SECTION I: SOCIAL CHANGE AND CIVIL SOCIETY MEASURING PERCEPTIONS QF "CORRUPTION" IN CZECH SOCIETY Raymond June, University ofCalifornia at Berkeley For several years now, the Czech and inter­ national presses have been awash with unrelenting stories of political and economic corruption scandals in the Czech Republic that have deeply wounded national pride. Often accompanied by unflattering statistics and international polling data that routinely place the country in the bottom rung of world governance and corruption rankings, I was somewhat amused to find taxonomies of another sort when I opened the Czech daily, Lidove noviny, on February 15 (2001 :l3-15). In a three-page spread, the paper published an informal statistical ranking titled "In What are We the Best ... And in What are We the Worst" ("V cemjsme nejlepsL..a v cemjsme nejhorsi") that measured how Czechs fared in 26 categories of social life compared to other European countries. The Czech Republic topped the rankings in the annual consump­ tion of beer, while it ranked last in religious observance, thus reaffirming long-harbored and intuitive national self-stereotypes. Flanking the results were several images of the fictional soldier Svejk, the titular prot­ agonist ofJaroslav Hasek's classic early twentieth-century novel beloved by most Czechs, who served in the Austrian army during World War One. Over the years, the image of Svejk has been resignified and mobilized for different purposes. For many, he is a folk hero who embodies pragmatic common sense that enables him to subvert (Habsburg) authority and discipline. Yet others view him as an embarrassing reminder of Czech incompetence and amateurishness. It seemed entirely appropriate, however, to find him juxtaposed in an ironic, tongue-in­ cheek manner with the newspaper's enumer­ ation of national rank. For the "omni­ present. .. specter of the Good Soldier Svejk" (Cervinkova 2003:91) haunts the country's landscape after socialism, implicitly pointing out the comic and grotesque effects of measuring national character and opinion, as well as the shadow of foreign domination, that have assumed different fonns in the new world order. This article is an inquiry into one such quantitative practice - opinion polling ­ which has come to objectify current perceptions of Czech corruption and bureaucratic dysfunction. I take as my focus the influential Corruption Perceptions Index (CPI) of the Berlin-based global anti­ corruption NGO, Transparency International (TI). Specifically, I address how this technology of democratic governance has been instrumentalized and made symbolically authoritative in the global imaginary of certain local and transnational elites by the key mediating labors of a particular segment of Czech professional anti-corruption crusaders. I contend that the reification of this prominent transnational opinion poll was made through a form of "boundary-work." As socio-cultural explorations in anthropology and science studies have suggested, the formation of political and scientific rationality is a process that significantly involves the making and re­ making of dualisms between science and non­ science (e.g., Gieryn 1983, Lave 1988, Nader 1996). Creating hierarchies of scientific knowledge is a strategy that actors often use to augment their professional and intellectual authority. This article engages with these insights by examining how cultural intermediaries at the Czech national chapter of Transparency International (TIC) helped to reify the representation of foreign expert opinion, thereby casting the CPI as superior to the knowledge practices of the Czech state at a moment of weakness in the local bureaucratic machine. Implicit in my argument is that these dichotomizing practices were often unselfconscious attempts by my 10 • interlocutors at manufacturing and enhancing • pre-existing uncertainty and a sense of crisis (e.g., Czech "corruption,,).l In other words, instead of simply responding to the crisis, •• they also actively created the perception of a crisis by interpreting and "translating" (Latour 1987) it as one about corruption, transparency, and accountability for their own • purposes and the interests of outside actors. TIC workers thus offered their practices2 • oriented to abstract international forms like the CPI as a tautological solution to the crisis of legitimacy in the state and capital markets. • This is not to say that boundary-maintenance • was a seamless, conflict-free process. On the contrary, I hope to show they were constituted in ambiguities and daily struggles within and among NGO professionals at TIC, Czech bureaucrats, and the broader pUblic. • This article is based on ethnographic materials culled from sixteen months of dissertation fieldwork (2000-01, 2003) at TIC and • elsewhere, examining why and how an indigenous, globalized "governance intelligentsia" mediate and configure the trnth-defining professional idioms of • transparency and accountability - which together underpin our notion of good governance - in the Czech Republic. The • analysis derives from my participation in the daily life of the TIC office as a volunteer for various projects. • • J As Y chouda Shcnhav (1994) argues, "uncertainty" is not always located outside rational orders but is oftcn exploited to strengthen scientific and political authority. • 2 Herc I am less concern cd with the more familiar critiques of polling's design and construction (e.g., • Bourdieu 1990, Paley 200 I) than with its performative deployment by internationalized governance specialists and state officials in mass media and public rclations circuits. I find myself in accord, then, with Susan • Herbst's (2003:587) reminder of the constitutive links bctwcen polling tools and institutions of publicity that are too often lcft undercxamined: "thc close rclationship of opinion measurcmcnt to the diffusion and heterogeneity ofthc mass media is often overlooked. Without an extcnsive mcdia infrastructure for manipulation ofpublic opinion - print and broadcast -­ • as well as a keen understanding ofthat complex infrastructure, survey research data are not particularly useful." My argument unfolds in five sections. First, I sketch the present disunity of government bureaucratic knowledge in the Czech Republic that has enabled the governance experts to achieve presence in the political­ cultural field. I then critique the salient features of international indices in governance, especially the cpr and its valorization of foreign professional opinion. In the third section, I discuss the everyday narratives and actions that contributed to constrncting the Index's trnth in representation by analyzing the dialectics between openness and secrecy embedded in the stories and practices of publicity among Transparency professionals. The fourth section briefly examines one example of the CPI's reception among local citizens. Finally, I analyze how this global socio-technical intervention that stands in privileged opposition to conventional state knowledge practices has been contested by the Czech government. The "state" of statistical knowledge and its discontents To provide a context for the role and uses of opinion polling in recent Czech governance, it is worth exploring how authoritative estimat­ ions of the population have been framed and received more generally before moving to the specific case of polls and their tracking of public opinion. In the Czech Republic, as in many other societies, the census is the most familiar and pervasive form of government knowledge. Like polls, the census is a technology of trnth that is central to national states and large-scale organizations.3 This 3 One important diffcrence, however, is that while thc census targets bodies and popUlations, public opinion polls "aim to chart desires and graph interiorized states. In this, thcy operate less by serializing, sorting. and identifying discrete individuals than by cstablishing averages and norms that, in effect, constitute an aggregate. Moreover, the aggregatc that polls constitute is not a population that can be regulated and monitored by the statc, but rather 'public opinion' which, by claiming to express preferences of respondents, can act as a powerful legitimating mechanism for political democracy and public policies" (Paley 2001: 137). II II D raises the question of how effective the Czech government has been in enacting this power. The 2001 census controversy is instructive. From February through March ofthat year, I was in the field on the occasion of the biggest nation-wide census conducted by the Czech Statistical Office (CSU) - an official statistical agency and arm of the state - every ten years (coinciding with a broader continental European-wide tabulation of census data). The state required its citizens to provide detailed personal information that ranged from their professional and educational to religious and marital status. Even foreigners were emolled in tllls comprehensive govermnent information­ gathering activity, as I discovered when a friendly census worker dropped off the questionnaire forms at my Prague apartment one afternoon. From the start, however, the state's attempt to classify and count the national population was marked by controversy. Many people objected to its lack of anonymity: information provided on the forms could be traceable to individual citizens based on their security identification number (rodne Cislo). Tms sparked fears that confidential information could be abused, especially by a bureaucracy that was widely viewed by the public as comlpt, inefficient, and unaccountable. Public rumors and tales of conspiracies that the census would be outsourced to a small private firm (where one of its employees stood accused of being a former secret police collaborator) fanned public concerns. In addition to circulating in informal conversations, people's fears and antipathy toward the census were made visible by the print and electronic media. Perhaps the most sensational public representation critical of the govermnent's enumerative practice was a provocative Internet image that was also widely circulated over e-mail, with a grapmc photo of an ossuary and the tongue-in-cheek message: "Do not be afraid ofbeing subtracted~!! ,Subtracting people, houses, and bunkers. CSU - Czech Sadistic Office" (see Figure 1). Figure 1. 12 - ----- - ~.-' I I The state found it difficult to mount an effective response to the accusations of I bureaucratic mismanagement and corruption I circulating in the public and official spheres. The Chair of the Statistical Office (and co­ founder of TIC), Marie Bohata,l appeared in I daily media reports for more than two weeks and tried to assure the public that the process would be transparent and above board. But by then, the public's ire and fears had super­ ceded Marie's good intentions: many citizens refused to fill out the census, much to the I state's embarrassment. Despite a massive public relations campaign to persuade the populace, the hollowness of statepower was I revealed. In effect, the public ridiculed the state and its efforts to gather comprehensive statistical knowledge. In so doing, they I signaled the profane and fragile qualities embedded in the official machinery of truth. This is not unlike the fetishism of the state I and ineffective character of political authority analyzed by Michael Taussig (1997) and Achille Mbembe (2001) in Latin America and I Africa respectively. The language of I enumeration was inserted into a discursive, socio-political, and historical field where quantification has long been penetrated, translated, and usurped by various social groups in struggles over who has the power to represent the truth of the body politic (Uri a I 1993, Lampland 1995, Paley 2001). The skirmish discussed above is but one instance of the Czech state's precarious hold I on administrative rationality and public trust in recent years. The ongoing indeterminacy of political reason has created a fissure in the I I field of power and encouraged new modes of technocratic knowledge and its agents to insert and energize themselves, to which I now turn. Honor, shame, disrepute: international I opinion of the Czech Republic The saturation and essentialization of Czech pathology and difference in the public sphere I I All full names cited for the first time are real, while I first names that are introduced arc pseudonyms. can be indexed by the numerous academic as well as investigative, often sensationalistic, media accounts of bank failures, political corruption, and privatization scandals (Altshuler 2001). It is worth noting, however, that this discursive landscape is routinely accompanied by international polling and survey data that ultimately entrench these stereotypes as objective facts. So, what precisely are these opinion surveys? Perception-based indices of governance and corruption have become a veritable cottage industry across the globe. The opinions of foreign experts, in particular, are the most frequently deployed diagnostic in assessing various levels of wrong-doing, disbursing aid funds, or shaping domestic political outcomes by the international governance policy, business, and development communities. A partial1ist of the enornlOUS and astonishing range of international indices to monitor and assess good governance in the public and private sectors would have to include the following:2 • Transparency International's Corruption Perceptions Index and Bribe Payers Index (BPI) • PricewaterhouseCoopers's (PwC) Opacity Index and Economic Crime Survey • World Economic Forum's Global Competitiveness Report • World Bank's World Business Enviromnent Survey (WBES) 2 There are also regional opinion surveys of governancc, including those that cover Central and Eastern Europe, such as MERIT Research's Corporate Governance Risk Projcct (organizcd and run by students at Charles University's Institute of Economic Studies), the New Europe Barometer, and more recently the joint TIC-Gtk Index V4 (the City Corruption Propensity Index covering the four Visegrad countries). A fcw international survcys have reccntly launchcd studies of domcstic general public opinion, including TI's Global Corruption Barometer, PwC's European Fraud Perception Survey, and a Gfk-TIC survey of the corruption climate in the Czcch Republic in 1999. Furthennore, some surveys combine perceptions ratings with "hard" cconomics data, such as PwC's Opacity Index. I 13 • World Bank Institute's Govemance Research Indicators • World Bank's and European Bank for Reconstmction and Development's Business Environment and Enterprise Performance Survey (BEEPS) • United Nations Development Programme's Intemational Crime Victims Survey and Country Assessment in Accountability and Transparency (CONTACT) • Freedom House's Nations in Transit Survey • Economist Intelligence Unit's (Elli) Country Risk Service and Country Forecast • Business Environmental rusk hltelligence (BERI) • The Heritage Foundation's Index of Economic Freedom While these surveys measure and rank weak or strong governance in comparative and cross-cultural perspectives, the massive intemational taxonomies are clearly aimed at developing and transition societies that are perceived by the transnational community of academics, investors, and aid agencies to have weaker and less transparent regulatory infrastmctures. But these indices are not just technical diagnostic tools; they are also an altemative form of moral accountability to the state. For in conjuring up the expertise of self-appointed global guardians of good govemance, the new moral science of intemational polling also serves as a shaming mechanism by legitimating and authorizing censure of the state constituted in global ethical standards. Thus it is corruption and governance as surveyed in "intemational opinion," rather than its domestic public counterpart, that is privileged by the intemational development and govemance elites. Their judgment is cmcialnot just to investment and aid decisions, but also to the making of new and durable hierarchies of values and structural inequalities in the post-Cold War global order. Marking the convergence of business, govemment, and academia, these polls are also a novel technology of accountability in which foreign expertise and sentiment have become a new global moral authority on and social scientific knowledge about the Czech Republic. As an example of the new ethico-scientific nonns that govem these polls, let us consider Transparency Intemational's aruma 1 CPI. TI views itself as a leader in the global spread of corruption and governance surveys. In particular, the iconic Index has acquired an increasingly powerful and pervasive presence in the public sphere, although its meanings vary considerably across social groups, institutions, and individuals. The CPI's stated purpose is to measure and assess "levels" or "frequencies" of political economic malfeasance by ranking countries from the lowest to highest levels of perceived corruption. It aims to serve as a policy instrument in strengthening professional anti­ corruption strategies, infonning investment risk analysis, monitoring govemance performance, and generating public debate and government action through the news media. Designed and led by a German economist at Gottingen University specifically for TI and originally launched in 1995, the CPI is considered a "poll of polls." That is, it is a compilation of several intemational surveys drawn from organiz­ ations like Gallup International, the World Bank, PricewaterhouseCoopers, and the Economist Intelligence Unit. The Index is overseen and reviewed by a TI Steering Committee composed of various academics and professionals from across the globe. So as to substantiate its scientific precision, TI justifies combining these disparate data sources (roughly fifteen per year) into a single composite index by claiming it increases the CPI's reliability and independence. For example, a minimum of three to four surveys based on the past three years are used to produce a country's annual average score and rank. Hence a bias or anomaly in one index can be balanced by other more representative and accurate surveys. This will supposedly minimize the possibility of misrepresenting a 14 I I I country. In fact, TI constantly seeks to refine serves the purpose of the global anti­ and perfect its methodology in order to corruption and good governance industry (see strengthen its scientific validity. It has been also Sik 2002:112-113). remarkably open about its methodology by But these perceptual ratings are about more publishing it on its website and in other media than just consciousness-raising, allocating aid, I outlets (Kotkin and Saj6 2002:28). In or pressuring national governments to reform. addition to revising and incorporating new They also concern the making of a country's sources into its data base so that the surveys moral worth and integrity in the global I "correlate well," the CPl has also expanded fraternity of nations. In press releases and the number of countries into its annual reports, TI consistently throws out dire ranking list, from an initial count of 50 warnings that national governments risk countries to more than one hundred today, to I ignoring the CPI at their peril (Eigen 2001). ensure the robustness of its comparisons. It proudly cites how diverse polities from Results are routinely cited in influential Pakistan and Malaysia to Argentina have used international and domestic publications, such I the Index as the basis of reform, even as the Wall Street Journal, the New York eradicating corrupt administrations and Times, and the Financial Times. entrenched political malfeasance. The CPI More significant and controversial is the has even affected election outcomes in I CPI's reliance on international expert opinion countries such as Nigeria. To ignore the to discipline national governments. Rather Index, the TI Director warns darkly, "means than elicit the views of citizens and the more impoverishment, less education, less I general public, the CPI pools the results of health care" (Eigen 1997). New rational I surveys based on expert evaluations by instruments and prescriptions ofbetter expatriate and foreign residential business behavior are now twinned with a liberal executives, risk analysts, and political humanitarian concern for human rights and analysts. (TI discontinued the incorporation good citizenship. of general public surveys in 2001 in order to However, for those countries that fare poorly I "streamline" its methodology. Since 2003, TI in the Index - and they tend to be and Gallup International have collaborated in disproportionately from the global South} developing a new, though less visible and their objective representation instills a new I well-publicized, study of general public and powerful sense of dishonor and shame. In opinion in 47 countries called the Global so doing, durable racial and national divisions Corruption Barometer.) Officially, the are recast along new hierarchies of belonging solicitation of foreign expert opinion is meant I in the international moral community. To to focus on influential players in investment earn their castigation by performing poorly on and trade who presumably have greater the CPI is to suffer new forms of exclusion, authority and legitimation in the global and I stigmatization, and asymmetrical power regional economies. That the CPI valorizes relations based on "universal" moral and free the perceptions of elite decision-makers, market grounds. Stated baldly: the Index is however, also suggests something of a I embedded with the neocolonial tropes of reversal in the logic of opinion-formation: the Western moral superiority and expertise that model "expert" citizen has rapidly crowded underwrite the production of East-West/ out the statistically average citizen in I democratic governance. Sounding out the opinions of foreign knowledge makers thus 3 In response to intense criticism that the cpr unfairlybecomes something of a tautological exercise targets and stigmatizes countries from the ~Iobal S~uth, I as they are the primary group that uses the TI inaugurated a Bribe Payers Index (BPI) m 1998 m results to make aid and investment decisions. order to focus attention on the predominantly advanced In this way, the recursive, reflexive quality of industrial and capitalist countries that give bribes. However, the BPI has received significantly less I (polling) numbers, and the CPI in particular, publicity than the attention-b'Tabbing CPL I 15 North-South hierarchies, the ranking of peoples, and the re-invention of Eastern otherness (Wolff 1994; for an examination of relations between public opinion research and scientific racialization, see Monis-Suzuki 2000). In conflating the multiple socio­ cultural meanings of "corruption," the CPI therefore reduces ethics to individual choice, personalizes blame in a given country, and often creates or exacerbates the problems it is supposed to solve. And so despite the strenuous claims by TI that the perceptions index is not necessarily an accurate measure of actual incidents of corruption and "notwithstanding the fact that perceived corruption is a one-sided indicator, once the powerful perception index is presented as an objective indicator of a country's comparative moral value, the index acquires its own dynamic" (Kotkin and Saj6 2002:29). The tarnish to the Czechs' "civilized" and "velvet" reputation has brought about dubious and unwanted attention from the international conununity. This should lead us to consider how and why these polls have granted international expert opinion such symbolic power over and within the contemporary Czech Republic. hlternational indices of good governance in Czech society have contributed to fixing the perception of pervasive corruption and its entrenchment in everyday life among many foreigners and some locals by frequently accompanying sensational and serious investigative reports. That is, the media produces and reinforces the world and phenomena that they construct through the hard social facts of international polling that saturate popular and scholarly publications. Table 1: Transparency International CPI (2003). Rank Country 1 Finland 2 Iceland 3 New Zealand Denmark 43 Cuba Jordan Trinidad & Tobago 54 Czech Republic Bulgaria Brazil 133 Bangladesh Score 9.7 9.6 9.5 9.5 4.6 4.6 4.6 3.9 3.9 3.9 1.3 In what has become something of an annual rite of passage, the Czech public reads about their country's latest free-fall in the CPI Surveys Used 8 7 8 9 3 7 6 12 10 12 8 Standard Deviation 0.3 0.3 0.2 0.4 1.0 1.1 1.3 0.9 0.9 0.5 0.7 rankingsl - finding themselves invariably I The Czech Republic's relative standing in the CPl scale has steadily fallen, from a rank of 37 in 1998 to 54 in 2003, while the overall coverage of countries continues to increase annually. 16 I I I grouped alongside or even below countries such as Bulgaria and Cuba - in technical, reified form (see Table 1). The perhaps not­ I so-subtle suggestion is that the Czech Republic is similar to a Third World country. Tabloid journalism and unsubstantiated I studies are thus granted greater legitimacy by virtue of their proximity to authoritative knowledge. I This objectification is further strengthened by the fact that foreign and comparative opinion polls behave implicitly as a tribunal of the international community and symbol of a higher rational and moral order "above" local I politics. Given the on-going struggles of Czech state and national intellectuals to administer retributive justice and accountability over the past decade, the moral I and juridical authority of local governance has been displaced to some degree on the exterior body of international polls and surveys. From I this perspective, independent surveys carried out by translocal institutions can be seen through the eyes of many global elites as I acting as an impartial and moral agent of the international community by putting the state on trial and meting out its own forms of I punishment. Evidence of the indices' power to reprimand can be found in their wider impact on I normative macro-economic indicators. For I one, they have affected foreign investments in the country since partners and grant agencies frequently draw upon studies commissioned by organizations like Transparency Inter­ national when making their decisions I (Altshuler 2001: 117). The Czech Republic I has also suffered the rebuke of the European Dnion (ED), which, in taking into account the CPI among other barometers of opinion, has made specific (sometimes unreasonable and hypocritical) demands for institutional and structural reforms as the country seeks I European integration. We would do well to recall the pragmatist dictum ofW. I. Thomas that if a social situation is perceived as real, it I is real in its consequences (Thomas and Thomas 1970[1917]). On a broader level, then, instruments of governance knowledge like the CPI can have a more dynamic locutionary force than the state, producing what Michel-Rolph Trouillot (2001: 132) calls "legibility effects," or techniques deployed by translocal, non-state agencies to aIiiculate public sentiments and regulate populations more successfully than the national bureaucracy. Daily narratives and pUblicity work Polling numbers, however, do not speak for themselves. They need to be mediated, interpreted, and codified through an institutional circuit - or what Dominic Boyer (2000:461-462) calls an "epistemic feedback system" - peopled by various professional experts who link the academy to market research organizations, international NGOs, and the media. Not surprisingly, the public representation of foreign opinion is de­ contextualized and reified in the process as journalists and policy-makers finally append and reproduce selected aspects of the opinion poll results to their press coverage. To investigate the authoritative production of the CPl, it is therefore worth examining how the poll was mediated and disseminated in routine everyday practice among the TIC staff. The construction of bounded, factual representation, as I show below, was reinforced through the mutual elaboration of openness and concealment that was integrated into the NGO's practices ofpublicity. Indeed, despite the professed transparency of polling, it was dependent on its constitutive other­ secrecy - to perform its power (see Geschiere 2002). As the centerpiece of Transparency's professional agitation efforts, the CPI was primarily woven into the textures of daily life at TIC in the form of paper and electronic documents. During my fieldwork, I was constantly struck by how the CPI and other international polls were silently filtered and deployed through an extensive documentary apparatus and paper trail:2 grant proposals, 2 The dense circulation and dissemination of documents have some interesting continuities with the unofficial communication activities ofCzechoslovak dissidents and their refoTIn efforts during the 1970s and 1980s. I thank KIista Hegburg for this point. 17 annual reports, and print media sources. My like faceless conduits. In an interview, he e-mail and office mailboxes were frequently explained: bombarded with articles and memos focusing on recent quantitative surveys of corruption and good governance. One electronic message that I received, for example, contained an attachment that summarized forthcoming international newspaper surveys "in case it helps [the national chapters] in their dealings with the media." Another time, the TIC Director distributed without comment a one-page photocopied article on the latest mathematical models for quantifying overseas country investment risk. More generally, my interlocutors found the CPI an easy shorthand to highlight the gravity of corruption in the Czech Republic for donors and the public. Many of them told me they found the Index an efficient tool that reduced a complex social phenomenon like corruption to information, thus making it simpler for them to communicate with various constituencies. "It's easy, if someone from the media calls me to ask about how deep corruption and bribery really are in the Czech Republic, I can just refer to the CPI without having to go into great detail," one TIC professional told me. The CPI and my interlocutors' role in deploying its results, however, also proved to be a deep source of ambivalence. The more concrete and frequent critiques that I heard in corridor talk or kitchen-table conversations as well as interviews was how the Index discounted their role as producers of knowledge. They often expressed feelings of being non-persons, as if cogs in the marketing and public relations machinery who merely transmitted information to the public. In addition, my interlocutors critiqued the Index for being shot through with secrecy (a charge made with particular forcefulness by the older generation of TIC professionals). As such, the CPI served as a screen for debates about and expressions of ambivalence toward their work more generally. Martin, the TIC project manager of the police project, often remarked on the sense of feeling We are not experts; we are just managers at Transparency. So any project that we do involves coming up with an idea, getting the money, and bringing people together who know something about the issues. We can do press conferences and publicize the work of the experts like the CPI. I think that Transparency is not doing anything else but putting this issue in the newspapers. It can do something more than just publicize things like the CPI, but it doesn't. ... The CPI and all those press conferences do not try to solve the problems of corruption. It's just mapping out the problem, and that is it. The disavowal of expertise and ambivalence towards managerial authority were familiar laments that I captured throughout the course of my fieldwork. It conveyed some of the tensions that pervaded the workplace. The management and dissemination of the CPI were imputed to be mechanized and disembodied labor. In this account, they perceived that their contribution was devalued or insignificant in contradistinction to the generation of data and hard numbers by "real" experts, whose work they merely publicized as appendages. Martin's colleague, Dana, echoed his sentiments. Like TIC's Director, Michal Burian, Dana had previously worked at the Prague branch of the global advertising agency Ogilvy and Mather. She had a no­ nonsense demeanor and performed her work with great efficiency and minimal fuss. Her criticism of the CPI was expressed clearly to me one day over coffee: One of (TIC's) main target groups is the public. And we have to do something about this. We can't work as a scientific organization (vedecka organizace) and do some surveys, such as the CPI. I don't think we are mentally equipped for this, and educationally equipped for this. We 18 • ••• II ••••••••••• can actually initiate some surveys, but not done by us. I think we can collaborate with some institutions, but we have to think about the project: how we are going to use the surveys, whatever it is, for the project that would be usefuL Because the survey has to be the first step, and the results of the survey have to be the project. The disappearance of claims to expertise from their self-portrayal seemed to me an expression of some of the conflicts and uncertainty that attended their work. Even though my interlocutors were endowed with impressive professional credentials and work experience, they were often beset with uncertainty over their epistemological authority. This tension manifested itself in the wildly oscillating stories I heard. Most TIC professionals expressed tremendous pride in their professionalism as a sign ofintegrity and self-discipline. They held that this made Transparency a highly respected and trustworthy institution, especially in comparison with their local counterparts. But a few others I spoke with also adopted more troubled expressions when talking about the technocratic orientation to knowledge. A number of them voiced concems that the CPI evacuated their sense of efficacy and expressed resentment that they could not engage with more participatory forms of research and organizational work with citizens and allied social organizations. I sensed that portraying themselves as non-experts spoke to this fundamental tension at TIC. Frequently articulated in tandem with the denial of self-agency were their feelings of mystification at the Index itself. Despite the professed openness and transparency of the international polls, they often claimed that the CPI and its sources appeared cloaked in secrecy. Iva was a particularly vigorous commentator on this point. A middle-aged woman of mixed Czech-Slovak-Hungarian descent, she presented herself as a Open Society Fund in Bratislava (Slovakia) before assuming her current post as a TIC project coordinator for the 10lh International Anti-Cormption Conference (lACe) held in Prague in October 2001. She stated: The CPI has to show you what is necessary to do in the field. The survey can show us what are the weak points, what are the strong points. But the cpr is problematic because it is very difficult to compare the development in Peru with the development in the Czech Republic. This is nonsense!. .. That's why for me it's a little bit scary. That's why I'm keen to know who will use this. Maybe some consultancy institute and some media opinion researchers, I don't know. I really don't know (Ja opravdu nevim). I would love to know who is really producing the CPI, too, and who is able to compare it. Although they were well-informed and intelligent, people like Iva frequently claimed that they did not know much about the Index and often "black boxed" its sources of production and its uses. And yet, their distance from the Index seemed more a critique of professional knowledge rather than an expression of sheer ignorance. The daily narratives that I heard about the CPI's obscurity made it visible as an object of critical ret1ection and debate. What, then, do such statements about the opaque workings of the CPI suggest? In some ways, it is not surprising that many of my interlocutors gave critical responses such as Iva's. Much of their service work relied heavily on secondary polling research conducted by different international and Western agencies, which the TI Secretariat research staff then cobbled together to make average composite scores for each country. Furthermore, these comments revealed the broader tensions between the TI head office in Berlin and its Czech national chapter that became increasingly visible in their long­ •• standing differences over the organization and cosmopolitan professional, having worked for a number of years in Germany and the financing of the 10lh lACC. I witnessed or overheard complaints on both ends about • 19 undue expenses, miscommunication, PR excess, and managerial conflicts, in addition to the usual personality clashes. All of this fueled the sense of remoteness and discOlmection from the CPI and its forms of knowledge. What must be stressed, however, is that by minimizing their intellectual authority and ceaselessly pointing out the CPI's hidden dimensions in their everyday narratives, they cast a veil over their own interpretive labors in translating and disseminating the CPI for the media and public. In their stories that I heard almost daily, making the ambiguities of their labor and the Index's secrecy the targets of scrutiny ironically deflected attention away from their own active work processes and suppressed my interlocutors' recognition of the daily mediating praxis in which they participated. As a result, they became implicated in the Index's master representation and reinforced the invisible hand of its authority. The backstage contingencies behind the Index were further modulated by the objectifYing practices of what Jiirgen Habennas calls "publicity work," which "is aimed at strengthening the prestige of one's own position without making the matter on which a compromise is to be achieved itself a topic ofpublic discussion" (1991 :200). Indeed professional techniques, such as public relations, have become increasingly central to the production of techno-political and moral authority by international NGOs and civil society organizations. As Andrew Barry reminds us, the art of politics is constituted by its technical and material features, with social and often "anti-political" effects: Devices such as press conferences, parliamentary debates, public demonstrations, public opinion polls, political analyses, electoral registers and so on are not incidental to politics. They play a critical role in making it possible for politicians, trade unionists, activists, lobbyists and citizens to act as political agents. The political actor does not come isolated into the political arena any more than the consumer comes isolated into the marketplace. He or she comes with a whole array of material devices and fornls of knowledge which serve to frame political action. There is a physics to politics .... [However], what is commonly termed politics is not necessarily - or generally political in its consequences. Politics can often be profoundly anti-political in its effects: suppressing potential spaces of contestation; placing limits on the possibilities for debate and confrontation. Indeed, one might say that one of the core functions of politics has been, and should be, to place limits on the political ... Such skills .. .in creating arrangements where consensus can be reached and in managing the press and public relations and so on, are often extraordinarily technical. .. [W]e might consider the importance of political science, political theory and public opinion research in justifying and informing the conduct of anti-politics and reproducing particular forms of anti-political action. (Barry 2002:269­ 271) My ethnographic example comes from the 2001 CPI press release, focusing on the highly fetishized and carefully scripted press conference that was the primary technical mechanism by which the Index was conveyed to the public. By adhering to the instrumental and rationalist ideals ofpublicity, my interlocutors' construction ofthe CPI and its representation contained local contingencies and suppressed attempts at public debates of the Index itself. In the Czech Republic, TIC and its PR apparatus were extremely important nodes within the entire range of relevant institutions and techniques that granted the CPI the status of representational truth. But conflicts over the CPI became increasingly visible during the week preceding the June 27, 2001 press 20 ~I • release date in Prague over the content of the Vaclavske namesti (Wenceslas Square) where • message they wanted to comey to the media the press conference would take place that and public. On the one hand, Marie Bohata afternoon, I asked him how he would address wanted to stress less the importance of the this issue. He responded, "I think it's •• CPI results than the encouraging news of important to show different sides to the Index continuing foreign investment behind them. and how it can be read in different ways. If too much emphasis was placed on the However, we also have to send a clearly negative aspects of the Czech Republic's poor defined message through the media, otherwise • • standing, she feared, then this would have a it will be confusing." snowball effect and business investment and Michal's testimony suggested that the primary confidence would drop off precipitously. The task of the press conference was not framing central question for her was how to spin the and managing the definition ofthe CPI, but results to reflect this meaning. • rather the aesthetic and performative tasks of On the other hand, there were others in the publicizing the Index. My initial suspicion organization who felt that this strategy would was confirmed when Michal reached into his undermine the CPI and evacuate its larger briefcase and pulled out a one-page • message. As I was helping Martin alert handwritten document that he had various media outlets a few days before the photocopied ofPR tips suggested by TI. As I press conference, for example, we talked skimmed the document, I found no reference • about how Marie wanted to manage the to the meaning or content of the CPI. For • message during a momentary lull in their instance, the first suggestion was to "find a routine. Martin put down his phone receiver good location for the press conference - good and said to me: "I know that the CPI is means convenient for the press." Other major problematic, but to achieve a meaningful tips included making "sure the conference is outcome, we can't just say the picture is not at a convenient time - that it does not clash as bad as portrayed by the polls. The CPI with other major events and it is not too close • should be used as a way to encourage people to media deadline" and to "have good, clear to improve by showing them that the copies of all documents available to hand corruption climate is not good. I raised this out." In order to ensure the clarity of the • with Michal (Burian), but he remained message, they were encouraged to "write a noncommitta1." Another TIC professional, five to eight minute press conference opening Jaroslav, interjected by calling the CPI and its statement. Introduce Tl. Explain the • methodology, "an embarrassment." CPI. ... ( and to) close with a short statement Nevertheless, he felt the press conference repeating your key points." By the time we should be an occasion to discuss both the reached the hotel entrance, Michal explained • problems and efficacy of the Index without that TIC's strategy was to run a PR media necessarily projecting the positive image that blitz in order to sell and heighten public Marie advocated. awareness of the CPl. As we approached the hotel, he said, "We only have enough time to • The challenge, then, was how to reconcile say so much. So we have to stay on message. many of these discordant perspectives that I want to include both sides, and maybe that were voiced throughout my fieldwork and will come up when the journalists ask • particularly during the week leading up to the questions. But the main goal for us is just to press release. What seemed to be at stake was get the results out and raise public how best to make the right judgment and awareness." With these comments and his • interpretation of polling statistics by framing treatment of the tip sheet as a canonical and communicating the CPI with a unified document, Michal clearly cast his professional and coherent message. Michal, while more role as that of a managerial technician who sympathetic to Martin and Jaroslav, wanted to • did not (at least overtly) question the ontology balance competing demands. Walking with the Director to the downtown hotel on • 21 of the Index and its PR apparatus.3 To him, the CPI press conference was a technical site whose main purpose was to disseminate and saturate the public sphere with a statistical model of expert opinion as if it were an infonnational product. It is no surprise, therefore, that the press conference shaded into an intense infonnation lobbying session and PR offensive that depoliticized the public representation of the CPI. The conference began with Marie and Michal briefly reciting their prepared statements, outlining the Czech Republic's comparative rankings, and drawing attention to TIC's own activities in combating corruption. Marie then invited questions from journalists. A few journalists wanted to probe more deeply into the methodology of the Index, such as the sources of infonnation used to rank the Czech Republic and the problems associated with quantifying subjective perceptions. But such perfunctory questions were already addressed in the press packet that was distributed to journalists (and posted on the organization's website). Bored looks registered on the faces of most participants. Towards the end of the Q & A, Marie noted in a fonnal bureaucratic manner: "I do not want to impeach this Index, but I find it striking the degree in which foreign investments continue to flow into the Czech Republic despite the negative evaluation of corruption here." When no one responded, the press conference ended swiftly, approximately thirty minutes after it had started. Whence the banality and purpose of the CPI press conference? For one, the incoherent and disconnected attempts to negotiate the numbers' ambiguity dissipated under the strain of publicity and its remarkable ability to flatten serious messages and multiple voices through sound bites. (Indeed, when I read the local newspapers later that week to track the media reporting, all of them except Lidove 3 Kimberley A. Coles (2004) charts a similar process of technicization in the democratic electoral techniques (ballots, polling stations, rule of law) and sC\f­ perceptions of international officials and electoral workers in postwar Bosnia-Herzegovina. noviny [June 28, 200 I :2] failed to provide critical commentaries or generate significant debate on the CPl. Rather, the stories mechanically reproduced the results without reflection and were accompanied by brief, often decontextualized, passages on the country's continuing corruption crises.) Furthern10re, the reification of the numbers as trustworthy and objective was not self-evident by virtue of their proposal and design alone. They also had to be brought into existence by being corporealized, dramatized, and confinned through the fonnal presentations of reports, tables, and statistics, thereby intensifying the decontextualization of the CPI and its representation of foreign opinion. In so doing, the press conference operated as a feedback loop by putting Czech "corruption" in the limelight and raising the profile of the anti-corruption agency in the Czech Republic through eye-catching results. Hence it would not be an exaggeration to say that the existence of the Index depended significantly on perfonnative publicity in order to heighten and objectify the "crisis" in Czech governance. Representation in dispute The public reception of the CPI, of course, was more unpredictable than the seductive, scientific representation of Czech/Eastern alterity. While tracing the effects of professional communication with any precision is elusive, I did witness one occasion in which the polished sheen of public relations was disrupted when the Czech weekly newspaper, Respekt, hosted a public debate in Prague's avant-garde Theater Archa (Divadlo Archa) on October 29, 200 I. In one of the rare instances when the public made its voices heard vis-a-vis my interlocutors' practices, the governance intelligentsia lost some grip over the mass-mediated representation of survey research. The civic discussion of corruption and the CPI showed how professional henneneutical control could 22 I I be derailed when put into direct public circulation with citizens.4 I The October debate on "'The Greatest Bribers in Europe,' or 1,000 Faces of Domestic Corruption and its Efflorescence in the I Oppositional Agreement Years" ("'NejvetSi uplatkafi v Evrope,' aneb Tislc tvan tuzemske korupce a jeji rozkvet v letech opozicni smlouvy") was part of a year-long series of I I public events sponsored by Respekt. Key local personalities from the world of politics, journalism, civil society, and academia were invited to converse and spar with the newspaper's editors and public on topical domestic and global issues ranging from I freedom of the press to the 9/11 terrorist attacks in the United States. The intimate theater, which was packed with I I some two hundred people in the audience, provided a forum that was desif,'1led to break down the barriers between technocrats and citizens. Facing the audience were four panelists and two moderators for the evening's program, all of whom were seated I on the stage. The setting was informal and resembled a town hall meeting. After some brief opening remarks, the Chief Editor of I Respekt, Petr Holub, asked the panelists if any of them had dealt with corruption on a personal level. Everyone demurred with I nervous jokes and laughter about corruption, as if it were the outer limit of human comprehensibility and knowledgeability. 5 I 4 Many TIC professionals, including the Director, often I yearned for more direct contact with the Czech citizenry. Yet, as these NGO workers would often remark, the public was too diffuse and incoherent for them to address their citizen-audience directly, thus forcing I people like Michal to speak through the media. And so I while several TIC activities did circulate in public, they were highly mediated, with mediatizing effects. Thanks to Krista Hegburg for this clarification. S Krista Hegburg, who also conducted (preliminary) dissertation fieldwork at TIC while I was in the field, was often struck by how almost everyone in the TIC I office would admit to her, in confessional mode, that thcy did not know how to go about bribing people. One professional told her that she would not know how one could even signal that you want to enter a bribe (personal communication, November 30, 2004). It was at tlus point that the discussion became more heated when the editor then asked them, "What is the worst part of corruption in the Czech Republic? And why are we descending in the world charts and rankings?" Transparency's Director was the first to respond. In a carefully orchestrated performance, Michal turned and pointed to graphs and tables ofthe CPI results that were projected on a large screen belund the panelists. In a dry, matter-of-fact tone, he commented: "This is a comparative index of perceived corruption across the globe by businessmen and investors. The results show that we have performed worse than many other countries, and we're not far from the Ukraine, Romania, and Bulgaria." Before Michal could continue, he was interrupted by Tomas Sokol, a prominent lawyer, who rebutted: "there are not enough facts to make this Index accurate. I don't want to denigrate it, but I think that there are many actions and behaviors that may be caused by reasons other than corruption, which the Index can't capture. Corruption certainly exists here, but as a lawyer and citizen, I camlOt say whether corruption is increasing or not. So how can the CPI measure its level?" Sokol's response was greeted by loud claps and cheers of approval from the audience. Clearly flustered, Michal responded predictably by acknowledging that the Index should be read with some reservation: "In every country, corruption is read in different ways. In Hungary and Poland, corruption in the health services is not considered corruption there." "Nevertheless," he continued, "the CPI is necessary to make the topic less taboo in public discussion." His defense was b'Teeted by a few whistles from the audience. A middle-aged man rose and retorted with anger, "We have seen in many cases of self­ enrichment that people, including our senators, have escaped facing consequences when in fact they should have been prosecuted. We need penal responsibility, not pseudo-scientific moralizing from the CPI!" More enthusiastic clapping and a smattering ofjeers followed before the raucous discussion turned to other issues. I 23 I This ethnographic illustration reveals how the tightly scripted public representation of international polling could easily be unraveled. While Michal packaged the CPI professionally so that its public framing and choreographed message would screen out external "noises," some members of the public and panelists exposed the internal debates and ambivalences by resignifying the CPI and rendering it suspect as a technical mechanism of accountability. In short, they penetrated and critiqued the Index and its image makers for partially constructing the sense of crisis. We can perhaps see more clearly, then, why institutions like Transparency International were so vigilant in managing their messages in order to suppress such surfeit meanings in the public sphere. State anxieties over representing the vox populi The ascendancy of international indices, such as the CPI, has forced the Czech state to acknowledge and respond to the new transnational technique of democratic governance. Some background notes on Czech public opinion research, however, are necessary in order to grasp the relation between international agencies and the state as one that significantly involves boundary disputes. The embrace of public opinion polls by the postsocialist Czech government, to date, has been rather uneven. Part of the reason is historical: although polling enjoyed a brief renaissance during the 1968 Prague Spring, the official practice of scientific polling surveys (tied, primarily, to developments in the discipline of sociology) moved into a different, perhaps more subterranean, realm after the Warsaw Pact invasion.6 Since the end ofthe Cold War, social scientists have struggled to rejuvenate empirical sociological research. In 1990, the Institute of Sociology was re-established as a branch of the state Academy of Sciences after a 20-year absence. For the remainder of the decade, public 6 For a more extensive historical discussion of polling and its precursors in the Czech lands, see June (2005). opinion surveys sponsored by the sociology institute and the newly formed Institute for Public Opinion Research began to increase and even flourish at times. Nevertheless, they were limited to predicting outcomes through exit polls in general election campaigns (although studies of other socio-political issues were periodically carried out). Moreover, the fluid post-1989 electoral landscape and the unstable multiparty coalitions and realignments made it difficult to cultivate government accountability and "predict" results through polls (Leff 1997: 1 04). Finally, the awkward institutional arrangement that separated the Institute of Public Opinion Research from the sociological institute made systematic research on behalf of the government difficult. It was not until 2001 when the newly named Center for Research on Public Opinion (Centrum pro ryzkum verejn