engagement and capacity building to promote water cooperation by helping to build localized social learning and knowledge construction practices to improve access to water and protect the quality of the water along transboundary rivers 2) identify the necessary information such as data collection, research, reports and mutually beneficial technical solutions to improve the quality of water resources, 3) convening meetings and negotiations to promote bilateral/multilateral communication across borders on managing water resources. democratic participation, and social learning can be solutions to clear local misconceptions, improve transboundary water governance and ecological stewardship in the kabul river basin and to build more trusting and peaceful relations among these neighboring countries. initializing localized transboundary civic-scientific social learning capacities and exchanges can be a powerful avenue to building a localized adaptive governance framework. references ali, s., & zia, a. (2016). transboundary data sharing and resilience scenarios: harnessing the role of regional organizations for environmental security. in zafar adeel, robert wirsing (ed.) imagining industan: overcoming water insecurity in the indus basin. heidelberg, springer. asian news international (ani). (2018). pakistan’s punjab province is poisoning sindh. the economic times. retrieved from https://economictimes.indiatimes.com/news/international/world-news/pakistans-punjab-province-ispoisoning-sindh/articleshow/63711014.cms awan, z. a. (2018). cpec: oppurtunities for chinese entrepreneurs in mining industry in pakistan. china dialy. retrieved from http://www.chinadaily.com.cn/a/201807/10/ws5b446ef2a3103349141e1e69.html baig, m. b., & al-subaiee, f. s. (2009). biodiversity in pakistan: key issues. biodiversity, 10(4), 20–29. https://doi.org/10.1080/14888386.2009.9712858 http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 118 university of bamberg press bigas, h. (ed.). (2012). the global water crisis: addressing an urgent security issue. in papers for the interaction council, 2011-2012. unu-inweh. butterworth, j., mcintyre, p., & da silva wells, c. (eds.). (2011). switch in the city: putting urban water management to the test. the hague, netherlands. retrieved from http://www.switchurbanwater.eu/outputs/pdfs/switch_in_the_city.pdf carius, a., dabelko, g. d., & wolf, a. t. (2004). water, conflict, and cooperation. environmental change and security project report, (10), 60–66. folke, c., hahn, t., olsson, p., & norberg, j. (2005). adaptive governance of social-ecological systems. annual review of environment and resources, 30, 441–473. https://doi.org/10.1146/annurev.energy.30.050504.144511 fontana, b. (1993). hegemony and power: on the relation between gramsci and machiavelli / benedetto fontana. minneapolis: minneapolis : university of minnesota press. gerlak, a. (2007). regional water governance and institutional arrangements around transboundary waters. in international studies association annual meeting. chicago: http://www. allacademic. com/meta/p179029_index. html gulab, s. a. (2016). waterborne diseases. pakistan observer. retrieved from https://pakobserver.net/waterbornediseases-2/ gupta, j. (2017). polluted waters bring disease and death to afghans. the third pole. retrieved from https://www.thethirdpole.net/en/2017/07/27/polluted-waters-bring-disease-and-death-to-afghans/ haddadin, m. j. (2001). water scarcity impacts and potential conflicts in the mena region. water international, 26(4), 460–470. https://doi.org/10.1080/02508060108686947 hearns, g. (2015). the helmand river and the afghan-iranian treaty of 1973 international water law project blog. retrieved from https://www.internationalwaterlaw.org/blog/2015/07/23/the-helmand-river-and-the-afghan-iraniantreaty-of-1973/ hussain, m. (2015). new film, among the believers, looks inside pakistan’s madrassas. the intercept. retrieved from https://theintercept.com/2015/05/16/among-believers-pakistans-education-crisis/ hussain, m. (2018). how buying baby powder helps fund the taliban and islamic state in afghanistan. the intercept. retrieved from https://theintercept.com/2018/05/22/afghanistan-mining-talc-taliban-islamic-state/ iucn. (2010). towards kabul water treaty: managing shared water resourcespolicy issues and options. karachi. retrieved from https://cmsdata.iucn.org/downloads/pk_ulr_d3_1.pdf jägerskog, a. (2012). transboundary waters: current status of cooperative efforts on transboundary waters. in background paper for the ministerial roundtable on transboundary waters, world water forum. marseille, france. jansky, l., pachova, n. i., & murakami, m. (2004). the danube: a case study of sharing international waters. global environmental change, 14(suppl.), 39–49. https://doi.org/10.1016/j.gloenvcha.2003.11.008 julien, f. (2012). hydropolitics is what societies make of it (or why we need a constructivist approach to the geopolitics of water). international journal of sustainable society, 4(1–2), 45–71. https://doi.org/10.1504/ijssoc.2012.044665 kakakhel, s. (2017, march 2). afghanistan-pakistan treaty on the kabul river basin? the third pole. retrieved from https://www.thethirdpole.net/en/2017/03/02/afghanistan-pakistan-treaty-on-the-kabul-river-basin/ kiani, k. (2013). pakistan, afghanistan mull over power project on kunar river. dawn. retrieved from https://www.dawn.com/news/1038435 kliot, n. (1993). water resources and conflict in the middle east. london: routledge. kugelman, m., alam, a. r., & bakshi, g. (2011). peace through water. foreign policy. retrieved from https://foreignpolicy.com/2011/12/02/peace-through-water/ koliba, c., meek, j. w., & zia, a. (2010). governance networks in public administration and public policy (1st ed.). boca raton, fl, usa: crc press, inc. lee, k. n. (1999). appraising adaptive management. ecology and society. https://doi.org/10.1201/9781420042597 lowi, m. r. (1993). the politics of a scarce resource in the jordan river basin. cambridge: cambridge university press. mashal, m. (2012). what iran and pakistan want from afghans: water. time. retrieved from http://world.time.com/2012/12/02/what-iran-and-pakistan-want-from-the-afghans-water/ micallef, j. v. (2015a). islamic state: its history, ideology and challenge. vancouver: antioch downs press. micallef, j. v. (2015b). afghanistan and pakistan: the poisoned legacy of the durand line. huffington post: the blog. retrieved from https://www.huffingtonpost.com/joseph-v-micallef/afghanistan-and-pakistan_b_8590918.html mirumachi, n. (2015). transboundary water politics in the developing world. abingdon: routledge. molle, f., & hoanh, c. t. (2011). implementing integrated river basin management in the red river basin, vietnam: a solution looking for a problem? water policy, 13(4), 518-534.https://doi.org/10.2166/wp.2011.012 mudabber, z. (2016). afghanistan’s water-sharing puzzle. the diplomat. retrieved from https://thediplomat.com/2016/11/afghanistans-water-sharing-puzzle/ mukhtarov, f., & gerlak, a. k. (2013). river basin organizations in the global water discourse: an exploration of agency and strategy. global governance, 19(2), 307–326. https://doi.org/10.1525/sp.2007.54.1.23. nagheeby, m., & warner, j. (2018). the geopolitical overlay of the hydropolitics of the harirud river basin. international environmental agreements: politics, law and economics, 18(6), 839-860. doi:10.1007/s10784-018-9418-9 oregon state university department of geosciences. (2011). transboundary freshwater dispute databases. retrieved from www.transboundarywaters.orst.edu/ http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 119 university of bamberg press pahl-wostl, c. (2007). transitions towards adaptive management of water facing climate and global change. water resources management, 21(1), 49–62. https://doi.org/10.1007/s11269-006-9040-4 pahl-wostl, c. (2009). a conceptual framework for analyzing adaptive capacity and multi-level learning processes in resource governance regimes. global environmental change, 19(3), 354–365. https://doi.org/10.1016/j.gloenvcha.2009.06.001 petersen-perlman, j. d., veilleux, j. c., & wolf, a. t. (2017). international water conflict and cooperation: challenges and opportunities. water international. https://doi.org/10.1080/02508060.2017.1276041 petersen-perlman, j. d., watson, j. e., & wolf, a. t. (2018). transboundary unbound: redefining water conflict and cooperation for contemporary challenges that extend beyond watersheds, regions, and water. in k. conca & e. weinthal (eds.), the oxford handbook of water politics and policy. oxford: oxford university press. price, g., alam, r., hasan, s., humayun, f., humayun, m., karki, c. s., mittra, s., saad, t., saleem, m., saran, s., shakya, p. r., snow, c., & tuladhar, s. (2014). attitudes to water in south asia. london: the royal institute of international affairs. retrieved from https://www.chathamhouse.org/sites/default/files/field/field_document/20140627watersouthasia.pdf rizvi, m. (2003). planned afghan reservoirs worry pakistan. inter press service. retrieved from http://www.ipsnews.net/2003/09/environment-planned-afghan-reservoirs-worry-pakistan/ ruiz, h. a. (2002). afghanistan: conflict and displacement 1978 to 2001. forced migration review, (13). retrieved from https://www.fmreview.org/sites/fmr/files/fmrdownloads/en/september-11th-has-anything-changed/ruiz.pdf saeed, b. a., hassani, s., & malyar, i. (2016). transboundary basin management under conditions of latent conflict: a multi-sectoral and multi-disciplinary approach towards the kabul river basin. islamabad. retrieved from https://pk.boell.org/2017/07/31/transboundary-basin-management-under-conditions-latent-conflict-multi-sectoraland-multi salman, s. m. a. (2007). the helsinki rules, the un watercourses convention and the berlin rules: perspectives on international water law. international journal of water resources development, 23(4), 625–640. https://doi.org/10.1080/07900620701488562 shroder, j. (2016). hydro-hegemony in afghanistan and surrounding countries. in j. shroder & s. j. ahmadzai (eds.), transboundary water resources in afghanistan (pp. 339-359). boston: elsevier. slaughter, a. m. (2011). international relations, principal theories. max planck encyclopedia of public international law, 7. https://doi.org/10.5305/amerjintelaw.107.2.0510 sneddon, c. (2013). water, governance and hegemony. in contemporary water governance in the global south: scarcity, marketization and participation (pp. 13–24). london: routledge. https://doi.org/10.4324/9780203076361 sosland, j. (2007). cooperating rivals: the riparian politics of the jordan river basin. new york: state university of new york press. subramanian, a., brown, b., & wolf, a. (2012). reaching across the waters: facing the risks of cooperation in international waters. water papers. washington, d.c. retrieved from https://openknowledge.worldbank.org/handle/10986/17256 thomas, v., azizi, m., & behzad, k. (2016). developing transboundary water resources: what perspectives for cooperation between afghanistan, iran, and pakistan? kabul. retrieved from https://areu.org.af/archives/publication/1607 trottier, j. (2003). water wars: the rise of a hegemonic concept – exploring the making of the water war and water peace belief within the israeli-palestinian conflict. paris: unesco ihp. un. (2013). promoting water cooperation: legal frameworks and institutional arrangements. in un-water annual international zaragoza conference. zaragoza, spain. retrieved from http://www.un.org/waterforlifedecade/water_cooperation_2013/pdf/info_brief_legal_frameworks.pdf un. (2014). international decade for action “water for life” 2005-2015. focus areas: water quality. retrieved from http://www.un.org/waterforlifedecade/quality.shtml un general assembly. (2014). convention on the law of the non-navigational uses of international watercourses: 1997. unep. (2008). un-water thematic paper transboundary waters: sharing benefits, sharing responsibilities. retrieved from www.unwater.org/downloads/unw_transboundary.pdf wolf, a. t., kramer, a., carius, a., & dabelko, g. d. (2005). managing water conflict and cooperation. state of the world 2005 redefining global security, 80–99. retrieved from http://tbw.geo.orst.edu/publications/abst_docs/wolf_sow_2005.pdf world bank. (2013). afghanistan resource corridor development: water strategy. final kabul river basin report, ver. 4.0 (vol 2.). executive summary. washington, dc. retrieved from http://documents.worldbank.org/curated/en/877711468187751455/pdf/797210v20bri0p0box0379789b00public0 .pdf wwap. (2012). managing water under uncertainty and risk: the united nations world water development report 4. paris: unesco. retrieved from http://www.unesco.org/new/en/natural-sciences/environment/water/wwap/wwdr/wwdr42012/ yoffe, s., wolf, a., & giordano, m. (2003). conflict and cooperation over international freshwater resources: indicators. water resources, 39(5), 1109–1126. https://doi.org/10.1029/2003wr002530 zeitoun, m., & warner, j. (2006). hydro-hegemony a framework for analysis of trans-boundary water conflicts. water policy. https://doi.org/10.2166/wp.2006.054 http://dx.doi.org/10.20377/cgn-80 complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 101-120 doi: http://dx.doi.org/10.20377/cgn-80 120 university of bamberg press zia, a. & hameed, k. (2014). politics of conflict in pakistan’s tribal areas: vulnerability reduction in violence-prone complex adaptive systems. in aman, m. m. & aman, m. j. p. (eds.), middle east conflicts & reforms. washington, dc: policy studies organization/westphalia press. zia, a., kauffman, s., koliba, c., beckage, b., vattay, g., & bomblies, a. (2014). from the habit of control to institutional enablement: re-envisioning the governance of social-ecological systems from the perspective of complexity sciences. complexity, governance & networks, 1(1). doi:10.7564/14-cgn4. http://dx.doi.org/10.20377/cgn-80 introduction water cooperation efforts in the kabul river basin theoretical underpinnings: building adaptive governance systems for water cooperation methodology brief overview of the study site—kabul river results hydro-hegemonic challenges to water cooperation in the kabul river basin historical aspects that fuel coercive and hegemonic compliance regimes: geographic capacity and covert coercive actions: neoliberal governance approaches and challenges to local water security: rural access to water and challenges to water security: knowledge gaps to water cooperation building localized civic-scientific social learning and engagement networks for water cooperation and adaptive governance community building and engagement for trust building: local capacity and civic-scientific social learning and local knowledge construction: transboundary engagement, exchange and advocacy on managing water resources: conclusion responsive governance and harmful microbial blooms on lake erie: an abm approach complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 24 university of bamberg press responsive governance and harmful microbial blooms on lake erie: an abm approach authors: d.g. webster and tyler pavolvich dartmouth college, environmental studies, new hempshire, u.s.a. e-mail: d.g.webster@dartmouth.edu, in general, decision makers tend to respond to problems rather than prevent them. in political science, this process is called responsive governance and is associated with complex dynamics such as availability cascades and punctuated equilibrium. most authors treat problems as one-time events, like oil spills or political scandals. in this paper, we use an agent-based model of the lake erie watershed to explore how responsive governance evolves along with an on-going but noisy environmental problem: harmful microbial blooms. this conceptual model features a two-level decision process based on jones and baumgartner (2005). meta-agents representing the individual level of analysis “perceive” blooms either directly via observation if they are near a bloom or indirectly through the media. as a meta-agent observes more blooms, their concern increases until it crosses an action threshold, at which point they use simple cost-benefit analysis to select from a range of options. one of these options is to send a signal to their policy agent, which aggregates these political signals based on a range of assumptions and then decides on actions in much the same way as the metapopulations themselves. we examine two major scenarios, one in which there is a single policy maker managing the entire region and one where there are 5 policy makers, each separately regulating a demographically and geographically distinct region. although the model is relatively simple, it lets us explore how variability in risk perception and responsive governance shape the functioning of the entire coupled human and natural system, including biophysical feedbacks. keywords: harmful algal blooms; governance treadmill; responsive governance; social ecological systems; lake erie introduction harmful microbial blooms (hmbs) cause many problems in aquatic systems, both for ecosystems and for human life. microbes like algae and bacteria are common in aquatic environments. most use photosynthesis for energy. thus, they need sunlight and nutrients like nitrogen and phosphorous to produce food. given sufficient nutrients and warm temperatures, populations of micro-organisms should be consistent between microbes vs. micro-organisms can grow rapidly, creating blooms that cover large areas. many blooms are healthy and can provide food for other organisms, but some blooms are harmful because they lead to oxygen depletion in the water or because the microbes in the bloom produce toxic substances. these are often referred to as harmful algal blooms, but we use the broader term harmful microbial blooms, which includes bacteria as well as algae (wilson et al., 2019). hmbs can be prevented through governance mechanisms that reduce the nutrient pollution that feeds blooms. once high levels of nutrients are in a system, reducing bloom activity is much more difficult, both because of natural factors (nutrient pollution stored in lake sediments) and human factors (increased economic reliance on polluting processes such as the use of chemical fertilizers). nevertheless, people and governments rarely intervene proactively to prevent blooms. instead, rules to reduce nutrient pollution are only put in place after a series of large-scale hmbs that have significant impacts on human health or welfare (bullerjahn et al., 2016; paerl & otten, 2013). according to webster (2015a), this process of responsive governance is common in other environmental issue areas and can lead to a cycling back and forth between more and less effective response in a dynamic that she calls the governance treadmill. http://dx.doi.org/ mailto:d.g.webster@dartmouth.edu complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 25 university of bamberg press support for the concept of responsive governance comes from several literatures. it is fundamentally based on how people and organizations deal with problems. research in psychology and decision science shows that, while people do respond proactively to risks that are easily imagined and important to them (practically or emotionally), the complexity of environmental issues tends to dampen the perception of risk until they experience the effects themselves or hear emotive stories about the effects on people like them (keller, siegrist, & gutscher, 2006; slovic, 1987). responsive behavior is also observed in organizations, and can be further delayed by high transactions costs and other barriers to information sharing (march & simon, 1993; simon & barnard, 1957). in political science, responsive governance is used most often to explain how governments change. for instance, higgs (1987) claims that as policy makers respond to problems they create bureaucratic agencies that persist, expanding the size of government unnecessarily. in his famous work, hirschman (1993) shows how public responses to economic, environmental, and social problems pressure different types of governments to become more or less authoritarian over time. in an interesting expansion on that work, kuran and sunstein (1995) show how public response can display complex properties, such as when information cascades led to rapid transition away from soviet dominance in eastern europe; as more people spoke out against these oppressive regimes, others were more willing to speak out as well, creating a positive feedback loop or cascade of response that culminated in the velvet revolution. for environmental management specifically, as early as the 1970s scholars showed that decisions makers tend to respond to environmental harm rather than prevent it, though patterns of responsive governance can be traced back much farther (walters & hilborn, 1978; webster, 2015a). more recently, political scientists started to develop theories around issue-attention cycles, showing that, in democracies, policy making tends to occur in response to high levels of attention from interests groups or the general public, usually in response to experienced effects or media coverage of some type of crisis situation. these authors adopted the concept of punctuated equilibrium, arguing that the policy process is punctuated by phases of high activity around periods of heightened attention (baumgartner & jones, 2009; jones & baumgartner, 2005a; kingdon, 2011). like higgs, authors in this field generally assume that problems do not persist for very long, either because they were ephemeral to begin with or because governments enact effective policies to solve them. as shown by webster (2015a), this assumption rarely holds for environmental issues, which are usually increasing in severity and are rarely “solved” by early responses. indeed, it may take multiple periods of crisis and response before decision makers arrive at an effective combination of governance mechanisms and, even then, there is no guarantee that these rules will remain effective in perpetuity. as a result of these shorter response cycles, in the long term environmental governance tends to shift between phases of more and less effective response in a process that webster (2015a) calls the governance treadmill. as shown on the left-hand side of figure 1, the treadmill starts in an ineffective cycle, where an increasing environmental problem sends more and more intense signals to scientists, decision-makers, firms, the public, and other actors. these signals are translated into political concern, which is increasing as long as signals are increasing. if policy remains ineffective, these cycles will continue until the system collapses. however, when concern is sufficient and solutions are available, governance can shift to a more effective cycle, where the problem is decreasing, signals are receding, and political concern is on the wane. the dampening of problem signals can generate a crisis rebound effect, which shifts policy back to the ineffective side of the treadmill. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 26 university of bamberg press figure 1 the governance treadmill. source: modified from webster (2015a, fig 1.1) figure 2 compares the change in policy activity over time as predicted by the theory of punctuated equilibrium and the basic governance treadmill shown in figure 1. as explained above, under punctuated equilibrium there is a flat trend in policy making that is punctuated by spikes of activity around crisis events that, while not completely random appear to occur randomly because they result from complex interactions within large socioeconomic systems. although the apparent randomness remains, the governance treadmill predicts that spikes of attention will occur around an increasing trend as long as response is ineffective, and the problem is still sending out stronger and stronger signals. attention and therefore policy-making will decrease substantially once an effective policy is implemented, and problem signals decline. each spike in figure 2 is a response cycle but only one round of the treadmill, or one transition from ineffective to effective governance is illustrated. if we extend the timeline long enough it is possible that policy will become ineffective again, either due to the rebound effect or because of exogenous changes like higher prices or new technologies that undermine existing policy. this would start another round of the treadmill, with growing spikes of response activity around an increasing trend. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 27 university of bamberg press figure 2 two perspectives on responsive governance over time of course, there are many variations on the example shown in figure 2. the governance treadmill can run smoothly, with relatively short periods of ineffectiveness and relatively long periods of effectiveness. this would be a healthy treadmill. when problems arise, they are dealt with in a timely and effective manner. treadmills can also function poorly, getting mired in ineffective cycles as inaction or the implementation of palliative measures ensures that the problem continues to worsen over time. these dysfunctional treadmills can result in stagnation, crisis, and even a total system break, such as an ecological shift or a political revolution. they usually occur due to disconnects that prevent problem signals from reaching or influencing decision makers in a timely and effective way. temporal disconnects like the threshold effects displayed by harmful microbial blooms (see below) are already well-recognized in the literature on complex systems, but there are many others. this paper focuses on power disconnects, which occur when those who receive the strongest problem signals have little political influence or, in an alternative framing, when those who have political influence are insulated from the relevant problem signals (webster, 2015a, 2015b). essentially, power disconnects ensure that problem signals do not create sufficient concern to generate a governance response, thereby divorcing the problem from those who have the power to implement solutions and flattening the policy cycles shown in figure 2. thus, where there are power disconnects we can expect effective response to be substantially delayed, keeping the treadmill in an ineffective cycle until a major crisis closes the gap between those who are affected by the problem and those who have the ability to mitigate it. such crises can be large-scale ecological failures that have broad-reaching impacts, but they can also be socioeconomic upheavals, such as the information cascades described by kuran and sunstein (1995) as explained above. either way, these crisis events usually carry high human and environmental costs and may be destabilizing to other components of the system. this is why it is important to understand how power disconnects affect the governance treadmill and what can be done to reduce them. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 28 university of bamberg press because it encapsulates complex social ecological systems or coupled human and natural systems, modeling the governance treadmill is difficult and interpreting highly detailed models is problematic. in this study, we created a stylized agent-based model (abm) of the treadmill for hmbs in the lake erie watershed that is relatively easy to understand and interpret. it will allow us to test two major scenarios, one in which there is a single policy maker managing the entire region and one where there are 5 policy makers, each separately regulating a demographically and geographically distinct region. power disconnects are relatively narrow in the single regulator scenario because all actors have similar levels of influence, meaning that people who are affected by blooms have similar levels of influence to those who contribute to the nutrient pollution problem. power disconnects are wider in the multidecision maker scenario because the farmers who contribute most to the problem of blooms and are affected the least by them are concentrated in a single governance region, giving them much more power over policy making in that region than they are able to wield in the single-decision-maker scenario. this mirrors state/province vs. national-level response as described in the next section, where we provide background information on hmbs in lake erie and a brief overview of governance response to this problem. we then describe our model and our testing methods. after presenting our results, we return to a discussion of the concept of the governance treadmill and methods for improving on this model in future work. background although they also occur naturally, hmbs are exacerbated by a number of human activities. three are most important. first, humans add nutrients to the system by allowing human or animal waste to enter waterways and by oversaturating the land with (mainly synthetic) fertilizers, which then wash off into lakes and streams. second and relatedly, land uses like farming and urban development increase runoff and erosion, moving more nutrients from the land into the water. third, anthropogenic climate change can increase blooms through both higher temperatures and through larger amounts of rainfall. as described by walker and salt (2006), among others, by increasing the nutrient loads both in the water and in the lake sediment, human activities can push lakes past an ecological threshold, setting off a cascade of changes that ultimately results in the eutrophication of the lake. essentially this means a rapid and sometimes irreversible transition from a high-oxygen plant and fish dominated system, which tends to be useful to humans, to a low-oxygen algae-dominated system, which reduces anthropogenic benefits such as recreation, fisheries, and clean water supply. because lake erie is very large, it is not as prone to such a complete transition, but the shallow western end of the lake is susceptible to blooms of microcystis cyanobacteria (cyano blooms) that display similar but reversible threshold effects. figure 3 shows a satellite image of a large-scale bloom event from 2011. although there is a visible depositing of sediment from lake sinclair via the detroit river, the majority of nutrient loading actually comes from the maumee river basin, which enters the western-most tip of the lake as it passes through toledo. blooms can spread from there as far east as cleveland. the canadian section of the lake can also be affected, though their contribution to the nutrient load in the lake is much lower than the inflows from the us side of the boarder. in 2011, this cyano bloom event was considered to be one of the worst since the 1970s, but these large-scale blooms have increased in frequency since that time, including massive blooms in 2013, 2014, 2015, 2016, and 2017. in 2016 and 2017, double bloom events were observed, with an early bloom in june/july followed by a second bloom in september/october (baker et al., 2019). there are other types of microbial blooms that occur in the deeper waters in the central and the north-eastern side of the lake, but these are not included in the model. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 29 university of bamberg press figure 3 satellite view of october 2011 cyano bloom in lake erie source: noaa coastwatch (modis) this region has gone through one and a half rounds of the governance treadmill, with a historical switch from ineffective to effective governance, similar to the pattern shown in figure 2, followed by a second transition back to the ineffective side. harmful microbial blooms first flared up in the lake in the 1950s. increasing bloom events caused growing socioeconomic costs that finally triggered management response in the form of improved sewage treatment and point-source controls in the 1960s and 1970s. in fact, large hmbs were well-publicized at the time and added to the national-level concern that culminated in the 1972 clean water act. these responses improved sewage treatment and other urban sources, effectively reducing nutrient pollution and related bloom events in the 1980s but starting in the 1990s, non-point source pollution increased greatly due to the expansion of agriculture and proliferation of highly soluble chemical fertilizers (wines, 2014). national-level legislation remains a primary driver of regulations to reduce hmbs in lake (wilson et al., 2019). the return of large-scale blooms described above sends a number of different types of signals, some of which are noisier than others. the most direct signals are those sensed by people in close proximity to the lake. large-scale cyano blooms turn the water bright green and they often are associated with a bad smell, green scums on the surface of the water and on beaches, and may cause irritation to bathers or pets that enter the water. large blooms can suffocate fish and other wildlife and, in some cases, the cyanobacteria produce toxic chemicals that poison everything in the water (paerl & otten, 2013). this sends direct signals when dead animals or plans are perceptible, but indirect signals such as scientific reports, media coverage, and government warnings are more frequent. scientists from the us national ocean and atmospheric administration (noaa) great lakes environmental research laboratory collect bi-weekly samples from over 40 stations on the western end of the lake. they also monitor blooms using satellite imagery. this information is provided to the public through their hab tracker web-page. though beaches are rarely closed due to hmbs, health advisories and other government warnings do occur. in fact, in 2014 a bloom forced the government to advise against using the municipal water supply, causing a severe water crisis for the city of toledo (neff et al., 2014; pearson, 2014). in that case, media coverage made the bloom an event of nation-wide concern. at a further remove from the core environmental problem, economic and political signals are also important. blooms tend to discourage lake-based tourism and recreation, depressing lake-side economies. as noted above, major cities can also be affected through higher taxes to pay for clean-up http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 30 university of bamberg press or alternative water sources. economic analysis has also showed that property values near the lake decline with more blooms (bingham, sinha, & lupi, 2015). people who experience these costs may send political signals to their representatives in several ways, most often through e-mails or phone calls, but also through protests, lobbying (esp. via professional organizations), and voting. of course, these are not the only voices that politicians hear. those people who might be affected by regulations may also send signals to protest potential new rules using similar channels. at the moment, most government action is aimed at reducing run-off from agriculture, so farmers are a very important interest group. while some farmers are environmentally concerned and already minimize their use of fertilizer and apply other practices to reduce run-off, other farmers feel economically threatened by regulations designed to curb habs (wilson, howard, & burnett, 2014). in this, they may receive signals from agro-business representatives and extension agents who are also politically opposed to regulations. decision makers must then aggregate all of these signals in some way. lake erie is embedded in a multi-scale governance system that ranges from community-based efforts at the local level to national-level legislation such as the clean water act of 1972. there is even some international coordination via the great lakes water quality agreement (wilson et al., 2019). at this stage we cannot incorporate all of these different levels into our analysis, and so instead focus on developing a set of models that incorporate increasingly complicated governance decision processes. we start with a single aggregator that simply sums the signals from all meta agents. this is somewhat representative of national-level response, although of course in the real world signals to the national government could come from other locations experiencing water quality issues as well. we compare this result to decisions made by 5 independent governance actors, each of which aggregates signals from different regions that have distinct geographic and demographic characteristics. notably, one particular region—representative of the maumee watershed in the erie case—contributes the majority of the nutrient pollution driving the cyano blooms but is also home to a large population of farmers who will generally resist being regulated. thus, our general expectation is that response will happen sooner in single decision maker mode and will be delayed or may not even occur in multiple decision maker mode. although the model is relatively simple, it lets us explore how variability in risk perception and responsive governance shape the functioning of the entire coupled human and natural system, including biophysical feedbacks. model & methods modeling responsive governance in a large, complex human and natural system like lake erie is a difficult task. analyzing a high-resolution, detailed model would also be difficult because of the many different sources of variation and emergent properties in the system. therefore, to maintain tractability in early stages of model development, we start with a relatively simple model that is easier to interpret. figure 4 illustrates the basic logic of our model (see appendix 1 for more detailed documentation of the model). additional detail is provided in subsequent sections. following the governance treadmill, the environmental problem (hmbs or bloom activity) sends signals to different metapopulations (metapops in figure 4) or agents that represent groups of similar individuals. we will refer to these types of agents as constituents throughout this text (e.g. farmers, coastal business owners, etc.). signals occur either through direct observation of blooms or their effects, or indirectly through media reports on factors like the size and location of blooms. if bloom signals are strong enough to generate high levels of public or interest group attention, policy maker agents will also receive indirect signals from their constituents, pressuring them to take action to reduce bloom activity. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 31 university of bamberg press figure 4 basic model diagram the internal decision processes for both constituent metapopulations and for policy makers are based on jones and baumgartner (2005b). we start by describing the constituent process. as shown in figure 3, signals from the bloom event and from the media increase the constituent’s level of concern regarding the hmbs problem. different types of signals are weighted differently across agents, depending on their pre-existing interests. thus, total concern is the weighted sum of concern from multiple signals. total concern will gradually dissipate, however, if no new signals are received by the agent. once total concern reaches a pre-set attention threshold, the agent uses its own cost-benefit analysis to select from a preset suite of optional actions. in this case, they can choose to do nothing or they can choose to send a signal to the policy agent(s). the policy agent(s) then use a similar process to aggregate constituent concern and, once they cross their attention threshold, they select a policy option. here, they decide on the maximum level of nutrients that farmer-agents can apply to patches within the watershed. changes in the nutrient level then alter the environmental conditions, reducing the likelihood of bloom activity. setup initial setup involves creating global variables, breeds of agents, and assigning patchand breedspecific variables (see appendix 2 for a list of all variables, their initial values, and the sources used to parameterize these values). the only global variables fixed in the setup code are related to lake and bloom variables. other global variables are controlled by sliders in the user interface. these variables relate to how different agents perceive and are affected by different signals. they include the http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 32 university of bamberg press magnitude of a signal, how susceptible an agent is to that signal, how long they care about that type of signal for, how quickly an agent loses concern, and at what level of concern an agent takes action (see appendix 3 for a list of these variables and the values used in the simulations described here). patchspecific variables include land-cover type and blooming conditions. they will be described in greater detail below. there are seven breeds of agents in the model: “urban.dwellers”, “rural.non.farmers”, “coastal.busines.owners”, “farmers”, “government.actors”, “media.sources”, and “scientists”. variables assigned to each breed primarily control their internal decision processes, particularly signal processing and internal levels of concern. of these breeds, “urban.dwellers”, “rural.non.farmers”, “coastal.busines.owners”, and “farmers” are metapopulations of constituents that respond to direct and indirect signals from bloom events as described above; “government.actors” aggregate signals from constituents in the areas that they control (see below) and from “media.sources” ; there is only one “media.source” agent that monitors bloom events and reports on specific characteristics to other breeds of agents in the model; “scientist” is also a single agent that uses bloom-related variables to calculate the range of nutrient application levels that are likely to minimize bloom activity and then passes this information to “government.actors”. patch setup is largely based on maps of the lake erie watershed and features deemed to be important based on our review of the literature on habs in the region (see background section). first, each patch is assigned a land cover type, delineating between “lake” patches, “coast” patches, and “land” patches. all patches outside of the lake erie watershed have no land cover type. next, major cities are placed on the map in approximately their location in real life. smaller cities are then distributed randomly throughout the remaining land area. “beaches” and “water intakes” are created in coastal and lake patches respectively, and are distributed in areas proximate to their real-world locations. of the remaining land patches, those within the approximate contour of the maumee river watershed are designated as “farms”. of course, there are many other farms outside this watershed, but the maumee is the primary source of nutrient pollution in the western basin, as described above, so we keep the analysis tractable by only modeling nutrient application on farms in this area. patches are given population characteristics according to their land cover type and the presence or absence of a city. these characteristics include population size and the proportions of urban dwellers, coastal business owners, farmers, urban non-farmers, and “tourists”. note that “tourists” are not agents in the model but are an attribute of coastal patches that changes with bloom activity. we populate the model landscape by placing constituent agents in their respective positions and assigning appropriate population proportions (see figure 5). “urban.dweller” agents are located in the major cities and represent the relevant proportion of population in each patch. for instance, toledo, oh, is represented by 5 patches, each of which sprouts an “urban.dweller” agent that is assigned to represent 1/5 of the total population of toledo. smaller cities and towns are only one patch, each of which sprouts a single “urban.dweller” agent with a population randomly drawn from a distribution based on the actual population sizes of small to medium sized cities in the region. coastal cities near beaches also sprout a “coastal.business.owner” agent that tracks the number of “tourists” in neighboring patches as well as other blooms signals. non-farm “land” patches sprout “rural.non.farmer” agents and “farm” patches sprout “farmer” agents. this nomenclature is a bit misleading, as “rural.non.farmers” can represent people who farm in areas outside of the watershed, but it follows with our designation of “farmers” as agents that control the level of nutrients in the http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 33 university of bamberg press maumee watershed only. the “media.source” agent and “scientist” are placed in patches outside of the erie watershed. figure 5 landscape and governance districts finally, regional boundaries are created (red lines in figure 5) and five “government.actor” agents are placed within their regions but outside of the area occupied by active watershed patches. regional boundaries are drawn and five “government.actor” agents are created on patches outside of the erie watershed. when the model is run with a single decision maker, the “government.actor” labeled “a” received signals from all constituent agents in the model and its policy decisions apply to all agents in the model. in the multiple decision-maker version of the model, each of the “government.actor” agents only receives signals from constituents in its region and can only apply policies to agents in said region. processes the processes run by the model proceed in the following order: 1) track time and seasons, 2) farming, 3) microbial blooms, 4) track tourists visiting beaches, 5) determine what signals the media will present, and 5) calculate the level of concern of and actions taken by urban dwellers, rural nonfarmers, coastal business owners, farmers, and the government. we should note that each agent type has a procedure for calculating concern and whether they take action. bloom-related procedures farming contains the decision of farmers about how much nutrients they apply while farming. this is relevant to the simulation in two ways: it is the source of nutrient pollution to the lake, and the level of nutrients applied can be regulated by policy. farmers apply nutrients to their fields during the growing season, between calendar weeks 18 and 42 to represent the ohio farming season. farmers apply 10 units of nutrients under normal conditions, and lower amounts if the government is regulating nutrient application. (see description of the nutrient regulation procedure http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 34 university of bamberg press below.) farmers have a certain probability of applying nutrients at each time step, controlled by the variable “fertilizer.application.rate,” whose values range from 0 to 1. microbial blooms can occur if the sum of all “nutrients” on the landscape times the “rainfall.factor” (takes values between 0-1) exceeds the threshold required to have a bloom (“nutrient.threshold”), and if the time of year is within the warm season (”temperature“ = “hot”; there are only 2 seasons, “hot” and ”cold”). when the conditions are met, a bloom is seeded in a specified number of lake patches (“bloom.seeds” number of patches set “blooming?” = 1) within the western basin of the lake (“pxcor” <= 46). the bloom then spreads to a certain number of its neighbors (“num.neighbors.bloom.spread”), which can be selected by the user using a slider in the interface. values for bloom spread range between 1 and 8, with a default level of 4. a patch remains blooming for a certain number of time steps (“max.time.blooming”), which can be set anywhere between 1 and 20 with a default level of 7. patches stop blooming if the temperature changes to “low” or nutrients drop below the threshold. coastal tourism declines if blooms occur near public beaches. if the season is warm and none of the lake patches neighboring a beach are blooming, the number of tourists is calculated using the following equation: 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = 𝑝𝑝𝑡𝑡𝑝𝑝𝑡𝑡𝑝𝑝𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑝𝑝 ∗ 𝑝𝑝𝑡𝑡𝑡𝑡𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 (1) where “proptourists” is the proportion of people at a patch that are normally tourists. as noted above, “tourists” are not agents but are a patch-related variable. if any of a beach’s neighboring lake patches are blooming, “tourists” are calculated with the following equation: 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = 𝑝𝑝𝑡𝑡𝑝𝑝𝑡𝑡𝑝𝑝𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑝𝑝 ∗ 𝑝𝑝𝑡𝑡𝑡𝑡𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 − 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡. 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡. 𝑏𝑏𝑏𝑏. 𝑏𝑏𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡 ∗ 𝑝𝑝𝑡𝑡𝑝𝑝𝑡𝑡𝑝𝑝𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑝𝑝 ∗ 𝑝𝑝𝑡𝑡𝑡𝑡𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 (2) where “tourism.loss.by.bloom” is the proportion of tourists who no longer visit beaches when there is a nearby bloom. there is also a “scientist” agent that provides advice to decision-makers based on bloom characteristics. specifically, this agent sets the maximum advisable level of nutrients per application as: 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑀𝑀𝑀𝑀𝑀𝑀. 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑀𝑀𝑀𝑀𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 = 𝑛𝑛𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑛𝑛𝑡𝑡.𝑡𝑡ℎ𝑡𝑡𝑛𝑛𝑡𝑡ℎ𝑡𝑡𝑜𝑜𝑜𝑜(1−𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟.𝑟𝑟𝑟𝑟𝑓𝑓𝑓𝑓𝑓𝑓𝑟𝑟+𝑟𝑟𝑛𝑛𝑓𝑓𝑟𝑟𝑟𝑟𝑛𝑛𝑟𝑟𝑓𝑓.𝑑𝑑𝑛𝑛𝑓𝑓𝑟𝑟𝑑𝑑) (𝑓𝑓𝑓𝑓𝑛𝑛𝑟𝑟𝑓𝑓..𝑟𝑟𝑟𝑟𝑟𝑟𝑓𝑓𝑛𝑛𝑟𝑟𝑓𝑓)/(𝑟𝑟𝑛𝑛𝑟𝑟𝑓𝑓𝑟𝑟𝑟𝑟𝑟𝑟𝑓𝑓𝑛𝑛𝑟𝑟.𝑟𝑟𝑎𝑎𝑎𝑎𝑟𝑟𝑟𝑟𝑓𝑓𝑟𝑟𝑓𝑓𝑟𝑟𝑓𝑓𝑟𝑟.𝑟𝑟𝑟𝑟𝑓𝑓𝑛𝑛) (3) http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 35 university of bamberg press in order to account for some uncertainty created by the potential build-up of nutrients on farms, the scientist agent sets a minimum advisable level of nutrients per application as: if rainfall.factor > 0.4: nutrients.min.recommended = nutrients.max.recommended 1 (4) if rainfall factor is <= 0.4: nutrients.min.recommended = nutrients.max.recommended – 2 (5) basically, when rainfall is higher, it is more likely that the nutrients on a given patch will be depleted within the two-week on average period between applications, so with more rainfall the buildup of nutrients is less likely, and the lower bound on scientific advice will be higher. with less rainfall, nutrient build-up is more likely, meaning that a lower minimum application rate is needed to prevent a bloom event is lower. this is counter-intuitive if you think about real-world systems, where rainfall is not a constant factor but rather higher average rainfall can mean more storm events that carry large nutrient loads into the lake. for tractability, we do not represent such realism in the model. rainfall is a constant percentage removed from the nutrients on the land and therefore sudden large inflows are more likely if nutrients are allowed to build up over several ticks in the model. concern “concern” is a variable held by each agent that calculates and stores their level of concern regarding the microbial blooms. concern at each time step is generated from the multiple signals agents receive according to the following equation: 𝑐𝑐𝑡𝑡𝑝𝑝𝑐𝑐𝑛𝑛𝑡𝑡𝑝𝑝 = ∑ 𝑡𝑡𝑡𝑡𝑡𝑡𝑐𝑐𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡𝑏𝑏𝑡𝑡𝑝𝑝𝑡𝑡𝑡𝑡𝑏𝑏𝑡𝑡𝑡𝑡𝑠𝑠𝑛𝑛𝑠𝑠𝑜𝑜.𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛 ∗ 𝑡𝑡𝑡𝑡𝑠𝑠𝑝𝑝𝑝𝑝𝑝𝑝. 𝑣𝑣𝑝𝑝𝑝𝑝𝑡𝑡𝑛𝑛𝑡𝑡𝑡𝑡𝑠𝑠𝑛𝑛𝑠𝑠𝑜𝑜.𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑡𝑡𝑡𝑡𝑠𝑠𝑛𝑛𝑠𝑠𝑜𝑜.𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑡𝑡 (6) for calculations of concern, there are 13 types of signals in the model. four types of signals are directly assessed by constituent agents in coastal regions through commands to ask neighboring patches for information or through information gathered from other agents (“signal.value.intake”, “signal.value.seeing”, “signal.value.tourism”, and “signal.value.regulation”). five types of signals are calculated by the media agent and then called by other agents in the model (“signal.toledo”, ”signal.sandusky”, ”signal.large.bloom”, ”signal.intakes”, and “signal.beaches”). each of the four types of constituent agents can also send signals to government agents, though these are interpreted using a somewhat different equation (see below). we describe each signal in greater detail in the next section. the “susceptibilitysignal.type” (range = 0-1) variable accounts for differences in how much people care about different aspects of the bloom. for example, farmers might not care if coastal tourism decreases (“signal.value.tourism”), whereas coastal business owners care a lot. in contrast, farmers often react negatively to regulations that limit their use of fertilizer (“signal.value.regulation”), while coastal business owners do not. each type of agent has a different level of susceptibility to each type of signal http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 36 university of bamberg press that they receive. thus, media signals, which are the same across all agent types, will affect types of agents differently because of the variation in susceptibility. this interacts with signal variability for direct experience with the bloom, which does vary by agent location. urban dwellers’ concern comes from four types of signals: contaminated water intakes, blooms that occur close enough to the shoreline of their cities for people to see, decreases in coastal tourism, and media signals on blooms. rural non-farmers’ concern comes from media signals only. this assumes they are not directly impacted by the blooms, but may care about the health of the lake or other people who are impacted. coastal business owners’ concern comes from three sources: blooms that occur close enough to the shoreline for people to see, decreases in tourism in adjacent beaches caused by nearby blooms, and media signals on blooms. farmers’ concern comes from two sources: regulation of nutrient application and media signals. media signals generate concern in two ways; farmers have concern for the health of the lake and that they may face further regulations that will hurt their business. government actors are susceptible to signals from the media, but they also receive signals from constituent agents directly. this variable is named with the following format in the code: “susceptibility.agent.type.government”. an example is “susceptibility.farmers.government” for government susceptibility to farmers. because constituent agents represent different proportions of the population and because population characteristics vary by region, we calculate government concern from signals received from each breed using this equation: 𝑐𝑐𝑡𝑡𝑝𝑝𝑐𝑐𝑛𝑛𝑡𝑡𝑝𝑝 = (𝑡𝑡𝑡𝑡𝑡𝑡𝑐𝑐𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡𝑏𝑏𝑡𝑡𝑝𝑝𝑡𝑡𝑡𝑡𝑏𝑏. 𝑝𝑝𝑠𝑠𝑛𝑛𝑝𝑝𝑡𝑡. 𝑡𝑡𝑏𝑏𝑝𝑝𝑛𝑛. 𝐺𝐺𝑅𝑅𝐺𝐺𝑅𝑅𝑅𝑅𝑅𝑅𝑀𝑀𝑅𝑅𝑅𝑅𝑁𝑁 ∗ 𝑐𝑐𝑡𝑡𝑡𝑡𝑝𝑝𝑡𝑡 𝑝𝑝𝑠𝑠𝑛𝑛𝑝𝑝𝑡𝑡. 𝑡𝑡𝑏𝑏𝑝𝑝𝑛𝑛 𝑤𝑤𝑡𝑡𝑡𝑡ℎ [𝑐𝑐𝑡𝑡𝑡𝑡𝑝𝑝𝑡𝑡𝑝𝑝𝑐𝑐𝑡𝑡. 𝑠𝑠𝑡𝑡𝑣𝑣𝑛𝑛𝑡𝑡𝑝𝑝𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡? = 1]/(𝑐𝑐𝑡𝑡𝑡𝑡𝑝𝑝𝑡𝑡 𝑝𝑝𝑠𝑠𝑛𝑛𝑝𝑝𝑡𝑡. 𝑡𝑡𝑏𝑏𝑝𝑝𝑛𝑛) (7) by using the proportion rather than the total number of constituents of a given type that contact government to calculate concern we are able to restrict the political influence of each agent type to the variable “susceptibility.agent.type.government”. while it is likely that groups with larger populations in a governance region will have more influence if they are activated, the scale of the difference in population sizes would wash out all impacts of the susceptibility variable at high levels of concern (where most constituents are contacting the government agents). thus, we instead assume that population size is one of the factors that determines the magnitude of “susceptibility.agent.type.government” while the proportion of the population contacting the government is a better indicator of the level of concern regarding the environmental problem for that type of agent. this also makes it easier to interpret the results of the model, though alternative methods of calculating government concern should be considered in future work (see conclusion). http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 37 university of bamberg press total concern for each government agent was calculated by summing across the weighted signals received. since farmer agents tend to be more concerned with the negative effects of regulation, their signals end up reducing rather than increasing government concern. for all signal and agent types, multiplying the signal value by a constant level of susceptibility implies that each additional unit of signal creates the same amount of concern. more complex models could be used if there were theory or evidence to support them. concern decays over time linearly. each agent type has a different rate at which concern decreases, referred to as “decay.of.agent.type”. 𝑐𝑐𝑡𝑡𝑝𝑝𝑐𝑐𝑛𝑛𝑡𝑡𝑝𝑝𝑡𝑡 = 𝑑𝑑𝑛𝑛𝑐𝑐𝑝𝑝𝑏𝑏 ∗ 𝑐𝑐𝑡𝑡𝑝𝑝𝑐𝑐𝑛𝑛𝑡𝑡𝑝𝑝𝑡𝑡−1 + 𝑐𝑐𝑡𝑡𝑝𝑝𝑐𝑐𝑛𝑛𝑡𝑡𝑝𝑝𝑛𝑛𝑛𝑛𝑛𝑛 (8) concern over each signal type disappears completely if enough time passes without a new signal; the time threshold is called “attention.span.signal.type”. signals the “signal.valuesignal.type” variable allows for variability in the strength of signals at different points in time. signal values range from 0 to 1. direct signal variables are binary while media signals are calculated as proportions. table 1 describes how each signal type is calculated in the model. table 1 calculating signal types in the model direct signals description signal.value.intake if any patches that has a water intake on it and is within a radius of four from an agent, the signal value for intakes equals one. otherwise, this signal equals 0. signal.value.seeing if any neighboring patches are blooming, the signal value for seeing the bloom equals 1. otherwise, this signal equals 0. signal.value.tourism if it is the hot, tourists are counted at coastal beaches. if there is a bloom in any of a beach’s neighboring patches, tourism declines by the proportion “tourism.loss.by.bloom”. the signal value for tourism loss is one minus the proportion of tourists currently compared to the number of that would be there had there not been a nearby bloom. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 38 university of bamberg press signal.value.regulation if government actors are currently regulating nutrients, the signal value for regulation is 1. otherwise, this signal equals 0. indirect (media) signals signal.toledo proportion of patches that are blooming in lake patches neighboring toledo’s coastal patches signal.sandusky proportion of patches that are blooming in lake patches neighboring sandusky’s coastal patches signal.large.bloom reports the proportion of patches blooming if the total size of the bloom is > “large.bloom.threshold” signal.intakes proportion of water intake patches where a bloom is occurring signal.beaches proportion of beach-adjacent patches where a bloom is occurring as noted above, signal values of constituents’ concern, received by government actors, are calculated as the proportion of constituents with “contact.government?” = 1. for instance, if every farmer was concerned enough to contact the government, the government would receive a signal of 1 from farmers. action there are two main types of actions in the model: constituents can send signals to government agents and government agents can regulate nutrient levels, and farmer agents change the nutrients applied to their patches once regulations are put in place. both types of action only occur when the agent’s concern crosses an action threshold, triggering their action routine. we limit the actions available to each agent type severely in this preliminary model to maintain tractability and transparency. more elaborate choices among different types of action could be modeled in future work. urban dwellers, rural non-farmers, coastal business owners, and farmers can take action by contacting the government (“contact.government?” = 1). they do so when their concern exceeds their type-specific threshold, “concern.threshold.agent.type”. these signals are then aggregated in government concern as described above. government actors take action by setting the regulated level of nutrient application. government agents only regulate once per year, in the 40th week (just before the end of the growing season). this is a simple representation, but it reflects the fact that decision makers can rarely change agricultural policy in the middle of a growing season and that in any case it often takes time to collect scientific http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 39 university of bamberg press information and move policies through the process. of course, regulation is often much harder to pass than is depicted in this model, but we believe this is a good place to start. if government concern remains less than their action threshold for an entire season, they choose not to regulate, leaving the nutrient level where it is. if no regulations are in place, this will be the default level of nutrient application. if there are prior regulations, then they will simply remain in place for another season. if government concern was greater than their action threshold during the season, government agents will choose to regulate by reducing the amount of nutrients that farmer agents are allowed to apply to their patches. to determine the new level of nutrients, the government agent follows a set of conditions that is designed to gradually align regulated nutrient levels with levels that are close to those recommended by the scientist agent: if the average nutrients per field is greater than the maximum amount recommended by scientists plus one (“nutrients.max.recommended” + 1), the government agents will set 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑅𝑅𝑅𝑅𝐺𝐺𝑅𝑅𝑅𝑅𝑀𝑀𝑁𝑁𝑅𝑅𝑅𝑅 = 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑀𝑀𝑀𝑀𝑀𝑀. 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑀𝑀𝑀𝑀𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 + 𝑛𝑛𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑛𝑛𝑡𝑡𝑡𝑡.𝑠𝑠𝑡𝑡𝑡𝑡𝑜𝑜𝑡𝑡𝑛𝑛𝑜𝑜−𝑛𝑛𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑛𝑛𝑡𝑡𝑡𝑡.𝑀𝑀𝑀𝑀𝑀𝑀.𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑀𝑀𝑀𝑀𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 2 (9) if the average nutrients per field is greater than the max amount recommended but less than or equal to the max amount recommended plus one, 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑅𝑅𝑅𝑅𝐺𝐺𝑅𝑅𝑅𝑅𝑀𝑀𝑁𝑁𝑅𝑅𝑅𝑅 = 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑀𝑀𝑀𝑀𝑀𝑀. 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑀𝑀𝑀𝑀𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 (10) if the average nutrients per field is greater than the minimum amount recommended but less than the maximum amount recommended, 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑅𝑅𝑅𝑅𝐺𝐺𝑅𝑅𝑅𝑅𝑀𝑀𝑁𝑁𝑅𝑅𝑅𝑅 = 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑡𝑡𝑛𝑛𝑑𝑑 − 1 (11) if the average nutrients per field is less than the minimum amount recommended, government agents will not change the regulated level of nutrients. in other words, the minimum level of nutrient application recommended by the scientist agent serves as a lower bound for the government agents’ regulatory algorithm. there are two options for the process of how government actors make decisions: either as a single decision maker for the entire watershed, or with five decision makers, each representing a defined http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 40 university of bamberg press geographic area. if there is a single decision maker, the concerns of all stakeholders affect the government actor’s concern and, hence, level of nutrient regulation. under the multiple-decisionmaker option, each government actor receives signals from the agents within its district, as well as media signals that are globally available. in this case, the regulated level of nutrient application is particular to each government actor and their district. farmers’ nutrient application is then governed by the regulation set for their district. parameterization, validation, and verification where possible, we used real-world data to parameterize the model (see appendix 2 for detailed descriptions of variables and sources). this provided some grounding in reality for factors like population size, but data were not available for many variables. for most of the variables in the decision process, we set parameters to reflect the relative importance of cyanos to the groups in question, using best-guesses based on our review of the literature (above). we then hand-tuned many of the variables to create our working model. because this is a stylized model, we were not able to validate against real-world behaviors. model verification was done through systematic evaluation of coding logic and model behavior (yilmaz, 2006). we started with bloom-related procedures, checking that nutrient application was occuring at expected rates, that the nutrient decay function was working properly, and that blooms were occuring as expected in the absense of regulation. at unregulated levels of nutrient application, even our simple model produced blooms that are physically similar to actual blooms observed in lake erie in recent years, though bloom duration tends to be longer, lasting for multiple ticks or “weeks”. we then tracked the relationships between bloom activity and the signals received by different agent sets and finally verfied that the regulation algorithm was working correctly and that farmer agents were applying only the regulated level of nutrients when regulations were in place. we found and eliminated multiple artifacts in this process. results model results generally support the expectations laid out in the introduction. response occurred more quickly in single-regulator mode than in multiple-regulator mode. when there was a single regulator aggregating signals from the entire area, regulation started after only 1-2 major bloom events and regulated nutrient levels quickly stepped down to scientifically recommended levels as new regulations were enacted each season. this, in turn, reduced bloom activity and represents a transition to an effective management cycle. in model runs with multiple decision makers, the disconnect between farmers and other actors in “region a” prevented a transition to more effective governance like the one shown in figure 2. instead, while all other regions quickly implemented regulations to reduce nutrient levels, the government agent in charge of “region a” was never sufficiently motivated to take action. because most of the nutrient pollution in the modeled lake comes from “region a”, the end result was that large bloom events continued much as in the unregulated version of the model. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 41 university of bamberg press no regulation figure 5 shows typical model results over a period of 1000 ticks (weeks). note the seasonal fluctuation in bloom activity, which can only occur when temperatures are above a certain threshold (summer months). at first, bloom activity is relatively low because there are not a lot of nutrients built up in the field patches but toward the end of the third “summer” the size of the bloom starts to increase. large bloom events like the one shown in figure 6 become common without regulation, though there is still a large degree of interannual variability in bloom size. this matches what is observed in the real world, though the mechanism is different; in reality large blooms tend to be correlated with major rainfall events but in our case large blooms reflect the build-up of nutrients in the soil and random/contagion effects in the bloom algorithm. figure 6 results with no regulation figure 5 also shows how various signals from the bloom are received by different agents. as expected, the media is tracking the overall size of the bloom (large bloom), and the proportion of blooming patches near major urban areas (toledo, sandusky), beaches, and water intakes. this information is distributed to other agents and is part of their concern function. the two charts on the bottom right track the level of concern for a randomly selected business owner (top) and farmer http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 42 university of bamberg press (bottom). note that the ranges are quite different for these agents but that these numbers are normalized and weighted by population size in the graph labeled average concern. the last graph shows the proportion of the agent set that is motivated to “take action” (send a signal to their decision maker) at any given point in time. urban dwellers and coastal business owners are the most motivated while farmers are the least motivated, as would be expected. figure 7 large bloom event with no regulation individual (national-level) regulator figure 7 shows the results with the same configuration as above but with an individual regulator collecting signals from all agents in the region and then regulating the level of nutrients applied by farmers. in this run a large bloom randomly occurs in the second “season”, triggering considerable concern by urban agents because it is near water intakes. this triggers the first stage of governance, setting the allowed level of fertilizer per application to the maximum recommended by scientists plus the mean difference between the current level and the scientifically recommended level. this reflects the responsive or trial-and-error nature of governance as well as the often observed desire of policy makers to cushion the impacts of regulations on business via gradual implementation. however, because this reduction is not sufficient, the government agent is forced to regulate again, gradually reducing the amount of fertilizer allowed per application until it is close to the minimum http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 43 university of bamberg press level recommended by scientists. nutrients applied go down with the regulations and eventually blooms are eliminated from the lake. figure 8 results with single government agent http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 44 university of bamberg press multiple (5 regional) regulators when we allow the five regional regulators to operate independently, the results change drastically, even though all other parameters remain the same. note that even though we ran the simulation for much longer, blooms continued to be a problem, much as in the unregulated case. the reason for this can be seen in the chart showing the “regulated nutrient level” by region. because they have no farmers, even though they are least directly affected by blooms, regions e and c (north-east and south-east) quickly set nutrient application limits at the minimum level recommended by scientists. however, as this had no effect on the flow of nutrients into the lake it does not reduce bloom activity. regions b and d (central north and south) do have some farmers, but signals from media, urban dwellers, and coastal businesses pushing for regulation outweighed signals from farmers against regulation. thus, these regions followed a gradual decline similar to that of the single-regulator described above. interestingly, region b, which has fewer coastal businesses took longer to enact minimum regulations, which would be expected based on relative levels of concern for these agents. here again, too few farmers are actually regulated to make much difference. finally, area a, (western basin) is dominated by farmer agents and so regulators never required any reductions in fertilizer rates. since this is where most of the farmers are located, nutrient pollution remained high and blooms remained a problem. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 45 university of bamberg press figure 9 results from the multi-regulator model http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 46 university of bamberg press discussion and conclusions the two scenarios we tested with our models (single regulator, and multiple regulators) show how the governance treadmill can switch to an effective cycle fairly quickly when disconnects are narrow (single regulator) but get stuck in ineffective cycles when disconnects are wide (multiple regulators). in the single regulator scenario, the signals from those agents that were most concerned about the bloom events (urban dwellers, coastal businesses) outweighed the signals from agents that were more concerned about the costs of regulation (farmers) and so response was relatively quick. in the multiregion scenario, the region that contributed most to the problem (a) was also politically dominated by agents that were most concerned about the costs of regulation and therefore the decision maker for that region never chose to regulate and blooms continued, even though other regions with narrower disconnects did choose to enact strict regulations after a time. while these results are interesting, they are not unexpected. the real value of this model is as a means to decompose the various elements of the governance treadmill and to plan for the development of more advanced models in the future. for instance, while we see seemingly random cycles of policy concern in the results, there is no positive underlying trend in policy attention (approximated by concern in the model) as shown in figure 2. this can be attributed to several elements of the model specification. most important, agents in the model had fairly short memories, so concern did not last much beyond the bloom events themselves. this may be indicative of public attention spans, but we know that some interest groups retain concern for much longer and could even evince step-wise increases in concern. on the other hand, people may also become desensitized to bloom events, such that susceptibility is actually decreasing with exposure to blooms. much more research is needed to be able to appropriately specify and parameterize these components of the model. in particular, longitudinal data on how people’s attitudes and willingness to participate in political processes or other response options with repeated exposure to extreme events is needed. the trend in concern was also relatively flat because of decisions that we made in order to maintain the transparency of the model. for instance, while there was some randomness in the application of fertilizer, its effect was minimal compared to either the farmer agent’s pre-set level of nutrient application or the allowed levels once regulations were in place. furthermore, we did not model any legacy nutrients in lake sediments, ensuring that after only a few years, bloom activity would be fairly stable at a level that could be generally predicted based on nutrient application rates. while this made it much easier to tell what was going on in the model, it assumed away a number of factors that would make bloom activity—and therefore agent concern—follow an increasing trend instead of the flat plateaus observed in our results. for instance, in the absence of regulation nutrient application rates have been increasing over time, rather than remaining constant, and even when regulations are in place their effects on nutrient levels are not uniformly effective. furthermore, nutrient storage in lake sediments can also lead to an increase in the severity of blooms and resulting concern even after regulations lead to lower inflows of nutrients (wilson et al., 2019). thus, instead of being a real result of the governance treadmill, the step-wise changes in regulation shown in the models where power disconnects are narrow are really an artifact of our programming. in other words, while we captured spikes in concern and resultant changes in the need to regulate, as predicted by the treadmill, the form of regulatory response was dictated by model specifications. this highlights the needs for additional data on representation and influence as well as better modeling of the policy process itself. a first step would be to allow for de-regulation in the current model. this would at least let us examine the rebound effect and multiple full cycles of the treadmill. more realistic representations could also include an array of possible policies, each of which has different cost and benefits for different sets of agents. then constituent agents could communicate not just their concern but also their support for particular regulatory options. the model could also be improved with more accurate representations of decision making, including periods in which decision http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 47 university of bamberg press makers could communicate or negotiate with each other and then make decisions using voting or some other rules. indeed, there are many avenues for enquiry through refinement of this simple model. we designed it specifically to serve as the foundation of a suite of experiments around the governance treadmill, starting with the lake erie case study but eventually generalizable to other regions and types of issues (janssen & ostrom, 2006). by keeping most of components relatively simple, it will be easier to observe the effects of changes or experiments in sections of the model where there are high levels of variation or uncertainty, like public attention or interest group influence. of course, model results must be interpreted accordingly, and it will still be difficult to trace causal connections as the model becomes more complex, but a systematic exploration of the governance treadmill through these models is still worthwhile. extreme events like harmful microbial blooms are expected to surge with population growth, increasing affluence, and climate change, so we need a better understanding of the way such events will affect environmental governance of social ecological systems. acknowledgements this work was supported by the national socio-environmental synthesis center (sesync) under funding received from the national science foundation dbi1052875. we greatly appreciate the suggestions and advice of our colleagues from these workshops. references baker, d. b., johnson, l. t., confesor, r. b., crumrine, j. p., guo, t., & manning, n. f. (2019). needed: early-term adjustments for lake erie phosphorus target loads to address western basin cyanobacterial blooms. journal of great lakes research, 45(2), 203–211. http://doi.org/10.1016/j.jglr.2019.01.011 baumgartner, f. r., & jones, b. d. (2009). agendas and instability in american politics (2nd ed.). chicago: university of chicago press. bingham, m., sinha, s. k., & lupi, f. (2015). economic benefits of reducing harmful algal blooms in lake erie. report 66. bullerjahn, g. s., mckay, r. m., davis, t. w., baker, d. b., boyer, g. l., anglada, l. v. d., … wilhelm, s. w. (2016). global solutions to regional problems : collecting global expertise to address the problem of harmful cyanobacterial blooms . a lake erie case study. harmful algae, 54, 223–238. http://doi.org/10.1016/j.hal.2016.01.003 higgs, r. (1987). crisis and leviathan: critical episodes in the growth of an american government. new york: oxford university press. hirschman, a. o. (1993). exit, voice, and the fate of the german democratic republic: an essay in conceptual history. world politics, 45(2), 173–202. janssen, m. a., & ostrom, e. (2006). empirically based, agent-based models. ecology and society, 11(2), 37. jones, b. d., & baumgartner, f. r. (2005a). a model of choice for public policy. journal of public administration research and theory, 15(3), 325–351. http://doi.org/10.1093/jopart/mui018 jones, b. d., & baumgartner, f. r. (2005b). the politics of attention: how government prioritizes problems. chicago: university of chicago press. keller, c., siegrist, m., & gutscher, h. (2006). the role of the affect and availability heuristics in risk communication. risk analysis, 26(3), 631–639. http://doi.org/10.1111/j.1539-6924.2006.00773.x kingdon, j. w. (2011). agendas, alternatives, and public policies (3rd ed.). boston: longman. kuran, t. (1995). private truths, public lies: the social consequences of preference falsification. cambridge, ma: harvard university press. march, j. g., & simon, h. a. (1993). organizations (2nd ed.). cambridge: blackwell business publishers. neff, m. r., bhavsar, s. p., ni, f. j., carpenter, d. o., drouillard, k., fisk, a. t., & arts, m. t. (2014). risk-benefit of consuming lake erie fish. environmental research, 134, 57–65. http://doi.org/10.1016/j.envres.2014.05.025 paerl, h. w., & otten, t. g. (2013). harmful cyanobacterial blooms: causes, consequences, and controls. microbial ecology, 65(4), 995–1010. http://doi.org/10.1007/s00248-012-0159-y pearson, g. (2014, august). don’t drink the water … in lake erie. retrieved may 9, 2016, from http://www.wired.com/2014/08/lake-erie-algae/ http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 48 university of bamberg press simon, h. a., & barnard, c. i. (1957). administrative behavior; a study of decision-making processes in administrative organization (2nd ed.). new york: macmillan. slovic, p. (1987). perception of risk. science (new york, n.y.). http://doi.org/10.1126/science.3563507 walker, b., & salt, d. (2006). resilience thinking: sustaining ecosystems and people in a changing world. washington dc: island press. walters, c. j., & hilborn, r. (1978). ecological optimization and adaptive management. annual review of ecology and systematics, 8(1978), 157–188. webster, d. g. (2015a). beyond the tragedy in global fisheries. cambrdge: mit press. webster, d.g. (2015b). the action cycle/structural context framework: a fisheries application, 20(1), 33. http://doi.org/10.5751/es-07272-200133 wilson, r. s., beetstra, m. a., reutter, j. m., hesse, g., fussell, k. m. d. v., johnson, l. t., … winslow, c. (2019). commentary: achieving phosphorus reduction targets for lake erie. journal of great lakes research, 45(1), 4–11. http://doi.org/10.1016/j.jglr.2018.11.004 wilson, r. s., howard, g., & burnett, e. a. (2014). improving nutrient management practices in agriculture: the role of riskbased beliefs in understanding farmers’ attitudes toward taking additional action. water resources research, 50, 6735– 6746. http://doi.org/10.1002/2013wr015200.received wines, m. (2014, august). behind toledo’s water crisis, a long-troubled lake erie. the new york times. retrieved from http://www.nytimes.com/2014/08/05/us/lifting-ban-toledo-says-its-water-is-safe-to-drink-again.html yilmaz, l. (2006). validation and verification of social processes within agent-based computational organization models. computational and mathematical organization theory, 12(4), 283–312. http://doi.org/10.1007/s10588-006-8873-y appendix 1 overview, design concepts, and details documentation overview this model was created to help explore the ways citizens perceive and respond to the risks associated with harmful algal blooms (habs), the way government officials respond to their constituents, and the policy and environmental outcomes of these processes. to simulate this action-response dynamic, we first setup a lake comprised of patches that can become colonized by harmful algae or “bloom.” the processes used to create the simulated bloom do not represent actual biophysical processes, but they do produce blooms that display common characteristics of real blooms (random seeding, threshold effects, contiguous spread). agents live in the land area surrounding the lake. there are several types of agents representing different types of people with different interests and incentives. each agent can perceive blooms in the lake through various signals. they can also become concerned about the blooms if they receive signals that the bloom will affect them negatively. if enough concern accumulates, the agents can take action by engaging with policy makers. policy makers, known as “government actors” in the model, receive signals from the media, scientists, and people who are concerned enough to contact the government about algal blooms. these signals create concern in the policy makers, and they can decide to regulate nutrient use by farmers if the government actors become concerned enough. setup global variables can be used by any patch or agent in the model. in this model, we created global variables for two reasons: to allow more than one agent type to use the variables’ values to run a procedure, or to give the model user the ability to change variable values through sliders on the user http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 49 university of bamberg press interface. the only global variables fixed in the setup code are related to lake and bloom variables; these are the nutrients in the lake ‘nutrients.lake’, the total nutrients on all farms ‘nutrients.farms’, a tracker of warm and cold seasons ‘temperature’, a tracker of precipitation ‘rain’, the number of patches that are farms ‘farm patches’, and the proportion of each of three types of farmers represented in the model ‘proportion.farmers.x’, ‘proportion.farmers.y’, and ‘proportion.farmers.z’. all other global variables are controlled by sliders on the user interface. these variables relate to how different agents perceive and are affected by different signals. they include the magnitude of a signal, how susceptible an agent is to that signal, how long they care about that type of signal for, how quickly an agent loses concern, and at what level of concern an agent takes action. these variables are described below in the section on the calculation of conern. next, seven different breeds of agents are created: ‘urban.dwellers’, ‘rural.non.farmers’, ‘coastal.busines.owners’, ‘farmers’, ‘government.actors’, ‘media.sources’, and ‘scientists’. the variables owned by patches and each agent type are setup next. these variables are either characteristics of the agents or elements of the world that are specific to individual agents and change over time. patches own a variable that corresponds to the map assigning patch type ‘land.cover.code’; the type of patch set by the code ‘land.cover.type’; ‘city.name’ which is left blank if the patch is not part of a city; binary variables that equal one if true named ‘major.city’, ‘minor.city?’, ‘farm?’, and ‘intake?’; the watershed to which the patch belongs ‘watershed’; represented human population ‘population’; the proportion of different agent types at each patch ‘prop.farmer’, ‘prop.urban’, ‘prop.rural.non.farmer’, ‘prop.tourist’, and ‘prop.coastal.business’; the nutrients at that point in time ‘nutrients.here’; the number of tourists at that point in time ‘tourists.now’; whether the patch has an algal bloom happening, which is only applicable for lake patches ‘blooming?’, and the number of time steps the patch has been blooming for ‘time.blooming’; and the district to which the patch belongs ‘pdistrict’. all agents or “turtles” own a few common variables; these are setup next. all agents have a ‘district’, a number of time steps they are affected by media signals ‘attention.span.media’, ‘conern’, and a threshold of concern over which they take action ‘concern.threshold’. agents also own variables specific to their type. these are defined next in the model, beginning with urban dwellers. agents keep record of signals they have received over the course of their attention span, coded as ‘signal.history.type’ where type is replaced with the type of signal. for instance, urban dwellers have four signal history variables: ‘signal.history. intake’, ‘signal.history.seeing’, ‘signal.history.tourism’, and ‘signal.history.media’. agents have variables that represent the amount of concern they have from each type of signal, coded as ‘concern.type’. for urban dwellers, these are ‘concern.intake’, ‘concern.seeing’, ‘concern.tourism’, and ‘concern.media’. the four agent types that represent constituencies, i.e. urban dwellers, rural non-farmers, coastal business owners, and farmers, have a variable that denotes whether their concern threshold has been crossed and if they are triggered to contact the government ‘contact.government?’. rural non-farmers have concern-related variables for media signals. coastal business owners have concern variables for intake, seeing, tourism, and media signals. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 50 university of bamberg press farmers have concern-related variables for regulation and media signals. they also have a variable denoting the type of farmer ‘farmer.type’, and the nutrients they applied to their fields at that point in time ‘nutrients.applied’. government actors own signal-history and concern variables for different signal types, with these signals coming from their constituents. these variables, ‘signal.history.type’ and ‘conern.type’, replace type with ‘ud’ for urban dwellers, ‘rnf’ for rural non-farmers, ‘cbo’ for coastal business owners, and ‘farmers.x’, ‘farmers.y’, and ‘farmers.z’ for farmers of type x, y, and z. government actors own a variable denoting whether they are the top decision maker, ‘top.decision.maker?’. they also own variables related to regulating nutrient application by farmers. whether they regulate nutrients at that point in time is tracked by the binary variable ‘regulate.nutrients?; the maximum amount of nutrients farmers can apply at any time step is ‘nutrients.regulated’; and if the conditions are met for government actors to regulate nutrients that year, the variable ‘need.to.regulate’ is set to one, and zero otherwise. media sources own variables that represent media coverage of different types of signals: ‘signal.large.bloom’, ‘signal.toledo’, ‘signal.sandusky’, ‘signal.intakes’, and ‘signal.beaches’. scientists own two variables that convey the scientific recommendations for the lower limit of nutrients famers should be allowed to apply, ‘nutrients.min.recommended’, and the upper limit ‘nutrients.max.recommended’. next, we setup the world. first, all information from previous runs is cleared. next, we set the size of the world to a 100 by 100 grid. we set the patch size to 7 so that the map is a usable size in the user interface. then, we set up land cover codes for the world and create beaches, major cities, minor cities, farms, and water intakes. the location of large cities – toledo, cleveland, erie, buffalo, detroit, and sandusky – and water intakes corresponds to their locations in the real world. beaches are distributed at intentional intervals along the coast to represent their approximate distribution in the real world. minor cities and farms are distributed randomly over the land. we then define watersheds and set patch colors according to their cover type. finally, we give patches population characteristics according to their land cover type and presence or absence of a city. these characteristics include population size and the proportions of urban dwellers, coastal business owners, farmers, and tourists. finally, we create seven different types of agents: urban dwellers, rural non-farmers, coastal business owners, three subtypes of farmers (“x”, “y”, and “z”), and government actors. in this version of the model (8 may 2018), farmers are considered a single group. these agents populate the modeled landscape in approximately the same way they populate the real-life landscape. however, the numbers of each agent type in the model does not represent the number of each type of people in real life. urban dwellers represent people living in major cities and small cities. rural non-farmers represent people who live in rural areas but do not farm. coastal business owners represent people who live near coastal tourist attractions, namely beaches. in the model, they occupy land patches with at least one neighboring patch that is a beach. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 51 university of bamberg press there is potential to add different types of farmers, but this has not yet been implemented. the three types of farmers represent categorizations made by robyn wilson from ohio state university, based on farmer surveys conducted in the maumee river basin, which is the main source of nutrient pollution in erie. they are currently conducting a latent variable analysis to generate up to three categories of farmers that will be distinguished by their age, income, education, and, most important, attitudes toward habs risk and nutrient-reducing policies. once this is finished we will use it to develop the farmer types. similar analysis is also planned from a public survey in the erie region that should help parameterize the other agent types. there are two versions of how the government operates: via single or multiple decision makers. in the single-decision-maker version, a single government actor receives signals and creates nutrientuse policy. in the multiple-decision-maker version, the landscape is divided into five districts, each with its own assigned government actor. here, policy is made in response to the concerns of constituents within each district, and that policy applies only to farmers within that district. more details about the differences between government processes are described below. processes the processes in the model proceed in the following order: 1) track time and seasons, 2) farming, 3) algal blooms, 4) track tourists visiting beaches, 5) generate media signals, and 5) calculate the level of concern of and actions taken by urban dwellers, rural non-farmers, coastal business owners, farmers, and the government. each agent type has a procedure for calculating concern and whether they take action). bloom-related procedures blooms only occur during warm months, roughly weeks 22 through 43 of the calendar year, which we designate with ‘temperature’ = “high” in the model. for all other months, ‘temperature’ = “low”. farming contains the decision of farmers about how much nutrients they apply while farming. this is relevant to the simulation in two ways: it is the source of nutrient pollution to the lake, and nutrient additions can be regulated by policy. farmers apply nutrients to their fields during the growing season, between calendar weeks 18 and 42 to represent the ohio farming season. farmers apply 10 units of nutrients under normal conditions, and lower amounts if the government is regulating nutrient application (see description of the nutrient regulation procedure below). we chose 10 units of nutrients as a starting point arbitrarily. farmers have a certain probability of applying nutrients at each time step, controlled by the variable ‘fertilizer.application.rate’, that ranges from 0 to 1. the default value for the fertilizer application rate is 0.5, i.e., every other week on average, which is also arbitrary. there are two possible scenarios for government structure and, therefore, regulation of nutrients. if there is a single government decision maker, the top decision maker sets the regulated levels for farmers in all districts. if there are multiple government decision makers, the decision maker from each district sets the regulated level for farmers in their district only. in this case, farmers from the different districts could all have different regulated levels. the procedure for farming proceeds as follows. first, we temporarily set the beginning and ending weeks of the farming season and the rate at which nutrients leave the patch where they were applied. next, farmers determine whether there are multiple or a single decision maker, which is based on a switch in the user-interface. the answer to this question determines which government actor farmers look to for the regulated nutrient level. then, if farmers are located in the western basin, representing the maumee river basin, and the week is within the growing season and a random draw between 0 and 1 is less than the http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 52 university of bamberg press fertilizer application rate, the farmer sets their ‘nutrients.applied’ variable to the ‘nutrients.regulated’ level of the appropriate government actor. if all three of those previously mentioned conditions are not met, farmers set ‘nutrients.applied’ to 0. patches calculate the amount of nutrients they have in the next step of the farming procedure. patches gain nutrients from the application by farmers, and they lose nutrients to natural processes we encapsulate with the term ‘nutrient.decay’ and by runoff from rainfall. if the farmer or farmers on a patch applied nutrients that week, the patch’s ‘nutrients.here’ variable is set according to the following formula: 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. ℎ𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡+1 = 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. ℎ𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡 − 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡. 𝑑𝑑𝑛𝑛𝑐𝑐𝑝𝑝𝑏𝑏 ∗ 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. ℎ𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡 − 𝑡𝑡𝑝𝑝𝑡𝑡𝑝𝑝𝑟𝑟𝑝𝑝𝑝𝑝𝑝𝑝. 𝑟𝑟𝑝𝑝𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 ∗ 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. ℎ𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡 + � 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑀𝑀𝐴𝐴𝐴𝐴𝑅𝑅𝐴𝐴𝑅𝑅𝑅𝑅 𝑓𝑓𝑠𝑠𝑡𝑡𝑓𝑓𝑛𝑛𝑡𝑡𝑡𝑡 if the farmer or farmers on a patch do not apply nutrients on a patch that week, the patch’s nutrients are set according to the following formula:𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. ℎ𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡+1 = 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. ℎ𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡 − 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡. 𝑑𝑑𝑛𝑛𝑐𝑐𝑝𝑝𝑏𝑏 ∗ 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. ℎ𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡 − 𝑡𝑡𝑝𝑝𝑡𝑡𝑝𝑝𝑟𝑟𝑝𝑝𝑝𝑝𝑝𝑝. 𝑟𝑟𝑝𝑝𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 ∗ 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. ℎ𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡 the procedure for producing harmful algae blooms has several components. first, two conditions have to be met: the nutrients entering the lake, ‘nutrients.farms * rainfall.factor’, must exceed the minimum threshold for a bloom to occur, and the temperature must be “high”. if so, and there isn’t a bloom already occurring, the western end of the lake where the x-coordinates are less than 55 has ‘bloom.seeds’ number of patches turned on by setting ‘blooming?’ = 1. the ‘time.blooming’ variable of the newly blooming patch is set to 1 as well. next, the bloom spreads to neighboring patches. a number, set by ‘num.neighbors.bloom.spread’, of lake patches also bloom. if the bloom occurs along the coast and there are fewer lake patches available to bloom than ‘num.neighbors.bloom.spread’, all of the neighboring lake patches bloom. within this sub-procedure, each patches’ ‘time.blooming’ variable advances by 1. any patches with ‘time.blooming >= max.time.blooming’ return to their regular state, i.e., ‘blooming? = 0’. if the conditions for a bloom are not met, i.e. ‘nutrients.farms * rainfall.factor <= nutrient.threshold’, or temperature is “low”, any blooming patches are returned to their regular state. coastal tourism declines if blooms occur near public beaches. if the season is warm and none of the lake patches neighboring a beach are blooming, the number of tourists is calculated using the following equation: 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = 𝑝𝑝𝑡𝑡𝑝𝑝𝑡𝑡𝑝𝑝𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑝𝑝 ∗ 𝑝𝑝𝑡𝑡𝑡𝑡𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 where ‘proptourists’ is the proportion of people at a patch that are normally tourists. if any of a beach’s neighboring lake patches are blooming, tourists are counted with the following equation: 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = 𝑝𝑝𝑡𝑡𝑝𝑝𝑡𝑡𝑝𝑝𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑝𝑝 ∗ 𝑝𝑝𝑡𝑡𝑡𝑡𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 − 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡. 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡. 𝑏𝑏𝑏𝑏. 𝑏𝑏𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡 ∗ 𝑝𝑝𝑡𝑡𝑝𝑝𝑡𝑡𝑝𝑝𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑝𝑝 ∗ 𝑝𝑝𝑡𝑡𝑡𝑡𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 53 university of bamberg press where ‘tourism.loss.by.bloom’ is the proportion of tourists who no longer visit beaches when there is a nearby bloom. media signals are calculated next in the netlogo script. we describe them within the “signals” section below. next, we calculate scientific advice on setting regulated nutrients levels. we set the rate at which nutrients leave the landscape and do not enter the lake at 0.1. the highest level of nutrient application scientists recommend is calculated according to the following equation: 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑀𝑀𝑀𝑀𝑀𝑀. 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑀𝑀𝑀𝑀𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 = 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡. 𝑡𝑡ℎ𝑡𝑡𝑛𝑛𝑡𝑡ℎ𝑡𝑡𝑝𝑝𝑑𝑑 ∗ (1 − 𝑡𝑡𝑝𝑝𝑡𝑡𝑝𝑝𝑟𝑟𝑝𝑝𝑝𝑝𝑝𝑝. 𝑟𝑟𝑝𝑝𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 + 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡. 𝑑𝑑𝑛𝑛𝑐𝑐𝑝𝑝𝑏𝑏)−1 ∗ 𝑝𝑝𝑡𝑡𝑡𝑡𝑏𝑏𝑛𝑛𝑡𝑡 𝑡𝑡𝑟𝑟 𝑟𝑟𝑝𝑝𝑡𝑡𝑡𝑡𝑛𝑛𝑡𝑡𝑡𝑡−1 ∗ 𝑟𝑟𝑛𝑛𝑡𝑡𝑡𝑡𝑡𝑡𝑝𝑝𝑡𝑡𝑓𝑓𝑛𝑛𝑡𝑡. 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑡𝑡𝑐𝑐𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑝𝑝. 𝑡𝑡𝑝𝑝𝑡𝑡𝑛𝑛−1 the lowest level of nutrient application scientists recommend differs based on the amount of rainfall, according to the following logic: if ‘rainfall.factor’ > 0.4, ‘nutrients.min.recommended = nutrints.max.recommended – 1’; if ‘rainfall.factor’ <=0.4, ‘nutrients.min.recommended = nutrients.max.recommended – 2’. concern ‘concern’ is a variable held by each agent that calculates and stores their level of concern regarding the algal blooms. concern at each time step is generated from the multiple signals agents receive according to the following equation: 𝑐𝑐𝑡𝑡𝑝𝑝𝑐𝑐𝑛𝑛𝑡𝑡𝑝𝑝 = � 𝑡𝑡𝑡𝑡𝑡𝑡𝑐𝑐𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡𝑏𝑏𝑡𝑡𝑝𝑝𝑡𝑡𝑡𝑡𝑏𝑏𝑡𝑡𝑡𝑡𝑠𝑠𝑛𝑛𝑠𝑠𝑜𝑜.𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛 ∗ 𝑡𝑡𝑡𝑡𝑠𝑠𝑝𝑝𝑝𝑝𝑝𝑝. 𝑣𝑣𝑝𝑝𝑝𝑝𝑡𝑡𝑛𝑛𝑡𝑡𝑡𝑡𝑠𝑠𝑛𝑛𝑠𝑠𝑜𝑜.𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛 𝑡𝑡𝑡𝑡𝑠𝑠𝑛𝑛𝑠𝑠𝑜𝑜.𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑡𝑡 urban dwellers, rural non-farmers, and coastal business owners have only positive levels of concern, representing how much they are worried about undesirable effects of the bloom. farmers concern can be positive or negative. positive concern in farmers represents concern that their farming business will be negatively impacted by nutrient regulation, and negative concern represents concern over undesirable effects from the bloom. the ‘susceptibilitysignal.type’ variable accounts for differences in how much people care about different aspects of the bloom. for example, farmers might not care if coastal tourism decreases, whereas coastal business owners care a lot. because all signals are between 0 and 1, with 1 being full signal strength, the susceptibility variables determine the maximum amount of concern a signal can generate. the susceptibility of government actors to a constituency contacting them is the maximum amount of concern that constituency can create in the government actor. note that this variable accounts for multiple factors such as number of constituents of each type and the political power they wield as a group. this variable is named with the following format in the code: ‘susceptibility.agent.type.government’. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 54 university of bamberg press an example is ‘susceptibility.farmers.x.government’ for government susceptibility to farmers of type x. a susceptibility of 1 would mean that if all farmers, for example, contacted the government, the government actor would become maximally concerned, i.e. ‘concern’ = 1. if urban dwellers were also concerned and contacted the government, no additional concern would accumulate, because there would be no ability to become more concerned. multiplying the signal value by susceptibility implies that each additional unit of signal creates the same amount of concern. more complex models could be used if there were theory or evidence to support them. concern decays over time linearly. each agent type has a different rate at which concern decreases, referred to as ‘decay.of.agent.type’. 𝑐𝑐𝑡𝑡𝑝𝑝𝑐𝑐𝑛𝑛𝑡𝑡𝑝𝑝𝑡𝑡 = 𝑑𝑑𝑛𝑛𝑐𝑐𝑝𝑝𝑏𝑏 ∗ 𝑐𝑐𝑡𝑡𝑝𝑝𝑐𝑐𝑛𝑛𝑡𝑡𝑝𝑝𝑡𝑡−1 + 𝑐𝑐𝑡𝑡𝑝𝑝𝑐𝑐𝑛𝑛𝑡𝑡𝑝𝑝𝑛𝑛𝑛𝑛𝑛𝑛 concern over each signal type disappears completely if enough time passes without a new signal; the time threshold is called ‘attention.span.signal.type’. in the model, a list is created for each signal type that keeps track of whether there was a signal at each time step for as long as the agent’s attention span for each signal type. if none of the lists contain any signals greater than one, no signals have occurred within the agent’s attention span and their concern goes to zero. urban dwellers’ concern comes from four sources: contaminated water intakes, blooms that occur close enough to the shoreline of their cities for people to see, decreases in coastal tourism, and media signals on blooms. rural non-farmers’ concern comes from media signals only. this assumes they are not directly impacted by the blooms, but may care about the health of the lake or other people who are impacted. coastal business owners’ concern comes from three sources: blooms that occur close enough to the shoreline for people to see, decreases in tourism in adjacent beaches caused by nearby blooms, and media signals on blooms. farmers’ concern comes from two sources: regulation of nutrient application and media signals. media signals generate concern in two ways; farmers have concern for the health of the lake and that they may face further regulations that will hurt their business. signals signal values, which feed into the calculation of agents’ concern, range from 0 to 1, with some variables being binary, i.e. either 0 or 1. the ‘signal.valuesignal.type’ variable allows for variability in the strength of signals at different points in time. if any patches that has a water intake on it and is within a radius of four from an agent, the signal value for intakes equals one. otherwise, this signal equals 0. if any neighboring patches are blooming, the signal value for seeing the bloom equals 1. otherwise, this signal equals 0. http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 55 university of bamberg press if it is the hot season, tourists are counted at coastal beaches. if there is a bloom in any of a beach’s neighboring patches, tourism declines by the proportion ‘tourism.loss.by.bloom’. the signal value for tourism loss is one minus the proportion of tourists currently compared to the number of that would be there had there not been a nearby bloom. if government actors are currently regulating nutrients, the signal value for regulation is 1. otherwise, this signal equals 0. media signals are calculated for five phenomena: if the bloom is 1) large (patches blooming > ‘large.bloom.threshold’); 2) near toledo (patches blooming and neighbor to toledo’s coastal patch); 3) near sandusky (patches blooming and neighbor to sandusky’s coastal patch); 4) at a drinking water intake; or 5) adjacent to any beach. the value or strength of each media signal is the proportion of patches blooming that meet the given condition relative to the number that could possibly meet the conditions. all agents have access to media signals because they are global variables. for government actors, signals from each type of constituent are calculated as the proportion of constituents with ‘contact.government?’ = 1. if every farmer was concerned enough to contact the government, the government would receive a signal of 1 from farmers. action agents take action if their level of concern reaches a threshold. taking action is binary; an agent can either take action or not. urban dwellers, rural non-farmers, coastal business owners, and farmers can take action by contacting the government (‘contact.government’ = 1). they do so when their concern exceeds a threshold, ‘concern.threshold.agent.type’. government actors take action by regulating nutrient application by farmers. at each time step, government actors either leave the amount of nutrients that can be applied the same or change it. the decision making process of government actors begins with the calculation of concern generated by signals from urban dwellers, rural non-farmers, coastal business owners, farmers, the media, and scientific advice. if these signals generate enough concern to cross government actors’ concern threshold, the government actors decide to regulate nutrients. the level of regulation set depends on a series of conditions. if the nutrients applied by farmers at that time step is greater than the amount recommended by scientists, 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑅𝑅𝑅𝑅𝐺𝐺𝑅𝑅𝑅𝑅𝑀𝑀𝑁𝑁𝑅𝑅𝑅𝑅 = 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑀𝑀𝑀𝑀𝑀𝑀. 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑀𝑀𝑀𝑀𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 + 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑡𝑡𝑛𝑛𝑑𝑑 − 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑀𝑀𝑀𝑀𝑀𝑀. 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑀𝑀𝑀𝑀𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 2 if nutrients applied by farmers is greater than the max amount recommended but less than or equal to the max amount recommended plus one, 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑅𝑅𝑅𝑅𝐺𝐺𝑅𝑅𝑅𝑅𝑀𝑀𝑁𝑁𝑅𝑅𝑅𝑅 = 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑀𝑀𝑀𝑀𝑀𝑀. 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑀𝑀𝑀𝑀𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 if nutrients applied is greater than the minimum amount recommended but less than the max amount recommended, http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 56 university of bamberg press 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑅𝑅𝑅𝑅𝐺𝐺𝑅𝑅𝑅𝑅𝑀𝑀𝑁𝑁𝑅𝑅𝑅𝑅 = 𝑝𝑝𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑛𝑛𝑝𝑝𝑡𝑡𝑡𝑡. 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑡𝑡𝑛𝑛𝑑𝑑 − 1 if nutrients applied is less than the minimum amount recommended, nutrients are not regulated. there are two options for the process of how government actors make decisions: either as a single decision maker for the entire watershed, or with five decision makers, each representing a defined geographic area. if there is a single decision maker, the concerns of all stakeholders affect the government actor’s concern and, hence, level of nutrient regulation. under the multiple-decisionmaker option, each government actor receives signals from the agents within its district, as well as media signals that are globally available, but which may be more or less important in their district (e.g. signals that a bloom is near a significant city in their jurisdiction). in this case, the regulated level of nutrient application is particular to each government actor and their district. farmers’ nutrient application is then governed by the regulation set for their district. there is no “cheating” in the model. farmers always apply the regulated amount of nutrients, though the total amount of nutrients will sometimes go above the regulated amount due to the probabilistic process farmer agents use to decide when to apply nutrients. appendix 2 model variables, values, and sources variable in code estimated/base value data source land.cover.code na lake erie map (see web address in notes) land.cover.type na lake erie map city.name na estimated position from google earth major.city? 5 major cities estimated position from google earth minor.city? 25 minor cities guestimate farm? 1846 farms determined by remaining patches after other patch types assigned beach? 8 beaches estimated number of beaches from google earth intake? 4 intakes https://www.google.com/mymaps/viewer?mi d=18-6xxsaz10doxcwfazrp_uqnw8&hl=en_us watershed lake erie map population 721 agents generated from pixelation of lake erie watershed map http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 57 university of bamberg press prop.urban varies by patch type guestimate prop.coastal.business varies by patch type guestimate prop.farmer varies by patch type guestimate prop.tourist varies by patch type guestimate nutrient.decay 0.1 none nutrients.here max ~31000 na tourists.now max 12776 na time.blooming calculated na blooming? reporter guestimate bloom.seeds 1 na num.neighbors.bloom.spread 3 na max.time.blooming 5 na signal.value.intake reporter na signal.value.seeing reporter na signal.value.tourism 2 na signal.value.regulation 2 na signal.toledo reporter na signal.sandusky reporter na signal.large.bloom reporter na signal.intakes reporter na http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 58 university of bamberg press signal.beaches reporter na signal.history.intake calculated na signal.history.seeing calculated na signal.history.tourism calculated na signal.history.media calculated na signal.history.regulation calculated na signal.history.ud calculated na signal.history.rnf calculated na signal.history.cbo calculated na signal.history.farmers calculated na susceptibility.media.toledo 2 set to values that reflect the variability in the impact of signals on agents susceptibility.media.sandusky 2 set to values that reflect the variability in the impact of signals on agents susceptibility.media.large.bloom 2 set to values that reflect the variability in the impact of signals on agents susceptibility.media.intakes 2 set to values that reflect the variability in the impact of signals on agents susceptibility.media.beaches 2 set to values that reflect the variability in the impact of signals on agents susceptibility.seeing 2 set to values that reflect the variability in the impact of signals on agents susceptibility.tourism 2 set to values that reflect the variability in the impact of signals on agents susceptibility.ud 2 set to values that reflect the variability in the impact of signals on agents http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 59 university of bamberg press susceptibility.rnf 2 set to values that reflect the variability in the impact of signals on agents susceptibility.farmers 2 set to values that reflect the variability in the impact of signals on agents susceptibility.cbo 2 set to values that reflect the variability in the impact of signals on agents attention.span.media 5 set to reflect a reasonable attention span attention.span.intake 5 set to reflect a reasonable attention span attention.span.seeing 5 set to reflect a reasonable attention span attention.span.tourism 5 set to reflect a reasonable attention span attention.span.regulation 5 set to reflect a reasonable attention span concern 0 calculatd in model concern.media 0 calculatd in model concern.intake 0 calculatd in model concern.seeing 0 calculatd in model concern.tourism 0 calculatd in model concern.regulation 0 calculatd in model concern.ud 0 calculatd in model concern.rnf 0 calculatd in model concern.cbo 0 calculatd in model concern.farmers 0 calculatd in model concern.threshold 20 tuned decay.of.ud 0.9 estimated to produce a reasonable decay curve http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 60 university of bamberg press decay.of.rnf 0.9 estimated to produce a reasonable decay curve decay.of.cbo 0.9 estimated to produce a reasonable decay curve decay.of.farmers 0.9 estimated to produce a reasonable decay curve decay.of.government 0.9 estimated to produce a reasonable decay curve apply.nutrients na arbitrary contact.government? na 0 regulate.nutrients? na 0 http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 61 university of bamberg press appendix 3 initial settings for all three regulatory options (none, individual, multiple) regulatory options initial setting base concern base.concern.ud 0.1 base.concern.rnf 0.1 base.concern.cbo 0.1 base.concern.farmers.x 0.1 base.concern.farmers.y 0.1 base.concern.farmers.z 0.1 base.concern.government 0.2 susceptibility susceptibility.intake.ud 0.75 susceptibility.intake.cbo 0.75 susceptibility.seeing.ud 0.5 susceptibility.seeing.cbo 0.75 susceptibility.tourism.ud 0.5 susceptibility.tourism.cbo 0.75 susceptibility.regulation.farmers.x 0.75 susceptibility.regulation.farmers.y 0.75 susceptibility.regulation.farmers.z 0.69 susceptibility.ud.government 0.35 susceptibility.rnf.government 0.35 susceptibility.cbo.government 0.35 susceptibility.farmers.x.government 0.35 susceptibility.farmers.y.government 0.35 susceptibility.farmers.z.government 0.35 susceptibility.media.intakes.ud 0.75 susceptibility.media.intakes.rnf 0.75 susceptibility.media.intakes.cbo 0.75 http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 62 university of bamberg press susceptibility.media.intakes.farmers.x 0.25 susceptibility.media.intakes.farmers.y 0.75 susceptibility.media.intakes.farmers.z 0.75 susceptibility.media.intakes.government 1 susceptibility.media.toledo.ud 0.75 susceptibility.media.toledo.rnf 0.75 susceptibility.media.toledo.cbo 0.5 susceptibility.media.toledo.farmers.x 0.25 susceptibility.media.toledo.farmers.y 0.66 susceptibility.media.toledo.farmers.z 0.75 susceptibility.media.toledo.government 1 susceptibility.media.sandusky.ud 0.75 susceptibility.media.sandusky.rnf 0.75 susceptibility.media.sandusky.cbo 0.75 susceptibility.media.sandusky.farmers.x 0.25 susceptibility.media.sandusky.farmers.y 0.74 susceptibility.media.sandusky.farmers.z 0.71 susceptibility.media.sandusky.government 1 susceptibility.media.beaches.ud 0.5 susceptibility.media.beaches.rnf 0.75 susceptibility.media.beaches.cbo 0.75 susceptibility.media.beaches.farmers.x 0.25 susceptibility.media.beaches.farmers.y 0.75 susceptibility.media.beaches.farmers.z 0.75 susceptibility.media.beaches.government 1 susceptibility.media.large.bloom.ud 0.5 susceptibility.media.large.bloom.rnf 0.25 http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 63 university of bamberg press susceptibility.media.large.bloom.cbo 0.75 susceptibility.media.large.bloom.farmers.x 0.25 susceptibility.media.large.bloom.farmers.y 0.75 susceptibility.media.large.bloom.farmers.z 0.75 susceptibility.media.large.bloom.government 1 susceptibility.media.regulation.farmers.x 0.75 susceptibility.media.regulation.farmers.y 0.75 susceptibility.media.regulation.farmers.z 0.75 attention span attention.span.intake.ud 66 attention.span.intake.cbo 66 attention.span.seeing.ud 66 attention.span.seeing.cbo 66 attention.span.tourism.ud 66 attention.span.tourism.cbo 66 attention.span.regulation.farmers.x 63 attention.span.regulation.farmers.y 63 attention.span.regulation.farmers.z 66 attention.span.media.ud 66 attention.span.media.rnf 26 attention.span.media.cbo 66 attention.span.media.farmers.x 66 attention.span.media.farmers.y 66 attention.span.media.farmers.z 70 attention.span.media.government 66 attention.span.ud.government 62 attention.span.rnf.government 62 attention.span.cbo.government 62 http://dx.doi.org/ complexity, governance & networks – vol. 5, no 1 (2019) special issue: adaptive governance of coupled social-ecological systems, p. 24-64 doi: http://dx.doi.org/10.20377/cgn-72 64 university of bamberg press attention.span.farmers.x.government 62 attention.span.farmers.y.government 62 attention.span.farmers.z.government 62 decay decay.of.ud 0.7 decay.of.rnf 0.8 decay.of.cbo 0.9 decay.of.farmers.x 0.9 decay.of.farmers.y 0.9 decay.of.farmers.z 0.9 decay.of.government 0.9 concern threshold concern.threshold.ud 0.5 concern.threshold.rnf 0.4 concern.threshold.cbo 0.2 concern.threshold.farmers.x 0.3 concern.threshold.farmers.y 0.3 concern.threshold.farmers.z 0.3 concern.threshold.government 6.02 bloom variable bloom.seeds 5 num.neighbors.bloom.spread 4 max.time.blooming 7 nutrient.threshold 1000 large.bloom.threshold 5 rainfall.factor 0.4 fertilizer.application.rate 0.5 nutrients.desired 15 nutrients.regulated.slidebar 2 http://dx.doi.org/