peton173 The Unsuspected Dynamics of the Regulative Pillar: The Case of Faute Inexcusable in France Hélène Peton � Stéphan Pezé Abstract. Institutional work translates actors’ capacity to bring about change in an institutionalised practice through its creation, continuation or destruction. This research examines political institutional work in order to understand actors’ activity in the regulative pillar of the institution. With its focus on the cognitive and normative institutional pillars, the existing literature has underestimated the regulative pillar’s contribution to institutional change, ignoring part of its role in institutional change. We propose to examine the status conferred on this regulative dimension through a qualitative study of an institutionalised practice, the legal concept of Faute Inexcusable in France from 1898 to 2012. We use secondary data including legal data, contextualised with other sources, and interviews with two key actors in the field. We show the existence of a genuine “regulative dynamic”; beneath its apparent stability, the regulative pillar is in fact the setting for institutional struggles that lead institutionalised practices to evolve. We highlight sequences of institutional work in which various forms of political institutional work interact. This research advances understanding of institutional dynamics in the regulative pillar, casting light on actors and mechanisms that have so far gone unreported. The article thus contributes to a rehabilitation of the regulative pillar’s role in neo-institutional theory. ! The question of the role played by actors in institutional change is central to recent debates in the neo-institutional stream of research. The concept of institutional work (Lawrence & Suddaby, 2006) offers a particularly fruitful perspective for understanding how the activities undertaken by actors lead to an evolution in the institution, and therefore to institutional change (Ben Slimane & Leca, 2010). These purposeful actions create, maintain or destroy institutions. Institutional work concerns the three pillars that make up an institution (Scott, 2008). While the cognitive and normative dimensions have been specifically studied, especially as regards the stage of institutional creation, the regulative pillar, which translates certain actors’ capacity, by virtue of their authority, to place constraints on the behaviour of other actors (Caronna, 2004), is only rarely studied in the literature on institutional change (Scott, 2008). This situation is less the reflection of a lack of interest than the result of an over-restrictive view of the regulative pillar, focusing principally on the State as the central actor and the law as the only instrument. This simplistic view of the regulative pillar leads to a peripheral analysis of the institution’s coercive dimension: institutional work is considered to be produced in the cognitive and normative dimensions, with the regulative dimension then passively registering these changes (Hoffman, 1999; Maguire & Hardy, 2009). As a weapon wielded by the dominant actors in the field, the coercive power of the rule encourages continuation of the status quo and institutional maintenance strategies (Russo, 2001). Finally, when it is a source of change, the law is perceived simply as an outside driven factor affecting the institution, leading to institutional work on the cognitive and normative dimensions (Oliver, 1992). But as Scott emphasises, there is no reason not to take a more specific, detailed look at the regulative pillar. Perkmann and Spicer (2008) have M@n@gement 2014, 17(3): 145-179 145 Université Paris-Est, IRG (EA 2354), UPEC, F-94010, Créteil, France helene.peton@u-pec.fr Université Paris-Est, IRG (EA 2354), UPEC, F-94010, Créteil, France stephan.peze@u-pec.fr www.management-aims.com © The Author(s) mailto:helene.peton@u-pec.fr?subject= mailto:helene.peton@u-pec.fr?subject= mailto:stephan.peze@u-pec.fr mailto:stephan.peze@u-pec.fr http://www.management-aims.com http://www.management-aims.com also highlighted the existence of a political institutional work specific to the regulative pillar, suggesting the existence of as yet little-explored underlying activity. This study proposes to examine the institutional struggles specific to the regulative pillar. Our reflection was guided by two main questions: who are the actors who shape the regulative pillar? How do they act and interact within that pillar? A study of the law on Faute Inexcusable1 in France between 1898 and 2012 offers a particularly interesting case to explore these questions. The legal concept of faute inexcusable has existed for a century in French law. It is used in disputes between an employer and an employee who suffers injuries as a result of or in the course of his work. Claims by the victim of faute inexcusable by the employer have serious consequences for businesses, because if the courts declare them guilty of the offence, the victims will be awarded additional compensation payable by their employer. For a long while the concept was rarely used, but lawsuits for faute inexcusable are now widespread and have become an institutionalised practice (Maguire & Hardy, 2009): the concept shapes perceptions and understanding shared by all actors in the field of occupational health. Labour relations and employers’ responsibilities are defined through faute inexcusable, and the concept is hardly ever challenged. Its stability over time confers on it the status of an institutionalised practice, i.e. it is taken for granted by all actors. ! By questioning the traditional understanding of the regulative pillar and its role in an institutionalised practice change, we make two main contributions to the neo-institutional literature: first, we show that the regulative pillar is more dynamic and complex than the literature makes out, since it is made up of a variety of actors able to conduct elaborate forms of political institutional work, which leads us to identify and define a “regulative dynamic”: next, we show that because of this regulative dynamic, the regulative pillar plays a continuous role in institutional change. It is not the auxiliary pillar depicted in the literature, acting only as a trigger or ratifier of institutional change. ! This article comprises five sections. The first section presents the theoretical framework from which our investigations emerge. The second section introduces the research design, presenting the choice of the case studied (faute inexcusable), then the methodology applied to collect and analyse data. The results are reported in the third section (which provides a description of the case followed by its analysis). The fourth section proposes an analytical summary of the case, leading to a final section containing a discussion of results with a view to determining the contributions, limitations and avenues for research raised by this study. THEORETICAL FRAMEWORK ! Institutions can be defined as shared rules and meanings that offer a framework for social interaction and the roles attributed to it (Fligstein, 2001). As Barley and Tolbert (1997: 99) undermine, institutions are an accumulation of shared codes that make it possible to impose a practice as legitimate. These institutionalised practices are made up of cognitive, normative and regulative elements that underpin the stability and meaning of life in society (Scott, 2008). The determinism of the earliest research (DiMaggio & Powell, 1983; Meyer & Rowan, 1977) has in the last ten years made room for greater consideration of the actors’ capacity for action (Lawrence & Suddaby, 2006). The concept of “institutional work” is part of this perspective, stressing the capacity of several actors to become involved according to their interests in creating, maintaining or destroying an institutionalised practice (Lawrence & Suddaby, 2006). An institutionalised practice is thus the subject of struggles, efforts, coalitions and M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 146 1. Due to the difficulty of finding a completely satisfactory English equivalent for the French term and concept of Faute Inexcusable (literally, “inexcusable fault”, sometimes translated as “gross negligence”), we have decided to keep the original French expression in this article. strategic actions. These institutional creation processes have mainly been analysed from the angle of the cognitive, and to a lesser extent normative, foundations of institutional change. The concept of framing, for example, attracted the interest of many researchers (Benford & Snow, 2000). This focus on the cognitive pillar has slightly obscured understanding of changes in the other two pillars. Today, the regulative pillar is the least studied in the literature, even though major authors (Scott, 2008) have emphasised the relevance of looking at it differently. There is no justification for its marginalisation: according to Scott (2001), none of the three dimensions, cognitive, normative or regulative is predominant over the others. It would therefore appear just as relevant to study the three pillars and thus gain a better understanding of the regulative dimension. THE TRADITIONAL CONCEPTION OF THE REGULATIVE PILLAR ! In neo-institutional theory, the regulative pillar refers to the certain actors’ capacity, by virtue of their authority, to constrain the behaviour of other actors in an institutional field (Caronna, 2004). The research conducted by North (1990) brings out the central role of formal rules, which are different from the informal rules specific to the normative pillar, in supporting the institution. This constraint is expressed through formal rules that can be used to control behavioural compliance with the institutional order in a field. Coercion is the essential mechanism of this pillar. A central actor: the State ! Within the regulative pillar, the State’s coercive power is central. It was long considered as a neutral actor (Milgrom & Roberts, 1992), but the emergence of the literature on institutional work opened up new perspectives. Drawing particularly on the work of Skocpol (1985), Scott (2008) underlines the state’s capacity for action within a field, thus calling into question its potential neutrality since institutional work is defined by the intentionality of the action taken (Lawrence, et al., 2009). The only channel for the state’s institutional work: the law ! The State is the lawmaker, but it also plays a role as a recognised authority, and coordinates implementation of the law. It is therefore invested with legislative power and executive power. The law, as a vector of legitimacy, has been studied in several empirical studies. Edelman’s (1992) study of the Civil Rights Act, for example, stresses the influence of the law on the way institutions operate: when faced with a new law, organisations are obliged to adopt organisational forms that can display signs of conformity. The regulative dimension thus constrained organisations in their operation. This is one of the sources of isomorphism in an organisational field (DiMaggio & Powell, 1983). In this sense, the law enables the State to constrain organisations. But that is not its only role. The law also plays a role in construction and developments in the institutional field. Hiatt, Sine and Tolbert (2009) show that a law on prohibition banning alcohol production in the USA in the early 20th century completely redefined the boundaries of the field and paved the way emergence of a new institution: the American soda industry. ! These studies paint the following portrait of the regulative pillar (see figure 1): the State, through the law, places constraints on the behaviours of actors in civilian society. Its role is thus comparable to the role of producer of a quasi-exogenous shock that causes change in behaviours in an institutional field. The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 147 Figure 1. The regulative pillar in the litterature ! We believe that this oversimplistic interpretation reflects the lack of studies of the regulative pillar. The force of the law and the coercive power of the State are, of course, unquestionable. And yet the State is an actor that in reality comprises a large number of entities. Also, it is not the only producer of formal rules that constrain action. Hoffman (1999), for instance, stresses the role of past court decisions in his analysis of a process of change in the consideration given to environmental matters by the US chemicals industry. The actor who is the source of these decisions, the judge, is rarely studied. There is thus room for doubt regarding the idea that the power of the regulative pillar is expressed exclusively by the law enacted by the State (McCann, 1994). The concept of institutional work enables us to take this reflection further. THE REGULATIVE PILLAR THROUGH THE LENS OF POLITICAL INSTITUTIONAL WORK ! To address these questions, we refer to the literature on institutional work. The concept of institutional work is founded on the concept of chains of actors whose interactions bring about change (Lawrence, et al., 2009). They therefore commit to action in accordance with their interests, and a degree of effort is required. More specifically, the concept of political institutional work (Lawrence & Suddaby, 2006) aims to redefine relations between actors, notably through a redefinition of the rules and rights in the institution. According to Perkmann and Spicer (2007, 2008), this work, which is typical of the regulative pillar, takes three major forms: advocacy to influence the State and the laws; defining, which helps to set the boundaries between the inside and outside of the field and therefore determines which actors are included and excluded; and vesting, which confers specific roles and rights on certain actors. ! These different forms of political work are conveyed by a variety of actors. Diplomacy, negotiation and lobbying are political strategies that enable coalitions of actors to defend their interests (Djelic, 1998; Botzem & Quack, 2006; Lounsbury & Crumley, 2007). The study of management fashions by Perkmann and Spicer (2008) shows the role of the State but also of large dominant firms M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 148 (Cole, 1985), consumer groups and NGOs (Baron, et al., 1986) – all institutions that represent civil society. Research on social movements also underlines that protestors generally take action against the law, but also can also take action with and by the law (Agrikolansky, 2010). The representatives of civil society such as unions or victims’ associations act on the institution to defend their interests, drawing on the regulative pillar (Botzem & Quack, 2006; Lounsbury & Crumley, 2007). As the legal arena is particularly difficult to grasp for civil society, they often have to go through professional practitioners of the law to make their voices heard. The “mercenaries”, independent specialist lawyers (McCann & Silverstein, 1998), are one group that play this role and can thus participate in institutional struggles. And yet apart from the role of lawyers, the strategies of actors in the legal and court sphere are rarely studied in the literature. ! Consideration of the three forms of political institutional work gives a more nuanced view than the literature’s initial representation of the regulative pillar: coercive action by the State is exercised via a political institutional work of defining and vesting carried by the law; meanwhile, various actors in civil society concerned by the law try to influence the State in accordance with their own interests, through a political institutional work of advocacy (Figure 2). Figure 2. The regulative pillar through the lens of political institutional work The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 149 ! This description of the regulative pillar enables further examination of the pillar’s place in the logics of institutional change and maintenance. In-depth knowledge of this pillar renews the question of its role in the life-cycle of institutions. TOWARDS INTEGRATION OF A COMPLEX REGULATIVE PILLAR INTO THE LITERATURE ON INSTITUTIONAL CHANGE AND MAINTENANCE ! The lifecycle of institutions is recognised in the literature as a succession of periods of change and periods of stability (Scott, 2008). Few studies so far have taken an interest in the continuity and succession of these moments (Dansou & Langley, 2012). These studies (Farjoun, 2002; Zietsma & Lawrence, 2010) highlight key moments explaining the passage from one state to the other. From an institutional change standpoint, institutional work in process is seen as a response to external events that are a source of instability (Krasner, 1984) or the emergence of a contradiction between the institution and its environment (Jepperson, et al., 1991). Through an interpretation internal to the institution, focusing on institutional pillars, Caronna (2004) also shows that institutional work can be triggered by a misalignment between the institutional pillars and the resulting instability. While the processes of creation and change in institutions have been studied from the angle of institutional work (Hargadon & Douglas, 2001; Kitchener, 2002; Townley, 2002), institutional maintenance has attracted less attention from researchers (Scott, 2008). ! Institutional maintenance, generally considered as a phase of stability between two periods of tension, says nothing about the actors’ strategies taking place between those periods. But as it does not mean a status quo, but a more or less stable state, institutional maintenance is a balance that can be contested but also defended (Lawrence & Suddaby, 2006). For example, Blanc and Huault (2010) highlight the role of institutional work in perpetuating cognitive frameworks that lead to institutional maintenance in the field of the music industry. Taupin (2012) studies the credit rating industry, and by a combined assessment of institutional work and sociology of critique shows that periods of contestation lead the actors in the field to implement justification work that can lead to a situation of institutional maintenance. These studies underline the relevance of taking an interest in institutional dynamics, particularly the institutional struggles that underlie institutional maintenance. ! This gap in the literature is particularly glaring for the regulative pillar. The existing literature on that pillar has tended to obscure the existence of institutional struggles within the pillar. As Hauriou (1925, adopted by Millar, 1995) underlines, there is an implicit consensus that the law – as a concept and in its practical forms – is a source of maintenance. As a result, no study has yet looked at the different internal struggles within the regulative pillar. ! To go beyond the current simplistic, static conception of the regulative pillar in the neo-institutional literature and contribute to a better understanding of institutional maintenance and institutional change, we focus our research along two complementary axes. We want to understand both who the actors of the regulative pillars are, and also what their role is within their pillar. By recognising the capacity for action vested in these actors, who to date are under-explored in the literature, we can highlight the political institutional work ongoing in this pillar and show the relevance of looking at it anew with the broader aim of making a contribution to the literature on institutional change. To do so, we follow on from processual analyses of change in institutions (Delacour & Leca, 2010; Hoffman, 1999; Leblebici, et al., 1991; Tolbert & Zucker 1983). This approach enables us to respond to the call in the literature for an interpretation of the life cycle of institutions in its continuity, and more specifically forms of institutional work that participate in this movement (Dansou & Langley, 2012). The longitudinal case M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 150 study of faute inexcusable in France will highlight the sequence of institutional struggles in a dynamic perspective. RESEARCH DESIGN THE CASE OF FAUTE INEXCUSABLE ! To address the issues described above, we study the case of an institutionalised practice for which the regulative dimension occupies a central role: the legal concept of faute inexcusable2. Like the use of DDT, studied by Maguire et Hardy (2009), faute inexcusable in France is a concept that embodies a number of shared representations beyond its legal definition. It is an institutionalised practice with stabilised features that form a focus for changes brought about by actors in the field. These are institutional struggles that we propose to trace. ! In practice, faute inexcusable is a legal concept that can be applied by the judges when ruling on a dispute between an employer and an employee (or his/ her heirs) who has suffered injuries or physical harm contracted due to or in the course of his/her work. When this is found to result from an occupational accident or illness, the employee receives a standard set amount of compensation for the temporary or permanent prejudice suffered. This compensation is paid by the national health insurance system, which then “reinvoices” it to the employers responsible. But if the employee considers his accident or illness resulted from a particularly serious fault or negligence by his employer, he can apply to the Social Security Court2 for additional compensation on top of this standard compensation. Recognition by the courts of the employer’s “inexcusable fault” entitles the plaintiff to this additional compensation, offering greater reparation for the prejudice suffered3. ! The faute inexcusable concept shapes the perceptions and understandings shared by all actors in the field of occupational health. It defines the interactions in the field, particularly in terms of taking responsibility for work-related risks. Its stability over time confers on it the status of an institutionalised practice, i.e. taken for granted by all actors. The particularly striking regulative factors in this practice provide justification for studying it in order to bring out the importance of understanding the regulative pillar. A single case is selected under the intensity sampling approach (Patton, 2002)4 or revelatory case approach as defined by Yin (2009), in other words a case with good potential for development of new information on a neglected phenomenon (the regulative pillar). To study the case we used a methodology combining longitudinal and qualitative research (Delacour & Leca, 2011; Langley, 1999). The data collection and analysis methods are presented in the following section. DATA COLLECTION ! Priority was given to legal sources (laws and regulations) and the judicial sources (past rulings) to identify the central elements of faute inexcusable and achieve triangulation. The laws defining faute inexcusable and its characteristics form a primary source in historians’ sense of the term (Prost, 1996). They were identified and collected by consulting specialist legal publishers’ resources: the “Dictionnaire Permanent Social” dictionary of labour law published by Editions Législatives and the “Mémo social” labour law guide published by Liaisons Sociales6 . But there was a difficulty in collecting all the texts through these legal resources: they provide an updated version of the current laws and thus automatically exclude older, repealed or simply amended versions. To make sure The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 151 2. Given the diversity of legal and court systems internationally, for example between civil law and common law countries, it must be borne in mind that this case specifically concerns the French context. 3. Tribunal des Affaires de la Sécurité Sociale (TASS). 4. For further details on the court procedure and the terms of compensation, see Graser, et al., 2004. 5. “An intensity sample consists of information- rich cases that manifest the phenomenon of interest intensely (but not extremely).” (Patton, 2002: 234). 6. After looking up “faute inexcusable” in the glossary of these books, we consulted the relevant articles and identified references to the original texts: articles of law codified in the French Employment Code or Social Security Code, and case law. We then consulted these sources via the government law website Legifrance (http:// www.legifrance.gouv.fr) http://www.legifrance.gouv.fr http://www.legifrance.gouv.fr http://www.legifrance.gouv.fr http://www.legifrance.gouv.fr we do not miss a major text, we therefore analysed articles by legal experts specialised in occupational health and safety (e.g. Bienvenu, 1938; Le Roux, 2010; Sargos, 2003). ! These articles, which formed a secondary source, are also useful as a way of dealing with a second difficulty inherent to the laws and regulations identified. Ultimately, those laws and regulations are the end result of a development process that is hard to discern from reading them. Other sources are thus needed, to have more details of that development and grasp the intervention by actors of the regulative pillar. So in addition to the specialist legal articles referred to above, we also consulted the official reports issued by the actors of the regulative pillar themselves: reports by members of the French Senate, National Assembly (Parliament) or Cour de Cassation (Supreme Court) (chiefly Dériot & Godefroy, 2005; Le Garrec & Lemière, 2006). Also, working on the principle that faute inexcusable represents significant financial stakes both for businesses and for victims of occupational accident or illness, and also for the State, we complemented these data by consulting official reports (by the Cour des Comptes, the National Institute for Safety research, etc). We also studied works by historians to complement our understanding of the historical context of faute inexcusable across time (e.g. Buzzi, et al., 2006; Machu, 2009). Data collection was complete when we considered we had reached completeness (complétude, Pires, 1997), i.e. we could assert that we had grasped all major events forming our case in relation to our research question. ! Finally, in order to complete these documentary data, like Maguire and Hardy (2009) we organised two research interviews with two French lawyers who made a notable contribution to changes in the definition of faute inexcusable, Maître Michel Ledoux and Maître Jean-Paul Teissonnière7. The purpose of these interviews was to fill an information gap concerning the origins of a 2002 ruling by the Cour de Cassation, which was important because it resulted in a redefinition of faute inexcusable. The documentary sources provided contextual information explaining that the compensation system for victims of occupational accidents and illness was being called into question and had to change, but we lacked information about the specific event that triggered that change. These two interviews completed our understanding of the origin of the ruling issued in 2002, in the lawsuits for faute inexcusable brought by Maître Ledoux and Maître Teissonnière seeking compensation for “asbestos victims”. ! Table 1 summarises all these data (see also the full list of references in appendix 1). ! The sources collected were critically examined twice under the principles of the historical method (Prost, 1996). Assessment of external validity to check the authenticity of sources was facilitated by the official nature of the sources: the authenticity of the primary sources (laws, court rulings and decision by the Council of State) is unquestionable, and the secondary sources (institutional reports, academic research and legal studies) were also considered reliable as a result of their authors’ status (and the principle of peer review for published scientific research)8. We also observed a triangulation principle to ensure internal validity and guarantee the veracity of information. For example, the fact that the Cour de Cassation took no firm position on the definition of faute inexcusable until 1932 is mentioned both by Bienvenu (1938) and Pic and Kréher (1938): we therefore considered that this was an accurate historical fact that could be used in our data analysis. M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 152 7. We met with these two lawyers personally in January and March 2010. The interviews lasted respectively 2 hours and 57 minutes, and were recorded and transcribed in full. 8. Three more unusual secondary sources (Association d’Assistance et d’Entr’Aide, 1949; Belhache and Belhache, 1999; Pinaud, 2002) were examined in more detail and the quality of the authors and institutions convinced us that the sources were reliable. 9. Volume assessed based on the number of pages discussing faute inexcusable (when only one chapter or section of a document was used, the relevant pages are given in brackets). The total volume is 1815 pages. Table 1. Documentary sources used to form the case study Source type Data type References Approximate volume9 Primary: Texts materialising the regulative pillar at time Law Law of 09/04/1898 Law of 30/10/1946 Law of 06/12/1976 Articles of the Social Security Code (L. 452-1 à 5) 9 pages 12 pages 2 pages (article 29) 3 pages Primary: Texts materialising the regulative pillar at time Court rulings Rulings by the Cour de Cassation: Ruling of 22 February 1932 Ruling of 04 May 1937 Ruling of 15 February 1938 Ruling of 15 July 1941 Ruling of 18 July 1980 23 rulings of 28 February 2002 Ruling of 11 April 2002 Ruling of 19 December 2002 Ruling of 29 June 2005 Ruling of 28 February 2006 Ruling of 05 March 2008 Ruling of 11 May 2010 4 pages 2 pages 1 page 5 pages 2 pages 190 pages 2 pages 3 pages 3 pages 3 pages 5 pages 10 pages Primary: Texts materialising the regulative pillar at time Other Decision by the Conseil Constitutionnel of 18/06/2010 9 pages Secondary: Texts providing context for the primary sources and explaining their changes Legal studies Association d’Assistance et d’Entr’Aide (1949) Belhache and Belhache (1999) Bienvenu (1938) Boisselier (2008) Chapouthier (2009) Douard (1961) Editions Législatives (2010) Guillemy (2006) Jaillet (1980) Le Roux (2010) Pic and Kréher (1938) Ray (2010) Rouast et Durand (1953) Rouast, et al. (1962) Sargos (2003) Sargos (2006) Seillan (2009) Tessonnière and Topaloff (2009) Vingiano (2011) 23 pages (9 - 31) 25 pages (32-33; 62-77; 134-140) 129 pages (1-129) 176 pages 2 pages 27 pages (37-61; 66-67) 40 pages 2 pages 420 pages 2 pages 48 pages 10 pages (157-168) 5 pages (527-531) 6 pages (405-410) 8 pages 2 pages 9 pages 32 pages 200 pages Secondary: Texts providing context for the primary sources and explaining their changes Institutional reports .Cour de Cassation (2002) Cour de Cassation (2007) Cour de Cassation (2010) Cour des Comptes (2002) Dériot and Godefroy (2005) Le Garec and Lemière (2006) Masse and Zeggar (2001) Laroque (2004) Vidalies (2011) Yahiel (2002) 10 pages (109-118) 16 pages 19 pages 24 pages (136-159) 26 pages (166-191) 63 pages (275-333; 462-463; 475-476) 50 pages 84 pages 75 pages 10 pages (14-23) Secondary: Texts providing context for the primary sources and explaining their changes Academic research (including by historians) .Buzzi, et al. (2006) Graser, et al. (2004) Machu (2009) Rosental (2008) 31 pages (10-40) 9 pages 13 pages 23 pages The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 153 DATA ANALYSIS ! Analysis of the data collected involved three steps: first, we reduced the volume of data concerning the history of the concept of faute inexcusable in a dual operation of (A) periodisation and (B) expression as a narrative; then, by successively examining the narrative of each period identified, we carried out (C) analysis of the regulative pillar itself. (A) Identification of the major periods in the history of faute inexcusable ! We began by reducing the volume of data collected, building up a narrative arranged by major periods. Rather than a strictly chronological division, we used temporal (Langley, 1999) periodisation, isolating separate sequences containing fairly stable developments consisting of action by the protagonists within the constraints of the existing context, ultimately leading to a change in the context. This change in turn affects the actions of protagonists – creating a new development identified in another period, and so on. Each period can give a separate analysis of the same phenomenon, in what Langley calls “internal replication possibilities” (1999:704). ! The three following periods were thus identified inductively through an iterative reading of our data (see also figure 3): (1) Earliest references to faute inexcusable (1898-1941). This period traces the foundations of this institutional practice, from its “birth” to the first stable definition of 1941 reached gradually by the Cour de Cassation. (2) Gradual shift (1941-2002). This second period is bounded by court rulings establishing the two successive definitions of faute inexcusable. In the period between the two, the laws were stable. Several movements related to changes in society, the legal framework of occupational risk prevention and organised action by asbestos victims and their lawyers gradually brought about a remarkable turnaround in court rulings. (3) Quasi-automatic declaration of faute inexcusable (since 2002). This last, contemporary period illustrates the importance of the event of the new turn in court rulings in 2002. This increased the importance of financial issues relating to compensation for the victims of occupational risks – and calls into question the very existence of faute inexcusable. Figure 3. The major periods in the history of faute inexcusable M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 154 (B) Writing the narrative of the periods identified ! To prepare our narrative, we noted all actors who in the data collected intervened in a range of capacities to use or attempt to modify the concept of faute inexcusable. This information was recorded progressively in a table stating the actor, the action taken, and the date or period of that action and its effects. This time-sensitive classification was then used to write the narrative in a combined synchronic and diachronic approach, to reconstruct the phenomenon that is the evolution of the concept of faute inexcusable over time, in all its complexity (Gombault, 2006). (C) Analysis of change in the regulative pillar ! We analysed the narratives in two stages: by describing changes in the regulative pillar specific to each period (via establishment of the sequence of forms of political institutional work that triggered a significant regulative change marking the transition to a new period), then by searching for patterns common to those changes with a view to bringing out and describing a generic regulative dynamic. ! Establishing the changes in the regulative pillar specific to each period. To describe changes in the regulative pillar we put the forms of political institutional work into sequence. This makes it possible to trace the origins of a significant change in the institutionalised practice that opens up another period. The procedure was as follows: first, we classified the type of political institutional work covered by the action by the actors mentioned in the narrative (actions and actors identified in advance during the narrative-writing work, see section B above); next, we reconstructed the sequence of the forms of institutional work by systematically noting the targets and effects of actions, to determine whether some of them were the triggers for other actions (causal link). For each period identified, based on the previously developed tables (cf section B), we classified the action noted according to the form of political institutional work undertaken by the actors involved. This classification was based on thematic coding using the three categories identified by Perkmann and Spicer (2007, 2008). The political work of advocacy corresponds to action undertaken by one actor to try and persuade another actor. These actions belong solely to the domain of attempted influence (whether or not they are successful, i.e. followed by a political institutional work of vesting or defining). Advocacy was identified in the data by the following question: was the purpose of this action to persuade one or more other actors? ! The political work of vesting corresponds to actions that directly create (or modify) obligations or prohibitions via introduction of a rule. These actions thus have a strong transformative capacity and require promulgation of a law recording the creation/modification of the rule. It was identified in our data by the following question: was the effect of this action to create or modify the very nature of the concept of faute inexcusable? ! Finally, the political work of defining corresponds to actions that (re)define the terms of application of the rules resulting from the political work of vesting. These actions have significant consequences, for they redistribute the beneficiaries (or targets) concerned by application of a rule. They also require promulgation of a law which, unlike the case of vesting, modifies not the existing rule, but its terms of application. This was identified in our data using the following question: was the effect of this action to change the terms of application of the concept of faute inexcusable? ! Next, we analysed the potential combinations of these actions, starting from their effects. Two types of effect emerged and were then sought throughout the data: political institutional work resulting in a change in the laws and regulation structuring the regulative pillar or with effective influence on other actors that leads them to engage in political action; and unsuccessful advocacy- The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 155 type forms of political institutional work with no decisive influence on any change in the regulative pillar. ! This analysis results in a numbered list of actions in chronological order, showing the sequence of events leading to a significant change in the regulative factors that constitute faute inexcusable, in other words: the change in the regulative pillar of faute inexcusable. This is presented in tables summarising the analysis, then modelled in a figure for each period, to illustrate the dynamic aspect of the sequence of forms of political institutional work. The operation is identical for each period studied, as such replication is possible with the temporal bracketing method used (Langley, 1999). ! Establishing a generic regulative dynamic. The analysis continued with a comparison of similarities and differences in the changes during the three periods, in order to produce a more abstract conceptual characterisation of the regulative pillar and its dynamic of change. In practice, we compared the three changes identified two by two, to identify patterns (following the inter-case analysis recommendations of Miles and Huberman, 2002). Three comparisons were thus undertaken (periods 1-2, periods 2-3 and periods 1-3) and brought out several recurring and divergent factors and three phases common to all three periods. A final re-analysis of the three periods was conducted to arrive iteratively at a more detailed description of the content of each of these factors. The outcome of this analysis was a generic regulative dynamic presented in a proposed summary model. ! In line with these analysis techniques summarised in figure 4, we now present the narrative and analysis of each of the three major periods identified as regards the changes in the concept of faute inexcusable, before proposing a summary conceptualisation of the regulative pillar and its dynamic. Figure 4. Schema summarising data analysis steps and their results M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 156 RESULTS PERIOD 1: EARLIEST REFERENCES TO FAUTE INEXCUSABLE (1898-1941) ! The legal concept of faute inexcusable originated in the French law of 9 April 1898. With the expansion of factories and industrial mechanisation the number of occupational accidents had risen, involving difficult investigations concerning the responsibility of the worker-victim and his employer. In the last few decades of the 19th century it was up to the victim to prove that the employer was responsible for the accident. When such proof was provided, there was full reparation for the prejudice, in accordance with the standard laws of the time. But injured workers (and their families) had little resources to win their case, especially as it was easy to argue that the accident was their own fault due to clumsiness or negligence, often fatigue-related (Boisselier, 2008). ! Several separate initiatives contributed to a major change, which was not easily won in a liberal industrial world that did not take kindly to the State’s attempts to interfere in its private affairs (Seillan, 2009). The road to change was prepared in a number of bills of law proposing to reinforce the employer’s liability in response to “strong feelings among the working population, whose electoral influence was becoming very serious” (Jaillet, 1980:39)10, court rulings that were increasingly favourable to victims, taking the principle of employer responsibility further, and other initiatives such as the Paris Chamber of Masonry’s formation of a mutual society to pay indemnities to injured workers as early as 1859. Gradually, these initiatives developed the “idea of occupational risk and demonstrated that good management of its insurance was possible, which ultimately led to a considerable legislative work in the form of the well-known law of 9 April 1898” (Boisselier, 2008:69). This law resulted from a turbulent process that began in 1880 (Bienvenu, 1938) and was marked by a hard-won agreement between France’s Chamber of Deputies (the former name for the National Assembly) and Senate on a common conception (Jaillet, 1980): introduction of the principle of automatic compensation for occupational accidents11, modulated by the nature and extent of the employer’s fault. Victims were no longer obliged to prove fault by their employer and its connection to the prejudice suffered. In return, a standard amount of compensation was awarded rather than full compensation, except in the case of faute inexcusable, which became the victims’ only possible grounds for suing their employer for a higher level of financial compensation (subject to a cap)12. ! This compromise was achieved after almost 20 years of to-and-fro discussions between the Chamber of Deputies and the Senate, and was longstanding despite several attempts to have it revised (Jaillet, 1980). Employees’ unions called on “the various commissions in charge of examining bills of law” (Machu, 2009:192) to encourage members of parliament to replace certain provisions of the law of 1898. These attempts failed, as did a bill presented in December 1925 aiming to define faute inexcusable as an employer’s failure to comply with the laws in force, which was initially accepted by the Chamber of Deputies but then rejected by the Senate Commission. “Once the legislative solution was impossible, the union’s action to broaden the classification of faute inexcusable continued before the courts” (Machu, 2009:201). The unions, whose advice was circulated through the union press (such as the journal Droit ouvrier, meaning Workers’ rights) and leaflets, supported employees in their court cases. They tried to influence the criteria for declaring that faute inexcusable had occurred. While the unions sought to open up those criteria, the employers were seeking exemption from responsibility by narrowing down the criteria around the deliberate nature of the fault. The law of 1898 does not define the criteria for recognition of faute inexcusable: “The The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 157 10. All English translations of quotations from sources originally in French are the authors’ own for the purposes of this article. 11. Occupational illnesses were incorporated into the measures from 1919. 12. “When it is proven that the accident is due to inexcusable fault [faute inexcusable] by the employer or the persons to whom he has delegated management, the indemnity may be increased, but the amount or total amounts allocated cannot exceed the reduction or the annual salary.” (Paragraph 3, Art. 20, Law of 9 April 1898). lawmaker did not wish to take sides on the nature and definition of the new concept it was introducing into our Law […] It was perhaps wiser, and that is our opinion, to leave the Courts the care of applying such a delicate concept” (Bienvenu, 1938: 34-35). This lack of definition was to have consequences for court rulings, as from 1898 to 1932 different courts took widely varying views of faute inexcusable, creating an arbitrariness such that neither the employer nor the employee could reliably foresee the outcome of a lawsuit. “In particular, it was to take a long time for case law to emerge from the most disappointing chaos. This was the period during which the Cour de Cassation refrained from playing its true role and took refuge behind the sovereign assessment of the judge of fact” (Pic & Kréher, 1938:15). The situation gradually stabilised after a ruling by the Chamber of Requests on 22 February 193213 in which the Cour de Cassation took classification of faute inexcusable back in hand. “It was soon clear that the Civil Chamber intended to differentiate more strongly between inexcusable fault [faute inexcusable] and serious or deliberate fault [faute lourde], with which certain decisions were tending to confuse it” (Jaillet, 1980: 102). Yet “for several more years, the Cour de Cassation would show great uncertainty and a regrettable hesitancy to break with the erring ways of the past” (Pic & Kréher, 1938: 36). ! Meanwhile, the members of parliament were aware of this legal uncertainty and the “erring ways” it caused, and set out to clarify the definition of faute inexcusable. On 29 December 1937, the Chamber of Deputies adopted a bill of law intended to define faute inexcusable in more detail14. But it was rejected by a Senate commission, mainly on the grounds that decisions by the Cour de Cassation were growing increasingly detailed (Bienvenu, 1938). Through decisions of 4 May 1937 and 15 February 1938 especially15, the Cour de Cassation had indeed begun to stabilise the criteria for accepting the existence of faute inexcusable. However, the writer of official reports on these two decisions (Gazier) explained that “the Cour de Cassation’s understanding of the law was completely changing and it continued tentatively, sometimes even slightly losing its way”, and wanted the question to be put before all the Chambers of the Court, to reach a final decision (Jaillet, 1980: 103). His wish was finally granted, when the Cour de Cassation united all its Chambers on 15 July 1941 in the following definition given in the ruling for the “Veuve Villa” case: “any exceptionally serious fault deriving from deliberate act or omission, from the awareness of the danger that the perpetrator should have, from the lack of any justificatory cause, distinguished by the lack of any intent16”. Five factors are thus required for classification as faute inexcusable: the deliberate nature of the act or omission, awareness of the danger, lack of any justification, lack of intent, and the exceptionally serious nature of the fault, generally resulting from “deliberate exposure of the victim to a danger, either by ignoring all the precautions or basic rules of care, or by violating a health and safety rule” (Graser et al., 2004: 135). From the date of this definition (and right up to 2002), the Cour de Cassation systematically overturned court rulings that did not refer to every point of the 1941 criteria (Douard, 1961; Belhache & Belhache, 1999). ! In short, four types of actor were involved in the gradual definition of faute inexcusable: parliamentarians17, employers’ and employees’ unions, victims of occupational accidents and illness, and the Cour de Cassation. They instigated court action corresponding to several forms of political institutional work, both successfully and otherwise, and contributed to the change in the regulative pillar that led progressively to the new definition of faute inexcusable (see table 2 and figure 5). M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 158 13. Cass. Req., 22 February 1932, D.P. n ° 1932.1.25, note A. Rouast. 14. “For instance, inexcusable fault [faute inexcusable] on the part of the employer exists when a failure to comply with the provisions of Title 2, Book 2 of the Labour Code on worker health and safety, having a direct connection with the causes of the accident, as noted in a regular report by an Employment Inspector, has been the subject of a legal sanction.” (quoted by Bienvenu, 1938:146). 15. Cass. rec., SIREY, 4 mai 1937, n° 1931.1.331 – R.G.A.T. 1937.1010 et Cass. civ., 15 February 1938, D.H. n° 1938.261. 16. Cass. chambres réunies, 15 July 1941, Dame Veuve Villa, JCP 1/41, n° 705. 17. “Parliamentarians” covers members of the French National Assembly and members of the French Senate, who vote on adoption of laws and oversee the activity of the government (Art. 24 of the French Constitution). Figure 5. Representation of the regulative pillar during the first period in the history of faute inexcusable Table 2. Actors of the regulative pillar and institutional work during the first period in the history of faute inexcusable Actors Actions Institutional work concerned Consequences Parliamentarians (1) Adoption of the law of 1898 Vesting Creation of the concept of faute inexcusable and a new compensation regime for victims of work-related accident and illness Employers’ unions (2) Lobbying of parliamentarians (end of 19th century) Advocacy Failure. Continuation of the law of 1898 and the principle of flat-rate compensation for accidents (favourable to employers) Employees’ unions (2) Lobbying of parliamentarians (end of 19th century) Failure. Continuation of the law of 1898 and the principle of flat-rate compensation for accidents (favourable to employers) Syndicates and victims of occupational accidents (3) Legal action (lawsuits) for faute inexcusable Advocacy Variable success: differing ruling, inconsistent acceptance of faute inexcusable Cour de Cassation (4) Ruling of 1932 Defining Reworking of classification of the criteria for accepting faute inexcusable, the relative legal uncertainty continues Parliamentarians (5) Bill of 1937 for a definition of faute inexcusable Advocacy Failure. The legal uncertainty continues Cour de Cassation (6) Rulings of 1937 and 1938, the “Veuve Villa” ruling of 1941 Defining Gradual definition of faute inexcusable The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 159 ! The source of this change was a factor internal to the regulative pillar: the legal uncertainty resulting from the lack of a precise definition of faute inexcusable in 1898. The original law was promulgated by the parliamentarians via a political institutional work of vesting (institutional work no. 1 in Figure 5), i.e. creating an obligation, in the event of faute inexcusable, to pay accident victims a higher amount of financial compensation. The unions tried to put pressure on the parliamentarians to broaden the situations in which faute inexcusable was deemed to have occurred (no. 2). This institutional work of advocacy was a failure – no new article of law was adopted. The unions then changed strategy and began bringing lawsuits for faute inexcusable before the courts (no. 3) – with varying success depending on the courts involved. This is a form of advocacy, in the sense of an attempt to influence the rulings issued. The accumulation of contradictory rulings – a form of legal uncertainty over the conditions for application of the law which affected both employers and victims of occupational accident/illness, who were unable to predict the outcome of a ruling in advance – had two consequences. First, in 1932, the Cour de Cassation took the definition of faute inexcusable in hand (no. 4). This was a political institutional work of vesting in the sense that it precisely defines the conditions for application of the concept of faute inexcusable – i.e. precisely defines the role of other actors (judges), and also their rights (victims of occupational accident/illness). Meanwhile, the parliamentarians, aware of the uncertainty resulting from the vagueness of the faute inexcusable concept, undertook internal political advocacy work (no.5) in the sense of back-and-forth discussions between the Chamber of Deputies and the Senate on articles of law proposing a more detailed definition of faute inexcusable. This work failed, because the Cour de Cassation was clarifying its rulings in the same period. In 1937, 1938 and finally in 1941, the Cour de Cassation gradually stabilised the criteria indicating the existence of faute inexcusable (no.6), removing the legal uncertainty surrounding this legal device and stabilising the regulative pillar, thereby opening up a new period. PERIOD 2: GRADUAL SHIFT (1941-2002) ! The definition of 1941 was the basis for court rulings until 2002. It was notably reasserted, unchanged, in a decision by a plenary session of the Cour de Cassation in 198018. It set the criteria for recognition of faute inexcusable - very strict criteria which in practice only led to a low number of cases of declared faute inexcusable (Jaillet, 1980; Sargos, 2003). The second period of faute inexcusable thus established a form of social peace inherited from the principles of the law of 1898, in which the victims automatically received a fixed amount of compensation. “Tirelessly since the decision of 1941 by all Chambers of the Cour de Cassation, the courts and the highest court itself have constantly repeated these five criteria that constitute faute inexcusable. Neither the legislator nor a dissident court has disrupted the harmony of this formula”. (Jaillet, 1980:115) Yet several related developments in society were to lead to redefinition of faute inexcusable: (1) observation of the ineffectiveness of the French prevention measures, (2) challenging of the standard compensation for victims of occupational accident/illness and (3) organisation of a court action strategy by two lawyers (Maître Ledoux and Maître Teissonière) to obtain indemnities for asbestos victims. ! (1) The social peace of standard compensation for victims, as established by the 1947 ruling, gradually eroded. This approach to reparation for accidents led to the idea that there was no need for an ambitious prevention policy, since the prejudice suffered would in any case be compensated within well-defined limits (Jaillet, 1989; Sargos, 2003). This fatalistic view weakened the principle of the employer’s duty to ensure safety (Seillan, 2009). ! Also, the spirit in which prevention should operate changed profoundly M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 160 18. Cass. assemblée plénière, 18 July 1980, n ° 78-12570. during the 20th century. Under the influence of European Union Directives, the employer was strongly recommended to take active steps for preventive health measures. Employee health and safety now had to be considered in terms of outcomes (Guillemy, 2006). Although the definition of faute inexcusable remained unchanged, its framework of application – the legislative requirements incumbent on the employer – evolved significantly. The Cour de Cassation actually makes explicit reference to this change and its influence on the nature of employer responsibility in its report of 2002: “deciding that the employer is contractually bound by a safety performance obligation19, if not "contra legem", at least goes beyond the law, whose balance is upset by this principle [of standard compensation for victims of occupational accident/illness], since in the absence of specific legislation an employee who suffers an accident at work or has a recognised occupational illness would be compensated for the entire prejudice simply through operation of contractual liability.” ! (2) In addition to the observed ineffectiveness in the preventive logic, there was the fact that the compensation regime for victims of occupational accident/ illness – a standard amount of compensation for all – looked out of step with other compensation systems based on full reparation, as shown by several reports issued by State bodies: a report remitted to the Minister of Justice by the Chairman of the Specialist Committee for occupational illnesses of the CSPRP20 (High Council of Occupational Risk Prevention) and a member of the l’IGAS21 (Masse & Zeggar, 2001), another report by an IGAS Inspector at the request of the Minister for Employment and Solidarity (Yahiel, 2002) and a third by the Cour des Comptes on the system of compensation for occupational accident/illness (Cour des Comptes, 2002). The legislator responded with the law of 1976, one of whose objectives was to “offer greater reparation to the victim and his heirs through higher compensation” (Jaillet, 1980:301), while still below a certain ceiling. ! Conscious that this gap was appearing, the Cour de Cassation gradually broadened the first definition of faute inexcusable through its rulings. In particular, it placed less importance on the exceptional nature of the fault (Chapouthier, 2009) to “make up for the gradual loss of the advantageous nature of the reparation system resulting from the law of 1898” (Boisselier, 2008: 71). (3) Among the appeals referred to the Cour de Cassation in the late 1990s were cases brought by victims of occupational illness related to inhalation of asbestos particles. These cases did not result from a collective move by individual plaintiffs; they were a deliberate court action strategy implemented by two lawyers, Maître Ledoux and Maître Teissonière, engaged by former members of the Jussieu anti-abestos collective which aimed both to have asbestos banned in France, and to achieve recognition and compensation for victims of asbestos. This social movement of the 1990s, which led to a ban on asbestos, relied extensively on the scale of the legal process set in motion: “In the criminal courts, we filed the suit straight away, and the cases are still being examined now. Then there was the social distress, which called for urgent action as regards compensation. (…) The priority was to have the employers declared responsible, and obtain compensation” (research interview with Maître Teissonière, March 2010) ! The broad media coverage of the earliest hearings gave a voice to the victims, who were still invisible, and afterwards hundreds of people came to attend the hearings. “So we started the proceedings for faute inexcusable despite the problems (…) And when people saw it worked, they came along.” (Maître Teissonière, March 2010) The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 161 19. Obligation de sécurité de résultat. 20. CSPRP: Conseil Supérieur de la Prévention des Risques Professionnels. 21. IGAS: Inspection Générale des Affaires Sociales. ! The lawyers had the idea of initiating lawsuits for faute inexcusable against two large asbestos processing companies in France, Everit and Eternit. This strategy was based on the large number of lawsuits: “So we plead the case, and things start moving all over the place. We win, they appeal, we plead the case again before the appeal courts and we win… (…) Further appeal before the Cour de Cassation by Eternit and Everit and… on the fateful day of 28 February 2002, the Social Chamber of the Cour de Cassation rejects the appeals and invents the well-known safety performance obligation. There have been 18 rulings by the Cour de Cassation… well, 17 of them were my Everit cases from Bordeaux.” (Maître Ledoux, January 2010) “Little by little the courts turned around, the appeal courts turned around, and the Cour de Cassation redefined faute inexcusable in 2002.” (Maître Teissonière, March 2010) ! The effect of the court action strategy developed by the lawyers brought the concept of faute inexcusable centre stage against the backdrop of a public health scandal, thus creating the space the Cour de Cassation judges needed to modify the definition of criteria for recognising this institutionalised practice: “Whereas by virtue of an employment contract with his employee the employer has a safety performance obligation towards the employee, particularly regarding occupational illness contracted by that employee as a result of the products made or used by the firm; whereas failure to fulfil this obligation is by nature an faute inexcusable as defined by article L. 452-1 of the social security code, when the employer was or should have been aware of the danger to which the employee was exposed, and did not take the necessary measures to protect him from it22. Of the old criteria, the only one remaining is the awareness of the danger; a second criteria is added, the fact of not having taken measures to protect the employee exposed to that danger (Teissonnière & Topaloff, 2002). “This takes us from proven fault to presumed responsibility” (Graser, et al., 2004:138). Winning lawsuits for faute inexcusable went from being difficult to practically automatic. However, this court action strategy was only successful because it was combined with two underlying trends – dissatisfaction with preventive health and safety policies and the out-of-step compensation systems – of which the Cour de Cassation was clearly aware when it redefined faute inexcusable: “the reparation system for occupational accidents, which was extremely favourable to employees when created, today reduces their rights to compensation” (Cour de cassation, 2002). ! In short, five types of actor were involved in this second period, which ends with the redefinition of faute inexcusable: the European Union, the parliamentarians, the state bodies (Cour des Comptes, IGAS and CSPRP), the lawyers for asbestos victims and the Cour de Cassation. They instigated action corresponding to several types of political institutional work, which fed an internal change in the regulative pillar that ultimately led to redefinition of faute inexcusable (see table 3 and figure 6). M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 162 22. Definition given in several decisions by the Social Chamber of the Cour de Cassation on 28 February 2002 (par ex. Cass. soc., 28 févr. 2002, no 99-18.389, no 837, Sté Eternit industries c/ Delcourt-Marousez et a.). Figure 6. Representation of changes in the regulative pillar during the second period in the history of faute inexcusable The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 163 Table 3. Actors in the regulative pillar and institutional work during the second period in the history of faute inexcusable Actors Actions Institutional work concerned Consequences European Union (1) Issuance of the 1989 European directive Vesting Promotion of more proactive legislation for prevention, with greater responsibility for the employer Parliamentarians (1 a) Promulgation of the law of 1991 Vesting Formal recognition in the law of new responsibility for company managers (C. trav., Art. L. 4121-1 et s.) State bodies (Cour des Comptes, IGAS, CSPRP) (2) Publication of reports (2000s) Advocacy Establishment of the ineffectiveness of prevention policies; observation that the compensation system for occupational accidents and illnesses is out of step with the standard legal system Lawyers for asbestos victims (3) Lawsuits brought before the courts for asbestos victims (1990s and 2000s) Advocacy Implementation of a court action strategy: increase in the number of lawsuits, putting the question of compensation for asbestos victims on the Cour de Cassation agenda Cour de Cassation (4) Turnaround in court rulings in 2002 Defining New definition of faute inexcusable, creation of the safety performance obligation, more favourable to victims ! The source of this change was external to the regulative pillar: it was the progressive misalignment between the regulative pillar and the cognitive pillar, the salient points being the two ideas that (a) the risk prevention system is ineffective and (b) differences in compensation awarded to victims under different legal regimes was creating an unfair situation. The first idea (a) led to a modification of the boundaries of employers’ responsibility for occupational health and safety. In practice, this modification was enacted through the European directive of 1989, in a political institutional work of vesting initially affecting French parliamentarians (no. 1 in Figure 6), who were obliged to transpose this directive in French law of 1991 (no. 1a), then employers and employees, by changing their respective rights and obligations. This law’s influence on faute inexcusable was indirect, because it did not concern the compensation systems for victims of occupational accident/illness. However, it was also a form of vesting in that it changed the more general framework for decisions concerning occupational health and safety. The second idea (b) is embodied in the regulative pillar by a political institutional work of advocacy carried by several State bodies including IGAS and the Cour des Comptes (no. 2). Their reports aim to influence parliamentarians by encouraging them to use the law to correct the injustices and inequalities resulting from the system derived from the law of 1898. This political work of advocacy targeting parliamentarians was a failure, but it was shared by the Cour de Cassation which, after a political institutional work of advocacy by the lawyers for asbestos victims (no. 3), was to redefine the criteria establishing faute inexcusable. This further example of the political institution work of defining by the judges of France’s High Court (no. 4) was explicitly driven not by legal uncertainty but by the gap that had gradually opened up between several different compensation systems – the misalignment between the regulative pillar and the cognitive pillar. The consequence of this political work is an increase in the levels of compensation received and thus a realignment of the pillars, opening up a new period. PERIOD 3: QUASI-AUTOMATIC ESTABLISHMENT OF FAUTE INEXCUSABLE (ONGOING SINCE 2002) ! The redefinition of faute inexcusable at the end of period 2 was soon extended to include occupational accidents23, with major consequences: “The asbestos affair drastically changed the right to safety at work. It was one thing before the asbestos affair, and something different afterwards” (Maître Teissonnière, March 2010). In proven cases of occupational accident/illnesses, declaration of faute inexcusable became practically automatic (Dériot & Godefroy, 2005): “It was no longer the exception to the rule […]. It can thus be supposed that under the new definition, any occupational accident or occupational illness recognised as such by the courts could involve the employer’s liability for faute inexcusable” (Pinaud, 2002: 11). The costs of compensating victims of occupational accident/illness thus rose substantially (Le Roux, 2010): in addition to the €600,000 - €700,000 that must be paid to a victim suffering 100% permanent partial incapacity, declaration of faute inexcusable generates additional compensation of between €38,000 and €335,000 depending on the court ruling (Pinaud, 2002). These consequences are considered as “a good prospect for insurers” offering employers insurance for faute inexcusable (Vingiano, 2011: 186). ! These consequences could grow further in the future. When the employer’s faute inexcusable is established, reparation for the prejudice suffered by victims is not full, but it is higher24. In order to encourage a principle of full reparation – and thus align the compensation system with other systems – the Cour de Cassation noted in its 2002 annual report: “The Social Chamber was thus moved, without departing from the legal rules which it is not within its power M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 164 23 . Cass. soc., 11 avril 2002, n° 00-16535, Hachadi c/Sté Camus Industrie et a. 24.. The victim may be awarded a (capped) increase in his compensation, and ask the employer for reparation of the prejudices to modify, to bring the compensation rules for victims of occupational accident or occupational illness closer to the standard rules.” In doing so, it upset the coherence between court rulings and the law, indirectly calling for a reaction from the legislator. ! In response to this shift in court rulings, several reports by the IGAS (Laroque, 2004), the Senate’s parliamentary commissions (Dériot & Godefroy, 2005) and the National Assembly (Le Garrec & Lemière, 2006) have considered that the legislator should intervene. One of the common scenarios proposed is the incorporation of the principle of full reparation for occupational accident/ illness, while restoring an offence of very serious fault for particularly negligent employers: “The law could redefine the specific fault for which the employer could be held liable before the courts, and restore its unusual nature opposite a compensation system for occupational accidents and illness, which would return to a mainly out-of-court system.” (Le Garrec & Lemière, 2006: 326). But these recommendations were not followed by legislative effect. In May 2010, the Cour de Cassation asked the Conseil Constitutionnel to verify that the provisions of the Social Security Code complied with the French Constitution. In its decision25, the Conseil Constitutionnel issued a reservation regarding the list of prejudices eligible for compensation: “in the presence of faute inexcusable by the employer, the provisions of [article L. 452-3 of the social security code] would by no means […] prevent those same people, from suing the employer before the same courts for compensation for all the prejudices not covered by book IV of the Social Security code”. The constitutional judge thus opened up the possibility for victims of occupational accidents/illness to seek reparation for prejudices not listed by the Social security code – with the compensation to be assessed on a case by case basis. Several judgements favourable to this full compensation logic have been issued: in 2010, a prejudice of anxiety was allowed, regardless of any effective outbreak of illness (it was enough to prove that an occupational illness was likely to develop at any time, and that this probability was the source of the anxiety26). A 2011 ruling by the Paris Appeal Court took up this prejudice of anxiety (awarding €15,000 per plaintiff) and also, for the first time, recognised a prejudice of “disturbed conditions of existence27” (€12,000 per plaintiff) for former employees likely to develop an asbestos-related occupational illness28. ! As a parliamentary report stresses, the Conseil Constitutionnel’s decision called for intervention by the legislator: “it is necessary for reasons of legal safety and intelligibility of the law” (Vidalies, 2011). The same appeal is made by the Cour de Cassation in the “Proposals for Civil law reform” in its 2010 report: “it would appear advisable to proceed without delay to a modification of the provisions of the social security code. The proposal made below seeks […] to set up full reparation as a principle under the standard legal rules for prejudice suffered by the victim of faute inexcusable” (p. 21)29. A bill of law aiming to incorporate full compensation into the Social Security Code was presented by some members of the French parliament, but rejected at the first reading by the National Assembly on 23 November 2011. In sum, the events of the third period in the history of faute inexcusable illustrate the major consequences of the turning point in case law in 2002 for firms and victims of occupational accident/illness. Although this period is not yet over, we can identify four types of actor: parliamentarians, state bodies (IGAS), the Cour de Cassation and the Conseil Constitutionnel (see table 4 and Figure 7). The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 165 25. Conseil Constitutionnel., 18 June 2010, no 2010-8 QPC. 26. Cass. soc., 11 May 2010, no 09-42.241, no 939, Sté Ahlstrom Labelpack c/ Ardilley et a. 27. trouble dans les conditions d’existence. 28. CA Paris, 1 December 2011, Sté ZF Masson. 29. The Cour de Cassation’s proposed amendment is as follows : “The provisions of the first paragraph of article L452-3 of the Social Security code are repealed and replaced by the following provisions : “Independently of the increase in the compensation allowance awarded under the previous article, the victim has the right to ask the employer before the social security court for reparation of all the prejudices not covered by that compensation, the increase and the indemnities stated in this book.” (2010:21) ! The source of this as yet unfinished change is internal to the regulative pillar: it upsets the balance of existing provisions, calling for reintroduction of consistency between the law and court rulings on compensation for victims of occupational accident/illness. First, the Cour de Cassation extended the definition of faute inexcusable to cover all cases of occupational accident/illness, in a political institutional work of defining (solid arrow no. 1 in figure 7) that had the effect of extending the scope of inconsistency between court rulings and the Social Security Code. In response to the existence of this inconsistency and its extension, several reports proposed changes in the legislation – an institutional work of advocacy of external origin (no. 2) but also internal origin (no. 2a) directed at the parliamentarians (reports issued by IGAS inspectors, members of parliament and senators). This work was unsuccessful: no law was adopted to change the Social Security Code. Meanwhile, the Cour de Cassation began a work of advocacy directed at the Conseil Constitutionnel (no. 3), and this was successful. In creating a reservation regarding interpretation of the article of the Social Security Code capping compensation, the Conseil Constitutionnel “unlocked” the list of prejudices eligible for compensation. This decision was a form of political institutional work of vesting, equivalent to the power of a law (no. 4). Another form of advocacy was also conducted in parallel towards parliamentarians (no. 4a), such that the legal inconsistency became a genuine legal uncertainty. Also, the Cour de Cassation was able gradually to accept new M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 166 Table 4. Actors of the regulative pillar and institutional work during the third period in the history of faute inexcusable Actors Actions Institutional work concerned Consequences Cour de Cassation (1) Court rulings (since 2002) Defining Reorganisation and extension of the principle of faute inexcusable State bodies (2) Publication of reports Advocacy (failed) Arguments in favour of a reform of compensation for occupational accident/illness on the principle of full compensation Parliamentarians (2 a) Publication of commission reports (2005; 2006) Advocacy (failed) Recommendation of the changes in the system for reparation and compensation for professional risks to adapt it to society (with no legislative effect) Cour de Cassation (3) Application to the Conseil Constitutionnel (May 2010) Advocacy Request for examination of the constitutionality of legal provisions for compensation of victims in cases of faute inexcusable Conseil Constitutionnel (4) Decision of 18 June 2010 Vesting Creation of a reservation regarding interpretation concerning the list of prejudices to be compensatedConseil Constitutionnel (4 a) Decision of 18 June 2010 Advocacy (failed) Creation of a reservation regarding interpretation concerning the list of prejudices to be compensated Cour de Cassation (5) Court rulings (since 2010) Defining Recognition of new prejudices (towards full reparation) Cour de Cassation (6) Report of 2010 Advocacy (failed) Recommendation for a change in the law to make it more favourable to full compensation Parliamentarians (7) Presentation of a bill of law – and its rejection (November 2011) Advocacy (failed) Examination of the bill of law by the National Assembly in November 2011 – rejected (continuation of compensation on a case-by-case basis) prejudices eligible for compensation, thus continuing its political institutional work of defining (no 5). Finally, in its 2010 report, the Cour de Cassation proposed a change in the law, in a new form of advocacy directly targeting parliamentarians (no. 6). And yet the rejection of a corresponding bill of law, which is a form of failure in an institutional work of advocacy internal to the parliamentarians (no. 7), perpetuated this unstable situation internal to the regulative pillar. Figure 7. Representation of changes in the regulative pillar during the third period in the history of faute inexcusable ANALYTICAL SUMMARY OF RESULTS ! The history of faute inexcusable has enabled us to highlight, for each of the three major periods studied, the actors who participated in the life of the regulative pillar and the type of institutional work undertaken. Putting these three periods into perspective offers a different conception of the regulative pillar, the actors it comprises and the dynamics of change in their political institutional work. A DIFFERENT CONCEPTION OF THE REGULATIVE PILLAR ! In the three periods identified, we distinguish several categories of coercive texts, formal rules that are sources of the regulative pillar’s influence: the laws, The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 167 court rulings and decisions by the Conseil Constitutionnel. The regulative pillar is thus more than just the law. One of the facets of its complexity is that it is in fact made up of a configuration of interdependent texts. These texts are interdependent in the sense that the court rulings clarify the provisions of law, and the Conseil Constitutionnel decisions cancel or uphold certain provisions of law. These texts form a moveable configuration, as court rulings and the law display varying degrees of consistency (they are most consistent during the second period, and least consistent during the other two periods). Finally, the regulative period is “flexible”: for a constant legislative scope (i.e. with no change in the law), court rulings may change and restrict, or alternatively extend the conditions for application of the laws (as happened at the end of the second period). As a result, the study of the regulative pillar must not be based solely on changes in law, but require (1) identification of repeal, creation or substantial amendments to any text in the configuration specific to the pillar at a given point in its history and (2) analysis of the effects of this change on the overall configuration of the texts that are a source of coercion for the actors in the institutional field. ACTORS OF THE REGULATIVE PILLAR ! In each period studied, we find several actors who influence (or attempt to influence) production of these texts: employers and employees – including the victims of occupational accident/illness, parliamentarians (European and French), the Cour de Cassation, the Conseil Constitutionnel, the lawyers for victims of occupational accident/illness, the unions, the Cour des Comptes, the IGAS and the CSPRP. Given their relation and the power of their initiatives, we put these actors into three groups for a more nuanced representation of the regulative pillar than the literature has indicated so far (see figure 8). Figure 8. Two representations of the actors of the regulative pillar: (1) in the literature and (2) resulting from the faute inexcusable case study ! A first group, consisting of actors in the “legal and court system”, is characterised by the capacity ultimately to produce coercive texts, even if the M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 168 production of texts is, strictly speaking, concentrated in a small number of actors in this category (see above). We identify two subcategories of actors in this group. First, the “Legal professionals30”: the Cour de Cassation capable of a political institutional work of defining, the Conseil Constitutionnel which holds a power of vesting, and the lawyers, whose role here is to defend the victims by pleading their case to the judges (advocacy). The second and third periods indicate that the work of advocacy may be internal to this subgroup: advocacy by lawyers to the Cour de Cassation, and by the Cour de Cassation to the Conseil Constitutionnel. But these actors can also carry out advocacy work directed at another subgroup: actors so far treated as part of “the State”: the parliamentarians who, through the National Assembly and the Senate, draw up reports (and thus conduct an internal work of advocacy), propose and vote on laws that form a political institutional work of vesting31. We call this subcategory the “Public authorities32”. ! A second group named “Administration” participates in the life of the institution through intervention by State bodies such as the Cour des Comptes or the IGAS. Through reports they provide information on the state of the current legislative arrangements in force. Their role is thus to influence the actors in the legal and court systems (institutional work of advocacy), particularly by helping to import into the regulative pillar the points of misalignment of the coercive texts with the ideas of the cognitive pillar (as the second period illustrates). Finally, a third group consists of actors from “Civil society”. Contrary to the first group, they do not intervene directly in production of the coercive texts and are the targets or actors primarily concerned by these laws: as such, they are not so much actors of the regulative pillar as other actors in the institutional field. They are employers and employees, including victims of occupational accident/illness, but also employers’ and employees’ unions. Their power of influence is the power of advocacy, principally towards parliamentarians or judges, via the lawyers defending employers or employees. TOWARDS A GENERIC REGULATIVE DYNAMIC ! In addition to the more complex conception of the regulative pillar and its actors just presented, analysis of changes in the regulative pillar during the three periods studied makes it possible to identify a generic “regulative dynamic” of three successive phases: (1) The emergence of a triggering event, (2) Attempts at influence and (3) Change in the configuration of coercive texts. (1) Emergence of a triggering event: in the three periods analysed, the regulative dynamics leading to a change in the regulative pillar initially result from a triggering event. We noted triggering events both “external” and “internal” to the regulative pillar. The regulative pillar can change following a misalignment with other institutional pillars; this is a source of change that is “external” to the regulative pillar. In the first period, this is the phenomenon that gave rise to the law of 1898. In the second period, the inconsistency in compensation regimes creating unfair situations also illustrates this phenomenon. We also identified two other dynamics that are “internal” to the regulative pillar: (a) the vagueness of the conditions for application of the laws and (b) the presence of inconsistencies between the texts of the pillar. First (a), as we observed during the first period, the interpretation of obligations created by the law depended on the degree of detail provided. In fact, the vagueness of the legal definition of faute inexcusable caused significant uncertainty regarding the situations in which it could apply. An internal dynamic then came into play to reduce this uncertainty. Also (b), as noted in the third period, the existence of inconsistency within the configuration of texts making up the regulative pillar was to lead to a dynamic aiming to secure consistency in this pillar33. The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 169 30. Corresponding to the professional family of “Legal professionals” comprising lawyers, notaries, legal advisers (apart from corporate lawyers), plus bailiffs and magistrates. 31. We can also include the European Union, which issues directives that must be transposed into French law to take effect – they thus apply via the law, decrees, etc. The government and parliament still have some margin for interpretation of these directives. 32. The public authorities comprise State bodies of the Presidency, the government and ministries, the National Assembly and the Senate, whose role is to produce and ensure compliance with the Law – i.e. the bodies invested with legislative and executive powers. 33. However, this outcome was not observed, as this period is still ongoing. ! (2) Attempts at influence: this second phase is principally characterised by the organisation of an institutional work of advocacy that can come from all actors making up the regulative pillar34. These attempts at influence may succeed or fail. In the case of failure, the configuration of coercive texts that make up the regulative pillar remains unchanged and it is perpetuated (for example, when the bill of law presented by parliamentarians was rejected); in the case of success, one or more of the actors in the legal and court system are influenced and respond by changing the configuration of the texts in the pillar. The regulative pillar continues unchanged from phases 1 and 2 (i.e. as long as there is no change in the configuration of texts making up one state of the pillar at a given point in its history). The change in the pillar forms phase 3 in the dynamic. (3) The change in the configuration of coercive texts: in this third phase, one of the actors in the legal and court system with the capacity to issue a text (in our case Parliamentarians, Cour de Cassation and the Conseil Constitutionnel), influenced by a work of advocacy, implements a political institutional work. Depending on the nature of the text produced, this work will be classified as vesting (for example, following the Conseil Constitutionnel’s decision of 2010) or defining (for example, following the rulings by the Cour de Cassation of 1941 and 2002). The production of a text would both close the current regulative dynamic and open up a potential new dynamic. In fact phase 3, which ends one regulative dynamic, is succeeded by another phase 1 of a new regulative dynamic. Note that while this political institutional work of vesting or defining creates an inconsistency or uncertainty in the configuration of texts (case of the two internal triggering events), phase 1 of the new regulative dynamic that follows this phase 3 will soon lead to another phase 2. For example, in the third period studied, the political institutional work of vesting by the Conseil Constitutionnel is a phase 3, and simultaneously an amplification of the legal uncertainty that was the source of the decision. It is thus the founding act of the constitution of a triggering event (phase 1 of a new dynamic), an uncertainty leading once again to institutional work of advocacy (phase 2 – in the form of a report by the Cour de Cassation and the filing of a bill of law) that was not followed by an institutional work of vesting. From this point of view, the third period is formed of two separate generic dynamics, the second of which is still unfinished and is currently in its phase35 2. ! The three phases identified form a generic dynamic which, once completed, continues in a further generic dynamic. One exception to this sequence is possible: when the regulative pillar, or more broadly the institution, is destroyed (for example in the potential case of a text repealing a legal provision supporting an institution, such as the abolition of the death penalty in France in 1981). The principal factors in this proposal of a generic regulative dynamic are summarised in figure 9. M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 170 34. This excludes actors of civil society who, since they are not actors of the regulative pillar, can only go though the actors of the pillar to engage in advocacy work to defend their interests. 35. These successive generic dynamics can be grouped into a single period for temporal bracketing (Langley, 1999). As the scale of the change brought about by the Conseil Constitutionnel, since it only intensifies the ore- existing legal inconsistencies, does not have the capacity to trigger any real change in the context that would mark the start of a new period. Figure 9. Proposal for a generic regulative dynamic ! We now discuss the implications of these results for the neo-institutional literature. DISCUSSION CONTRIBUTIONS TO UNDERSTANDING OF THE REGULATIVE PILLAR: COMPLEXITY AND REVELATION OF ACTORS ! Our first contribution is to reveal actors previously left out of analysis of the regulative pillar. These actors form a group covering the law and the courts. All of them have statutory political competence that confers automatic legitimacy in the field36. Therefore, contrary to the suggestion of Perkmann and Spicer (2008), they do not need to acquire “political skills” they already hold by statute in order to participate in institutional struggles. The actors external to the regulative pillar, who do not hold these skills, are the people who, in order to exercise a political institutional work, must either try to acquire those skills or call on actors who hold them. ! We also complement the research by Perkmann and Spicer (2007; 2008) by proposing a more dynamic reading of interactions between the actors of the field. This clarifies the interaction between the actors of civil society and the actors of the legal and court world – especially through “mercenary” lawyers (McCann & Silverstein, 1998) who sell their statutory political skills. The literature expects the advocacy activity exercised by the actors of civil society to directly target certain actors – such as the parliamentarians; we show that they can also use intermediaries who have special access to the actors who produce the texts. Furthermore, we stress that the same actor can draw on different forms of institutional work, depending on his desired aim. While Perkmann and Spicer (2007, 2008) highlight the different types of political institutional work and associate specific actors with them, we underline the greater complexity of this pillar, in which different actors can use different forms of political institutional work according to the interests they are pursuing. ! The political action of actors in civil society has already been studied (Stinchcombe, 1965): the main political strategies highlighted are diplomacy by governments (Djelic, 1998) or industrialists (Lounsbury & Crumley, 2007), and The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 171 36. National bar qualifications for lawyers, graduation from the National Magistrates’ School for magistrates, elections for members of parliament and appointments within the administration. negotiations or coalitions (Botzem & Quack, 2006) to influence legislative life. The inclusion of actors from the legal and court sphere explains new forms of lobbying. Actors with no particular political skills may either appeal directly to the public authorities to change the laws in their favour, or develop a legal strategy relying on lawyers to try and orient rulings in line with their interests. This dual operation – also accessible to employees who are victims of occupational accident/illness – shows that court action is not only a defence reaction, since it is also a form of lobbying aiming to secure a stakeholder’s interests in the law. HIGHLIGHTING A REGULATIVE DYNAMIC ! The traditional interpretation of institutional change is driven by two perceptions: on one hand, the regulative pillar is perceived as the trigger of institutional change (the law acts as an exogenous shock, Greenwood, et al., 2008); on the other hand, regulative factors mark the de-institutionalisation of the institutionalised practice (the law puts an end to a previous institutional order, Hiatt, Sine & Tolbert, 2009; Maguire & Hardy, 2009). Extending the work of Blanc and Huault (2010) and Dansou and Langley (2012), our work shows that this interpretation ignores the underlying institutional struggles exercised outside the boundaries of the institution’s life cycle. This brings out the existence of a regulative dynamic which, by modifying various categories of coercive texts, contributes to change in institutionalised practice in the same way as cognitive and normative factors. We use the term regulative dynamic to mean a process specific to the combination of the three forms of political institutional work undertaken in the regulative pillar. This dynamic can be modelled as a sequence that carries a notable change in the regulative pillar and is characterised by the succession of three phases. Several sequences, of variable duration and intensity, may follow each other throughout the institution’s life cycle. This dynamic reveals the power of actors who contribute to shape, incrementally, the institutionalised practice by their political institutional work. (1) First phase: Emergence of a triggering event  ! This first phase enables us to confirm and complement the current literature on the emergence of a phase of institutional change. While the regulative pillar evolves following the gradual formation of a misalignment between it and the cognitive and/or normative pillars (Caronna, 2004), we show a second source of institutional change resulting from the intrinsic operation of the pillar and the configuration of the texts (laws and court rulings) that carry obligations. Also, this change is part of a succession of regulative dynamics: the internal or external misalignment that brings change is the direct consequence of previous dynamics (the uncertainties or insecurities originate in coercive texts produced at the end of the previous regulative dynamics). We thus contribute to the literature on institutional change by documenting a second source of change due to an internal shift in the pillar. (2) Second phase: attempts at influence. ! In this second phase, the nature of the advocacy work responds to the issues of phase 1. It can be carried out by one group of actors towards another (for example, administration group actors can draw up reports recommending changes in the texts intended for actors in the legal and court system) or within the same group of actors (for example, one group of parliamentarians can attempt to influence other parliamentarians by presenting a bill of law). This work of advocacy reveals the actors’ strategy and their power within the institution. What can happen is that the pillar’s unstable balance is maintained: attempts at influence to achieve a new balance fail. Behind the apparent status quo, maintenance of the regulative pillar is the setting for institutional struggles. M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 172 To achieve real change, the actors in the legal and court system, influenced by the previous phase, decide to implement defining or vesting work with the aim of modifying coercive texts. (3) Third phase: change in the configuration of coercive texts ! After a successful work of advocacy, the actors implement a political institutional work of vesting or defining. The consequence of this political work is the amendment of a text creating obligations or prohibitions, or a redefinition of its scope of application. This is a response to the triggering event (alignment of the pillar with the other pillars; reducing the uncertainty or inconsistency inside the pillar). ! The revelation of a regulative dynamic offers perspectives for understanding the nature of the sequence of political institutional work in its various forms. CONTRIBUTIONS TO UNDERSTANDING OF INSTITUTIONAL CHANGE ! Our analysis in terms of regulative dynamic places this research in the neo-institutional research agenda for understanding institutional change. As Dansou and Langley (2012) suggest, the study of institutional change in its continuity made it possible to underline the interactions between different forms of institutional work. The institutional dynamic clarifies the interaction between advocacy, defining and vesting. This new look at the regulative pillar emphasises the relevance of not focusing solely on crisis periods (Zietsma & Lawrence, 2010): the institutional maintenance period can be interpreted as a failure of the advocacy work undertaken by actors seeking a change in practice. Behind the apparent maintenance, actors are constantly defending their interests (Blanc & Huault, 2010) - and the actors in the legal and court sphere have plenty of them. They seize on the form of institutional work that will enable them to achieve their objectives and interact with actors from civil society if necessary. And so the regulative pillar is not an entirely closed world, and can be a setting for institutional work challenging the institutionalised practice. One exception to this sequence can be envisaged: when the regulative pillar, or the institution more broadly, is destroyed (for example, in the potential case of a text repealing a provision of law that supports an institution, such as abolition of the death penalty in France in 1981). De-institutionalisation can thus be explained by a total disruption of the regulative pillar. LIMITATIONS AND AVENUES FOR FURTHER RESEARCH ! This research has shown the complexity and dynamic nature of the regulative pillar, but it is still necessary to examine its limitations and propose avenues for further research. ! First, our research is based on a single case study that emphasises secondary sources from the law, legal professionals and the courts, choosing a methodology that provides validity of a “theoretical” rather than statistical nature (Miles & Huberman, 2002:62). Also, in view of this institutional form and our research question, we focused on sources from the law, legal professionals and the courts. It was not possible to observe or interview all the actors who helped to shape the long history of faute inexcusable. ! To achieve generalisation and confirmation, we can only encourage our fellow researchers to continue this work by further exploration of the role of other types of actor and other areas of law (for example commercial law). Another interesting way to extend this research would be to study another major source of French law: collective agreements. Collective agreements specific to certain The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 173 sectors and other types of company agreements are the sources of law that are developed closest to the company by employee and employer representatives, and complement the other sources of law that are the laws and official regulations. Longitudinal case studies in organisations that have begun negotiations of this kind would enhance our understanding of the way this regulative dynamic is, within certain boundaries, coproduced (not simply undergone) by organisations due to social dialogue. Also, still with a view to casting further light on the complexity of the dynamics of institutional change, this future research could take a closer look at the pattern of the regulative dynamic, in other words the moments of calm, pause, resumption and intensification of political institutional work specific to each of its phases – and factors that explain these temporalities. ! At international level, other case studies would be useful to explore the variety of legal and court systems and identify, through comparison, the recurrences and divergences of actors and regulative dynamics they initiate. As Lejeune (2011:229) points out, such case studies require “consideration of national and local specificities, namely the legal culture, the relations between legal professionals with the public authorities and the various types of State”. Finally, the new appreciation of the regulative pillar provided here opens up the possibility of a new analysis of relations between the institutional pillars. We suggest these relations would be studied using the idea of alignment between pillars (Caronna, 2004; Maguire & Hardy, 2009). A complementary study could look in more depth at the role the regulative pillar plays, through a complex regulative dynamic, in the modalities of alignment and/or misalignment between the three pillars, which is a source of cohesion or institutional change. Between two laws, court rulings are a source of gradual adjustments between the dimensions of the institution. The turnaround in court rulings in 2002 was a response to cognitive shifts (it echoes the asbestos scandal and the broadly shared observation that compensation systems were unfair) and normative shifts (it is part of a redefinition of the employer’s responsibility as regards occupational risks). These results suggest that the relationships between the three institutional pillars are more dynamic than the literature shows. The regulative pillar is capable of evolving without any change in the law, and can for example soften or accentuate the misalignments between pillars and show a more detailed process of institutional change or maintenance, which of course cannot happen if the regulative pillar is confined to the State and the law. The case of the death penalty in France in 1981 could be an interesting field of study to this end. This case offers factors of a cognitive nature (such as the question of the power of life and death), normative nature (execution methods) and of course regulative nature (the type of penalties depending on the crime). Also, the case studied here appears to highlight a regulative pillar that is “ahead” of the cognitive and normative pillar: the regulative dynamic was proactive, and brought the other pillars to evolve; this conclusion is contrary to the current interpretation of the role of this pillar, and confirms the relevance, defended in this study, of looking differently at the regulative pillar. REFERENCES Agrikolansky, E. (2010). Les usages protestataires du droit. In O. Fillieule et al. (Eds) Penser les mouvements sociaux. Conflits sociaux et contestations dans les sociétés contemporaines (pp. 225-245). Paris : Editions la Découverte. Barley, S.R. & Tolbert, P.S. (1997). Institutionalization and Structuration: Studying the Links Between Action and Institution, Organization Studies, 18(1), 93-117 Baron, J.N., Dobbin, F.R. & Jennings, P.D. (1986). War and Peace: the evolution of modern personnel administration in U.S. industry. American Journal of Sociology, 92(2), 350-383. Ben Slimane, K., & Leca, B. (2010). Le travail institutionnel : origines théoriques, défis et perspectives. Management & Avenir, 37,53-69. M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 174 Benford, R.D. & Snow, D.A. (2000). Framing processes and social movements: an overview and assessment. Annual Review of Sociology, 26: 11-39 Blanc, A., & Huault, I. (2010). Reproduction de l'ordre institutionnel face à l'incertitude. Revue Française de Gestion, 203(4), 85-99. Botzem, S. & Quack, S. (2006). Contested rules and shifting boundaries: International standard setting in accounting. In M.-L. Djelic & K. Sahlin-Andersson (Eds), Transnational governance: Institutional dynamics of regulation. (pp. 266-286). Cambridge: Cambridge University Press. Caronna, C. A. (2004). The Misalignment of Institutional “Pillars”: Consequences for the US Health Field. Journal of Health and Social Behavior, 45(Extra Issue), 45-58. Cole, R.E. (1985). The macropolitics of organizational change : A comparative analysis of the spread of small- group activities. Administrative Science Quarterly, 30(4), 560-585. Dansou, K., & Langley, A. (2012). Institutional Work and the Notion of Test. M@n@gement, 15(5), 502-527. Delacour, H., & Leca, B. (2011). Grandeur et décadence du Salon de Paris : une étude du processus de désinstitutionnalisation d’un événement configurateur de champ dans les activités culturelles. M@n@gement, 14(1), 47-78. 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. Djelic, M.-L. (1998). Exporting the American model: The post-war transformation of European business. Oxford: Oxford University Press. Edelman, L. B. (1992). Legal Ambiguity and Symbolic Structures: Organizational Mediation of Civil Rights Law. American Journal of Sociology, 97(6), 1531-1576. Farjoun, M. (2002). The Dialectics Of Institutional Development in Emerging and Turbulent Fields: The History of Pricing Conventions in the On-Line Database Industry. Academy of Management Journal, 45(5), 848-874. Fligstein, N. (2001). Social Skill and the Theory of Fields. Sociological Theory, 19(2),105-125. Gombault, A. (2006). La méthode des cas. In P. Roussel & F. Wacheux (Eds.) Management des ressources humaines. Méthodes de recherche en sciences humaines et sociales (pp. 31-64). Paris: de Boeck. Greenwood, R., Oliver, C., Sahlin, K., & Suddaby, R. (2008). Introduction in. R. Greenwood, C. Oliver, K. Sahlin, & R. Suddaby, (Eds). Handbook of Organizational Institutionalism, (pp. 1-46), Sage: London. Hargadon, A.B., & Douglas, Y. (2001). When Innovations Meet Institutions: Edison and the Design of the Electric Light. Administrative Science Quarterly, 46(3), 476-501. Hauriou, M. (1925). Théorie de l'institution et de la fondation. Essai de vitalisme social. In. M. Hauriou (Ed.) Aux sources du droit : le pouvoir, l'ordre et la liberté. Cahiers de la Nouvelle Journée, N°23, Paris. Hiatt S.R., Sine W.D., & Tolbert P.S. (2009). From Pabst to Pepsi : The Deinstitutionalization of Social Practices and the Creation of Entrepreneurial Opportunities, Administrative Science Quarterly, 54(4), 635-667. Hoffman, A. (1999). Institutional Evolution and Change: Environmentalism and the U.S. Chemical Industry. Academy of Management Journal, 42(4), 351-371. Jepperson, R.L. (1991). Institutions, Institutional Effects, and Institutionalism. In W.W. Powell, P.J. DiMaggio (eds.) The New Institutionalism in Organizational Analysis (pp. 143-163), Chicago, IL: University of Chicago Press. Kitchener, M. (2002). Mobilizing the Logic of Managerialism in Professional Fields: The Case of Academic Health Center Mergers. Organization Studies, 23(3), 391-420. Krasner, S.D. (1984). Approaches to the state: Alternative conceptions and historical dynamics. Comparative politics, 16(2), 223-246. Langley, A. (1999). Strategies for Theorizing from Process Data. Academy of Management Review, 24(4), 691-710. Lawrence, T. B., & Suddaby, R. (2006). Institutions and Institutional Work. In S. R. Clegg, C. Hardy, T. B. Lawrence, & W. R. Nord (Eds.), Handbook of organization Studies, 2nd Edition (pp. 215-254). London: Sage. Lawrence, T. B., Suddaby, R., & Leca, B. (2009). Introduction: theorizing and studying institutional work. In T. B. Lawrence, R. Suddaby & B. Leca (Eds.), Institutional Work: Actors and Agency in Institutional Studies of Organizations (pp. 1-27). Cambridge: Cambridge University Press. Leblebici, H., Salacncik, G.R., Copay, A., & King, T. (1991). Institutional Change and the Transformation of Interorganizational Fields: an Organizational History of the U.S. Radio broadcasting Industry, Administrative Science Quarterly, 36(3), 333-363. Lejeune, A. (2011). Les professionnels du droit comme acteurs du politique : revue critique de la littérature nord-américaine et enjeux pour une importation en Europe continentale. Sociologie du Travail, 53(2), 216-233. Lounsbury, M. & Crumley, E.T. (2007) New practice creation: An institutional perspective on innovation, Organization Studies, 28(7), 993–1012. Maguire, S., & Hardy, C. (2009). Discourse and Deinstitutionalization: the Decline of DDT, Academy of Management Journal, 52(1), 148-178. McCann, M.W. (1994). Rights at Work :Pay Equity and the Politics of Legal Mobilization. Chicago, IL: University of Chicago Press. McCann, M., & Silverstein, H. (1998). Rethinking law’s ‘allurements’. A relational analysis of social movement lawyers in the United States. In A. Sarat, S. Scheingold (Eds.), Cause Lawyering. Political Commitments and Professional Responsibilities (pp. 261-292). Oxford: Oxford University Press. Meyer, J. W., & Rowan, B. (1977). Institutional Organizations: Formal Structure as Myth and Ceremony. American Journal of Sociology, 83(2), 340-363. Miles, M.B., & Huberman, M.A. (2002). Analyse des données qualitatives. Bruxelles: De Boeck. Milgrom, P. & Roberts, J. (1992). Economics, Organization and Management, Englewood Cliffs, NJ: Prentice Hall. Millar, E. (1995). Hauriou et la théorie de l’institution. Droit & Société, 30/31(1), 381-412. The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 175 North, D. C. (1990). Institutions, institutional change, and economic performance, Cambridge University Press, MA: Cambridge. Oliver, C. (1992). The Antecedents of Deinstitutionalization. Organization Studies, 13(4), 563-588. Patton, M.Q. (2002). Qualitative Research & Evaluation Methods (3rd edition). Thousand Oaks, CA: Sage. Perkmann, M., & Spicer, A. (2007). 'Healing the Scars of History': Projects, Skills and Field Strategies in Institutional Entrepreneurship. Organizaion Studies, 28(7), 1101-1122. Perkmann, M., & Spicer, A. (2008). How are Management Fashions Institutionalized? The role of Institutional Work. Human Relations, 61(6), 811-844. Pires, A. (1997). Echantillonnage et recherche qualitative. In Poupart, J., Deslauriers, J.-P., Groulx, L.-H., Laperrière, A., Mayer, R., Pires, A. (Eds). La recherche qualitative. Enjeux épistémologiques et méthodologiques (pp. 113-169). Montréal: Gaëtan Morin. Prost, A. (1996). Douze leçons sur l'histoire. Paris: Points Seuil. Russo, M.V. (2001). Institutions, Exchange Relations, and the Emergence of New Fields: Regulatory Policies and Independent Power Production in America, 1978-1992, Administrative Science Quarterly, 46(1), 57-86. Scott, W. R. (2008). Institutions and Organizations, 3rd Edition. Thousand Oaks, CA: Sage. Stinchcombe, A. (1965). Organization and Social Structure. In March J.G. (Ed.). Handbook of Organization (pp. 141-193). Chicago, IL: Rand McNally. Skocpol, T. (1985). Bringing the State Back In: Strategies for Analysis in Current Research. In P. Evans, D Rueschemeyer, & T. Skocpol (Eds.), Bringing the State Back In (pp. 3-43). NY: Cambridge University Press. Taupin, B. (2012). The More Things Change…Institutional Maintenance as Justification Work in the Credit Rating Industry. M@n@gement, 15(5), 528-562. Tolbert, P.S., & Zucker, L.G. (1983). Institutional Sources of Change in the Formal Structure of Organizations: The Diffusion of Civil Service Reform, 1880-1935. Administrative Science Quarterly, 28(1), 22-39. Townley, B. (2002). The role of competing rationalities in institutional change. Academy of Management Journal, 45(1), 163-179. Yin, R.K. (2009). Case Study Research (4th edition). Thousand Oaks, CA: Sage. Zietsma C., & Lawrence T.B. (2010). Institutional Work in the Transformation of an Organizational Field: the Interplay of Boundary Work and Practice Work. Administrative Science Quarterly, 55(2), 189–221. M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 176 APPENDIX A. REFERENCES OF DATA SOURCES •Association d’Assistance et d’Entr’Aide (1949). La Faute inexcusable en matière d'accidents du travail, Imprimerie Centrale de l’Ouest : La Roche sur Yon. •Belhache, C. et Belhache, S. (1999). La faute inexcusable de l’employeur, Sofiac : Paris. •Bienvenu, J. (1938). La Faute inexcusable dans la législation des accidents du travail, des maladies professionnelles et des inscrits maritimes (Thèse de droit). Imprimeries Editions M. Vilaire : Le Mans. •Boisselier, J. (2008). Naissance et évolution de l’idée de prévention des risques professionnels, 2nde édition, INRS, ED 926 : Paris. •Buzzi, S., Devinck, J. C. & Rosental, P. A. (2006). La santé au travail. 1880-2006, Paris: La Découverte. •Cass. Req., 22 février 1932, D.P. n° 1932.1.25, note A. Rouast. •Cass. rec., SIREY, 4 mai 1937, n° 1931.1.331 – R.G.A.T. 1937.1010. •Cass. civ., 15 février 1938, D.H. n° 1938.261. •Cass. chambres réunies, 15 juillet 1941, Dame Veuve Villa, JCP 1/41, n° 705. •Cass. assemblée plénière, 18 juillet 1980, n° 78-12570. •Cass. soc., 28 février 2002, n° 99-18.389, Sté Eternit industries c/ Delcourt- Marousez et a. (voir également les autres arrêts rendus le même jour par la Chambre sociale de la Cour de cassation) •Cass. soc., 11 avril 2002, n° 00-16535, Hachadi c/ Sté Camus Industrie et a. •Cass. soc., 19 décembre 2002, n° 01-20.447, dame Hervé c/ CPAM Angers et a. •Cass. soc., 29 juin 2005, no°03-44.412, no 1698, Sté Acme Protection c/ Lefebvre. •Cass. soc., 28 février 2006, n° 05-41.555, no 835, Deprez c/ Sté Cubit France technologies : Bull. civ. V, no 87. •Cass. soc., 5 mars 2008, n° 06-45.888, no 429, Sté Snecma c/ Synd. CGT Snecma Gennevilliers. •Cass. soc., 11 mai 2010, n° 09-42.241, no 939, Sté Ahlstrom Labelpack c/ Ardilley et a. •Chapouthier, A. (2009). La faute inexcusable de l’employeur en santé et sécurité du travail, Travail et Sécurité, 695, 48-49. •Conseil Constitutionnel., 18 juin 2010, no 2010-8 QPC •Cour de cassation. (2002). Rapport annuel 2002. La responsabilité. La Documentation Française : Paris. •Cour de cassation. (2007). Interprétation et portée des arrêts de la Cour de cassation en matière civile. Bulletin d’information de la Cour de cassation, 661, 6-21. •Cour de cassation. (2010). Présentation de la Cour de cassation, http:// www.courdecassation.fr/institution_1/savoir_plus_institution_2845/ presentation_cour_cassation_11982.html, consulté le 22/11/2010. •Cour des comptes. (2002). La gestion du risque accidents du travail et maladies professionnelles. Rapport au Président de la République suivi des réponses des administrations et des organismes intéressés, février, 276 p. •Dériot, G. & Godefroy J. P. (2005). Le drame de l'amiante en France : comprendre, mieux réparer, en tirer des leçons pour l'avenir. Rapport d'information n° 37 (2005-2006) fait au nom de la mission commune d'information du Sénat, déposé le 26 octobre, 333 p. •Douard, E. (1961). La Faute inexcusable dans le régime de sécurité sociale, Les Editions Sociales Françaises : Paris. •Editions Législatives. (2010). Etude Santé et sécurité au travail, Dictionnaire Permanent Social. The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 177 http://www.courdecassation.fr/institution_1/savoir_plus_institution_2845/presentation_cour_cassation_11982.html http://www.courdecassation.fr/institution_1/savoir_plus_institution_2845/presentation_cour_cassation_11982.html http://www.courdecassation.fr/institution_1/savoir_plus_institution_2845/presentation_cour_cassation_11982.html http://www.courdecassation.fr/institution_1/savoir_plus_institution_2845/presentation_cour_cassation_11982.html http://www.courdecassation.fr/institution_1/savoir_plus_institution_2845/presentation_cour_cassation_11982.html http://www.courdecassation.fr/institution_1/savoir_plus_institution_2845/presentation_cour_cassation_11982.html •Graser, M., Manaouil, C. & Jardé, O. (2004). La faute inexcusable de l'employeur dans la réparation des accidents de travail et des maladies professionnelles. Médecine & Droit, 69, 133-141. •Guillemy, N. (2006). La nouvelle approche réglementaire pose le principe d’une obligation générale de sécurité. Réalité prévention, 12, 4-5. •Jaillet, R. (1980). La faute inexcusable en matière d’accident du travail et de maladie professionnelle, Librairie Générale de Droit et de Jurisprudence : Paris. •Laroque, M. (2004). La rénovation de la réparation des accidents du travail et des maladies professionnelles. Rapport remis à M. François Fillon, Ministre des Affaires Sociales, du Travail et de la Solidarité, no 2004 032, mars, 84 p. •Le Garrec, J. & Lemière, J. (2006). Rapport n° 2884 fait au nom de la mission d’information sur les risques et les conséquences de l’exposition à l’amiante. Enregistré à la Présidence de l'Assemblée nationale le 22 février (Tome 1), 564 p. •Le Roux, D. (2010). Le coût de la faute inexcusable va s'alourdir pour les entreprises. ActuEL-HSE, 2 juillet. •Machu, L. (2009). Entre prévention et réparation : les syndicats ouvriers face à la question des risques au travail pendant l’entre-deux guerres. In C. Omnès, et al. (Eds.), Cultures du risque au travail et pratiques de prévention. La France au regard des pays voisins (pp. 189-201). Rennes: Presses Universitaires de Rennes. •Masse, R. et Zeggar, H. (2001). Réflexions et propositions relatives à la réparation intégrale des accidents du travail et des maladies professionnelles. Rapport à l’intention de Madame Elisabeth Guigou, Ministre de l’Emploi et de la Solidarité, 26 juin, 50 p. •Pic, P. et Kréher, J. (1938). La Faute inexcusable en matière d'accident du travail, Imprimerie J. Verdier : Saint Etienne. •Pinaud, P. (2002). Faute inexcusable de l'employeur en matière d'accidents du travail et de maladies professionnelles : portée et conséquences financières des arrêts de la Cour de cassation du 28 février 2002, CERAL : Lyon. •Ray, J. E. (2010). Droit du travail, droit vivant 2010-2011. 19e édition actualisée et augmentée, Paris: Editions Liaisons. •Rouast, A. et Durand, P. (1953). Précis de Législation Industrielle (droit du travail), Librairie Dalloz : Paris. •Rouast, A., Durand, P. et Dupeyroux, J.J. (1962). Sécurité Sociale, Dalloz : Paris. •Rosental, P. A. (2008). La notion de "risques professionnels". Système actuel et exemples. In J.-M. Mur (Eds.), L'émergence des risques (pp. 19-41). Paris: INRS. •Sargos, P. (2003). L’évolution du concept de sécurité au travail et ses conséquences en matière de responsabilité. La semaine juridique, édition générale, 4, 121-128. •Sargos, P. (2006). Le souci du juge est d’assurer l’effectivité du droit dans une perspective de prévention. Réalité Prévention, 12, 2-3. •Seillan, H. (2009). De la fatalité à la prévention. Hygiène, sécurité et santé au travail. Plus d’un siècle d’évolution. Conférence organisée par le Groupe Régional d’Ile de France du Comité d’Histoire de l’Administration du Travail, de l’Emploi et de la Formation Professionnelle, 14 octobre, (pp. 6-14). •Teissonnière, J. P. & Topaloff, S. (2002). L’affaire de l’amiante, Semaine sociale Lamy, 1082, 2-33. •Vidalies, A. (2011). Rapport n° 3922 fait au nom de la commission des affaires sociales sur la proposition de loi relative à l’amélioration de l’indemnisation des victimes d’accidents du travail et de maladies professionnelles. Enregistré à la Présidence de l'Assemblée nationale le 9 novembre, 75 p. M@n@gement, 17(3): 145-179! Hélène Peton � Stéphan Pezé 178 •Vingiano, I. (2011). L’incidence de la jurisprudence sur la garantie et l’indemnisation de la faute inexcusable de l’employeur. Aix-en-Provence: Presses Universitaires d’Aix-Marseille. •Yahiel, M. (2002). Vers la réparation intégrale des accidents du travail et des maladies professionnelles : éléments de méthode. Rapport à l’intention de Madame Elisabeth Guigou, Ministre de l’Emploi et de la Solidarité, 26 avril, 55 p. Hélène Peton is a Lecturer (Maître de Conférences) at Université Paris-Est Créteil and a member of the Institut de Recherche en Gestion (IRG, EA 2354). Her research concerns the role of actors in institutional change, mainly in cases of major controversies. Stéphan Pezé is a Lecturer (Maître de Conférences) at Université Paris-Est Créteil and a member of the Institut de Recherche en Gestion (IRG, EA 2354). His research concerns the construction of identity, occupational health and managerial situations and trials. Acknowledgements. We wish to thank Emmanuel Josserand and the two anonymous reviewers for the depth and quality of their useful, constructive comments. We also wish to thank Hélène Delacour for her crucial support in the early stages of writing, and Isabelle Huault for her wise comments. This article was also improved by feedback at the AIMS 2011 conference, when its first version was awarded the M@n@gement prize. Finally, we are grateful to scientific councils of IRG and UPEC University for the funding of the translation of the paper. The unsuspected dynamics of the regulative pillar! M@n@gement, 17(3): 145-179 179