The making of a European civil society: “Imagined”, “practised” and “staged”

Abstract

The paper tries to go beyond normative debates on what should count as civil society and who is to be identified as being part of civil society by relating such normative claims to the social sites where they are produced and claimed. It starts with the idea of conceiving civil society as a script which is used by a series of collective actors. This performance is structured first by general background conceptions of what a civil society looks like and which are taken up by all those engaging in claiming to be a civil society actor, and secondly by the public to which such performances are addressed. This second dimension is further explored assuming that the public civilizes civil society in terms of accepting or rejecting the performance of particular actors. Thus civil society is reformulated as a process in which the question of who is a legitimate part of civil society is permanently contested by a public that is addressed by these collective actors.

1 Situating the issue

The following remarks¹ are an attempt to relate the normative claims of a civil society to its sites, i.e., to the places where actors construct what we call civil society. This is not meant to defend a particular normative notion of civil society (e.g., a liberal or republican notion). As a social scientist, I am interested instead in the different modes in which the idea of “being civil society” is claimed by real actors in time and space. Such a sociological perspective is nonetheless helpful to assess the role of normative claims as the cultural and institutional software that enables ongoing social processes. Ideas intervene into social reality. In this sense, civil society will be conceived in this chapter as a “site” of staging normative claims that inform its own social practice as well as the social processes taking place in its environment.²

Without saying where civil society is to be located, we can already propose the theoretical assumption that “civil society” refers to a mode of justification of social relations, or, in the terms Boltanski and Thévenot, to “un ordre de la grandeur” (Boltanski & Thévenot, 1991). Its mode of justification equally works for the reproduction of power as it works for generating the critical capacity of actors, individual as well as collective ones (Boltanski & Thévenot, 1999). The word “grandeur” is well chosen, since normative justifications refer to big narratives which we invoke in order to make our practices appear as good practices (and even best practices in this new mode of speaking).³ My first claim here is, following Boltanski and Thévenot, that civil society can be talked into existence by reference to “ordres de la grandeur” (in the plural), which compete with parallel claims for “grandeur” made by the family, religion, the state, or the market. To be distinct from these other sites civil society needs a privileged claim for such grandeur which could be called the “civil” mode of justification.⁴ This civil mode is contested since it is also claimed by other sites. Civil society therefore has to appropriate this grandeur for itself through its actions, which need to take place somewhere, i.e., they need to be located in a social site.

At this point it is still left open whether and how civil society can be substantiated. Is civil society merely a justificatory arrangement, or is it also something that is enacted somewhere? In order to approach this latter dimension, a second theoretical assumption comes in: civil society, it is argued, emerges from a particular type of social relations. Civil society is, in my understanding, a site where people construct a social bond which demands solidarity from all to realize the common good. This specification deviates equally from the Hegelian/Marxist position, which denounced civil society as the site of pure interest or egoism, as well as from the liberal account which takes civil society as the sphere of individual freedom (private and political) that is protected against state intervention. Civil society is more than the realm of private interest and less than the realm of the state. It is located between Staat and Gesellschaft (or “bürgerliche Gesellschaft” in the sense of Hegel)⁵ and claims a civility that is not reducible to these two realms of social relations. Civil society represents a site where a social bond reducible neither to the private bonds among rational individuals nor to the abstract bond binding them to a political order (the “state”) is enacted. This bond can be called the bond of solidarity. Thus civil society is the site where solidary bonds are enacted and performed.

Departing from this definition, it follows that civil society appropriating solidarity as its mode of justification will enter into conflict with other sites such as the family or the state which engage equally in such appropriation.⁶ This brings us to the question of real civil societies as the site of imagined civil societies. The theoretical argument is that both notions are not simply referring to analytically distinct dimensions (e.g., structural versus cultural) of some real entity called civil society (which I find to be a very confusing notion of civil society), but that both notions are two sides of the same coin. Civil society has not only to be imagined, it also has to be performed. Thus the basic methodological twist of my argument is that civil society is an idea that is made real through performative practices.⁷

From this follows a first methodological rule: in order to know what civil society is we should not create an inclusive or exclusive list of actors that we consider to be the elements out of which civil society is made. We should rather ask what kind of practices we consider from an analytical point of view as performances that are specific for indicating a civil society. In this sense anybody, even the pope, could be involved in a performance that is producing and reproducing a social relation constitutive for a civil society.

Thus there is a reality beyond the order of justification, which is the reality of practical performances taking place in certain sites and being embedded in networks of social relations. Such performances follow a script, and it is in the script that we can identify the logic that governs the practice of civil society. Scripts are performed and address a public defined as networks of social relations. It is this public that judges the practice of civil society, thus giving to civil society performance the power of influencing its environment, i.e., the system of private interests and the system of the state.

We thus arrive at a triple concept of civil society; as “imagined” (background beliefs underlying performances), as “practised” (on a stage) and as “staged” (before a public). This triple concept will guide the answers to the following four questions:

1.	What is the script of civil society actors? How do they behave when they tell us that what they do is “civil society”?
2.	What are the practices in which these actors engage to make their performance “convincing” for a public?
3.	Does civil society have an impact, and if so, how can one evaluate such impacts?

These three questions remain implicit in civil societies which are co-extensive with the nation-state. They become explicit in the case of the European Community, in which the processes of imagining, practising and staging of civil society have gained autonomy from the collusion of civil society and national political community. Therefore, turning to Europe means retelling the analytical story of a civil society in a different context and considering whether it still can make sense of what is going on in Europe.

2 Transnationalizing civil society: the case of Europe

The debate on civil society has emerged in the context of the fall of the socialist states of Eastern Europe. Civil society was considered to be that part of society that stood up, silently or publicly, against the repressive use of socialist power. Civil society has been more than just a public sphere. It has also included private spheres in which opposition to dominating regimes was learnt and practised. In Western European societies, this concept served as a bandwagon concept for all those political activities that emerged in the course of the rise of the new social movements after the 1960s. Civil engagement, protest against nuclear power plants, or the struggle against reducing individual rights, were summarized under the concept of civil society.

With the end of Communism and with the decline of the new social movements, civil society did not disappear; rather it was adapted to quieter times without revolutions and without social movements. What remained was the claim that civil society acted as a factor for fostering democracy and as a watchdog against state-based intruders into the realm of free citizens. What was left from the “high times” of the Eastern revolutions and of the new social movements was the idea of a force that is a necessary ingredient of democracy; a civil society, securing democracy by constructing a boundary against state power and by creating strong social relations among those making up civil society.

Alexander (2006) has proposed a distinction between three forms of civil society. These are; civil society I (CS1), which is the “bürgerliche Gesellschaft”, emancipating itself from traditional bonds and creating a realm of individual freedom, civil society II (CS2), which is the civil society denounced by Marx and a century of social thinkers, and finally civil society III (CS3), which is the resurrection of civil society in the course the Communist revolutions and its re-import into Western societies as a solution to growing social problems. If we accept the distinction that civil society has historically had a diversity of meanings, as Alexander argues, we might be tempted to assume still another meaning of civil society, civil society IV (CS4), which is the form that civil society takes in the context of non-national forms of political institutions. The main difference occurring in such contexts is that civil society turned from an instance of defence against state power to a partner of political power. Civil society is co-opted into these emerging institutions of political power by providing special procedural rules for privileged access to political power. This is CS4, and Europe is the exemplary case where this type of CS emerges.

The debate on European civil society has therefore rightly focused on the notion of civil society as a partner in a system of governance. “Governance” provides the key concept for describing this changing relationship between political power and civil society. In a system of horizontal coordination among political actors, civil society is no longer only the addressee or the contestant of government but becomes an intrinsic part of the structure of governance. It is seen as a particular network of collective actors entering the more encompassing network of political actors in the “postnational” situation.⁸ CS4 is, from a functional point of view, equivalent to the European Parliament, both associated with the political power built into European political institutions. Its legitimacy is based on a mechanism equivalent to that of the European Parliament, on a mechanism of “representation” of the people, the minimal requirement for democratic legitimacy.

This functional equivalence with the parliament has at times been noted, yet with different ways of evaluating their position in the system of political power. Some favour the theory of the primacy of the Parliament, others the primacy of collective actors such as “civil society actors” (Kriesi, 2008).⁹ Some simply argue for a kind of equal division of labour between civil society and the European Parliament. These debates have primarily focused on the way in which collective actors relate strategically to each other on the European level.¹⁰ The remaining theoretical problem is the role of normative beliefs which provide legitimacy to these collective actors and their interaction games.

The term civil society mobilizes normative claims that differ from the claims built into the mechanism of representation by election. Since the electoral mechanism does not fully apply, the European Union is generally described to suffer from a democratic deficit. Others use civil society to specify alternative mechanisms of representation that compensate for the democratic deficits of the EU. This alternative way of legitimation is frequently linked to modes of functional representation of collective actors and good governance as a way to reconcile input with output legitimacy (Majone, 1998; Moravcsik, 2006). The basic dispute then is whether parliaments and with them political parties still are the privileged source of democratic legitimacy or whether this legitimacy comes from other social forms of organizing political action, such as interest groups or even protest movements, which is finally civil society (CS4). Thus the issue of the democratic deficit is finally contingent on the issue of whether civil society is a functional equivalent to political parties and electoral procedures in terms of generating democratic legitimacy.

It all depends then on what civil society is exactly. Therefore the analytical question of what civil society is and how it is performing becomes decisive. First, civil society represents “people”. This is specified in terms of background beliefs about democratic legitimacy. Second, it competes with other political actors and institutions in the context of European governance. This new way of relating civil society to political power (and thus to their social positioning in decision-making processes) is still mainly framed in terms of shifting actor alliances that are somehow related to the people as their legitimating basis.

In the context of theory and research on European governance civil society is predominantly analysed in terms of a principal-agent model in which the principal (a more or less well defined social “group”) gives a mandate to an agent, here parliamentary representatives or organized interest groups or even collective protest actors, to act in their interest. This model could be extended by adding feedback effects, which either tighten or loosen the links between principals and agents. Gains or losses of interdependence between principals and agents depend on the kind of legitimacy that is generated in these interactions. Yet this does not lead to a theory that explains why some legitimating frames work better than others. Normative conceptions are imported as options to justify an institutional design. In this vein governance structures are often linked to normative conceptions of deliberative democracy (Kohler-Koch & Finke, 2007; Neyer, 2006) whereas classic government structures are linked to liberal models of representative democracy. These linkages might hold, but what are the processes that make them hold? Why do certain constellations work better with some specific normative conception than others?¹¹

Here the following arguments in the literature dealing with European civil society and European governance try to offer a remedy. It is empirically taken for granted that “partnership” has become a key term in EU-civil society relations and needs to be taken into account as a second important source of representation of people (see Kohler-Koch, in this volume). It is equally taken for granted that the public that is represented in these partnership networks provides legitimacy. It is thirdly taken for granted that there is a limited set of normative conceptions that can be used for generating or withholding legitimacy in respect of these “agents”. The modelling of civil society along the principal-agent model needs to make a lot of unproven assumptions on the direction which processes of cooperation and legitimacy generation take. This is the black box of this tradition of representing European civil society. The open question is how these processes work and in which direction they work.

The theory of civil society proposed takes a strictly sociological perspective. It proposes a sociological translation of the principal-actor model. Firstly, it embeds the civil society agents and their interaction games with other agents (such as parties, etc.) in a system of social relations among the principals, i.e., in structured networks of social relations constituting a public for the agents.¹² These principals constitute a public with particular expectations on the agents.

The theory secondly looks at the “games” of these agents which are seen as “performances” before a public. Civil society in this sense is a performance before a shifting public for which the idea of games among rational actors is clearly insufficient. The civil society game rather follows particular scripts which speak to the public (at a given time in a given space) and which organize the presentation of the civil society agents in relation to other agents on the stage. Civil society can be seen, the theory proposes, as an attempt to engage in a meaningful interaction with other agents, while respecting background beliefs of democratic performance of those involved and while looking at the shifting public from which it tries to draw its legitimacy.

This conception is neither normative nor rationalistic. It is based on the analytical distinction between social relations among principals and agents which are mediated by meaningful codes on the one hand and narratives that mediate equally the interactions among agents and principals on the other hand. In the following, the main focus will be on civil society as a process of ongoing interactions following a script on the European stage. The claims of normative quality that are communicated in and through such scripts will be identified and related back to the public these performances address.

This theoretical proposal takes up recent developments in relational sociology as well as cultural sociology. It abstracts from the classic description of the “people” in terms of social classes or status groups and takes people as networks of social relations that produce boundaries around them.¹³ Therefore it prefers the term “networks” (not in the sense of networks somebody has, but in terms of networks to which people are related) with shifting boundaries and shifting identities (White, 2008). It also takes “games real actors play” (to take up the phrase of Scharpf, 1997) as full of meaning, as mediated by narratives which are enacted in performances in which background beliefs are used for relating to the public, i.e., for re-establishing the link between the agents on the stage and their principals (Alexander, 2004).

By presenting a model of civil society as imagined, practised and staged, a bottom-up perspective of the making of a civil society is taken up. The setup of European civil society as a bounded case of civil society can only be assessed by a model that re-embeds the game actors play in a process of sense-making to which these actors are bound in a constitutive way. Actors do not only play, they follow a script which makes sense of their games to the others involved in these games. This theoretical hunch is realized in the model that will be presented in the following. It starts with the images civil society actors have of civil society which inform their practices which again are staged (often with strategic intentions in mind) in front of a public the feeds back onto the process of sense-making. This process will be distorted to the extent that in a top-down process staging is imposed from above creating a world of empty signifiers which inform actors to practice empty images on a diversity of stages. Whether European civil society follows the “normal model” (from images to staging) or the deviant model (from staging to images) can be assessed by looking more closely at the stages available for civil society actors in Europe which are the Commission (including its comitology system) or the European Parliament. We will surely not be able to understand these differences by the strategy of disembedding the analysis of civil society from the process of meaning generation that dominates the analysis of European civil society.

3 The civil society script

A script always involves a series of actors that relate to each other. Therefore the first task is to identify civil society in terms of the actors that are involved in such a script.

The list of civil society actors is in principle open-ended. Even a tango club could decide one day to engage in a type of action which is considered to be “civil”, such as saving the culture of lower-class Argentine music as a particular mode of establishing solidarity among underprivileged people. If the aim is saving a part of world heritage, then we would have to decide whether world heritage is within the range of a justificatory script that we consider civil. Instead of enumerating a list of good civil actors¹⁴ we should rather ask what actors do when they perform a civil society. We should expect that in modern societies we will have an increasing number of specialized organizations that do nothing else but perform this script. Thus we have potential performers, but we need to identify the scripts that tell us something more about where to find a civil society.

Conceptualizing civil society in this way, we can add variation in civil society performance: some perform better than others, some try to lie or denounce other performers. The script of a civil society performance also provides some analytical keys regarding the evolution of such performances over time by marking points of continuity or points of rupture in ongoing performances. The script engenders a story in which the performances either corroborate a script or indicate turning points in this script that might be the beginning of a changed mode of performing civil society.¹⁵ Variations can also be introduced in the time dimension.

Civil society is a performance by actors becoming “civic” not by their function in society (such as being a trade unionist or as a member of a manager club or whatever) but by claiming for “civicness” and thus participating in the “civil society performance”. What turns these actors into civil society actors is that they follow a script in which the criteria of appropriateness differ from those that hold in a family, in the state, in science or in the economy. The script tells the story of a secular social bond which holds independently of membership in a polity (i.e., as citizen), independently of religious beliefs, independently of family roles and independently of economic position.¹⁶ In this sense it is a script staging a universal social bond which in principle includes all; it is a bond of solidarity with other human beings. Solidarity distinguishes the script of civil society from the script that actors follow in the market or the state (Alexander, 1998a, 2006). Given this universalistic principle, civil society is ideally a “global civil society” (Anheier, Glasius, & Kaldor, 2001).

The script works better the more it can be linked to background beliefs about democratic government or governance which require the participation of the people in decisions of shared concern. It can be assumed that there is consensus in Europe about the background belief of democratic political participation as a necessary ingredient for a good society. Given such background beliefs, the performance of the civil society scripts fuses this script with democratic background beliefs. It thus produces narratives of a people participating in the social self-organization and self-determination of social bonds among human beings.

To be a convincing actor of civil society one has to perform according to a script which dramatizes the threat of solidarity in a society by the state, by science, by religion or whatever other sphere tends to mobilize its justification for shaping the social bond. This varies with the context of civil society performances, according to the strength and weakness of non-civil society scripts. Contextual factors come in, such as proximity to and distance from other types of social relations, especially the distance from social relations of power which are institutionalized in the state. Bringing in context allows the transition from ideal civil societies to “real civil societies” (Alexander, 1998b).

This transition from ideal to real civil societies is the transition from background beliefs to real practices. Performing civil society is translating normative beliefs (as background beliefs) into scripts assigned to the agents in game positions and relating them to the other agents in the performance. This performance does not require that everybody, i.e., the people, “participates” on the stage, as ideal democratic theory assumes. Only the elected or self-appointed representatives are moving on the stage delivering a performance that can be evaluated by all the others precisely because they also know the script, including its rule of presentation and its rules of representation. Thus civil society is not a series of actors, but it is a script imposed upon and performed by agents. The success or failure of civil society performance depends on the stage on which civil society has to enact this script. Bringing back this context allows the transition from ideal to real civil societies.

4 Fusing performance and normative narratives: the making of a successful real civil society actor

Civil society comes into existence when we see actors performing the “civil society script” in the name of the people. Individual and collective actors appear in this performance and claim to do something for the common good. They organize their voices in order to be heard and mobilize resources for that goal. They compete in the market of political claims. Some actors are more organized, better networked and more experienced than others. Enacting a script means drawing boundaries in relation to other actors, thus producing relations of competition and power with non-civil society actors, notwithstanding that competition and power relations emerge even between civil society actors. Power emerges as differential access to the stage, which emerges in spite of the equalizing background assumptions that the civil society script uses.¹⁷ In addition, the civil society script itself distributes roles which distinguish between perpetrators and victims, between the good and the bad. This is symbolic power built into the civil society script. Enacting the script on a stage produces real civil society.

Performances require material resources, especially organizational resources.¹⁸ These organizational resources can be loosely structured as networks (thus approaching the form of social movements) or highly formally organized (thus approaching the form of the rational enterprise). They can be task-specific (such as environmental organization) or task-unspecific (such as religious groups who have also other things to do than to produce political claims for common goods). These organizations interact, often in a competitive way. Civil society is a socially structured performance in which we find winners and losers, some who are more successful than others in performing on the stage. Environmentalists, for instance, can relate to human rights activists to coordinate their actions in a joint performance that makes use of the civil society script. This commonness is linked to general ideas about a universal secular bond among human beings which provides the necessary consensus for performing well on the stage.

When civil society is performed it enters into interactions with non-civil society, mainly the state. The relation between civil society performers and state performers unfolds in a competitive game as a story in which at least two scripts are connected: the civil society script and the state script.¹⁹ The civil society script allows particular actors to assume privileged roles in this competitive game; the hero who stands for justice and for the common good, who acts in the name of the people and who persecutes the perpetrators who do harm to the people, and so on. In these performative practices the actors on the stage have to distinguish between enemy and foe, between true friends and false friends, between assumed perpetrators and real perpetrators, which complicates the story that unfolds in the course of time.

Thus the civil society script is embedded in permanent movement in space and time; North against South, East against West, centre against periphery and so on. And it varies in time: generations come and go and with them ideas change. Real civil societies are as shifting in time as any other social sphere in which human beings act together. They are not exempt from power structures and inequality, especially not from inequalities in discursive power, the most important weapon of civil society.

From this three conclusions follow. Civil society is a script which is highly context-sensitive. What a civil society is depends on the situation where some actors perform in the name of the people. Secondly, it mobilizes background beliefs about democracy which are used in performing the civil society script. Thirdly, the actors performing the civil society script act as agents, i.e., in the name of a people watching these performances. This leads to a further element in the performance: the issue of the relation between actors on the scene and the public, which is the theoretical translation of the relation between principals and agents. This relationship will be analysed using the notion of fusion between performers and the public they address (Alexander, 2004).

5 The fusion of civil society and the public sphere

5.1 The public of public performances

In terms of performance civil society agents refer to themselves as representatives of the people. Whether the represented accept this self-definition or not cannot be decided a priori in a normative fashion, but is itself a social process of what will be called the fusion of civil society and the public sphere. Ideal representation is to be seen as the total and complete fusion of performers and their public, a border case in real-world politics. This raises the decisive question of how the quality of the performance is evaluated. Instead of taking a normative position on the issue, it is proposed to take into consideration the judgments that publics make of such performances. Here the notion of a public that attends performances is used to clarify further the issue of the link between principals and agents of civil society.

With this idea a rather widespread analytical confusion can be tackled. I refer to the use of two concepts often used interchangeably; the concept of “civil society” and the concept of “public sphere”.²⁰ Civil society is a mode of scripting social relations that refer to general democratic background beliefs. Apart from these general background beliefs, civil society is a sphere of justice and solidarity towards the realization of a common good. The democratic performance is followed by publics which are addressed by this performance. Publics are like the principals checking what civil society actors as their agents do. The criteria of good performance are, apart from the emotional flow, cognitive criteria of argumentative rationality. The public has to be convinced that it participates in a good performance. Civil society does not civilize itself; it has to be civilized by a public.

This link between civil society and the public sphere is implicitly taken by Tocqueville (and his followers) who emphasized the dual process of intense of political participation by some and their control by a public. Tocqueville saw this as the key to explaining the path of democracy taken by the United States of America (Tocqueville, 1986). By adhering to associations and groups which foster social relations beyond the realm of the private and cultivate social capital, the members of a society become competent public actors controlling those who are acting in their name. In doing so, such publics develop a kind of “natural” inclination toward the support of a democratic political regime. But what exactly is the democratizing mechanism that is said to be built into civil society?

Civil society produces effects of democratization because its inbuilt performative practices resonate with democratic background beliefs that are held in and by the public. In the following, I will describe these practices in terms of representation, open debate and neutralizing social differences.²¹

5.2 How the public can civilize civil society

5.2.1 The practice of representation

The first practice through which civil society claims democratic quality is the performance of public representation. In the case of civil society, representation is principally based on emotional identification, i.e., on “surrogate representation” (see Trenz, in this volume). Civil society is not (as political institutions in part are) based on the equal representation of all. Civil society actors act in the name of an abstract constituency, in the name of the people as such. They are not necessarily elected (they can be elected); they are simply “called” for doing that kind of job. The key problem of such performative practices is to provide opportunities for making the performers accountable. The solution is the permanent performance of acting for the people. Practically speaking, civil society has to defend and argue for its stakes via mass media, on talk shows, on the Internet, wherever there is a chance to involve a public in such a way that it can express emotional identification. By permanently claiming to speak for the people the agents of civil society are linked back to the public. The performance of civil society in this sense reproduces the differences that exist in society and provokes the dialectic of agreement and disagreement in the course of these performances. Instead of being exposed to the cyclical form of electoral control, representation becomes a daily control of the agents. Such control easily connects to background beliefs of democratic participation through representation. Civil society thus provides a stage for representing permanently the cleavages that permeate society at large. By exposing its own performances to a public for applause, civil society generates democratic legitimacy for itself and indirectly for those with whom its representatives interact.

5.2.2 The practice of open debate and argumentation

The second practice through which civil society claims democratic quality is the performance of public deliberation. Deliberation means that performances not only generate a flow of emotions, but also a flow of rational arguments. “Good reasons” become part of civil society performance. As such, they have to pass the test of public argumentation. This requires a debating public, which produces critics and which knows how to utter disagreements and disavowals.²² Again, this idea of argumentation as a mechanism for civilizing civil society is built into the assumption of a public sphere in which validity claims are made and addressed to those who perform before a public, here to those performing civil society. The democratic quality of civil society thus resides in its resonance to public argumentation.

5.2.3 The practice of neutralizing social differences

The third practice is performing a society of equals under the veil of ignoring the differences (social, cultural, intellectual, etc.) among those involved. Civil society provides a stage for performing diversity without violating the principle of equality. Civil society is a form of staging differences. The civil society script is full of roles relating to nationality, class, race, religion and gender. It even stages and emphasizes these differences. Yet by linking these diverse role positions in the script of an ongoing story, differences are related forcing the participants in the performance to uncouple their positions in the story from their unequal original position. Cultural differences thus change positions in the unfolding story of the performance. Since the end of the story is open, nobody knows who will be in which position in the end. The agents on the scene act under a kind of veil of ignorance: their differences are narrated without being fixed in the beginning of the story—the hero of the story does not have to be a white male upper-class Western European. The story leaves this open; it even provides means for moving the roles of the script such that power differences can, if necessary, be rectified, turned upside down, thus continuing the story with a different arrangement of the involved actors. It finally has to become a story of solidarity in time.

The theory of the public sphere contains such a practical element: it requires the equality of access to the means of production, distribution and interpretation which is taken up by the public watching the performance of civil society. This element enlarges the public sphere from which the public for civil society actors emerge. This public sphere becomes as large as the social bond created by civil society, i.e., the bond of solidarity which does not know boundaries, which is a universal social bond. Thus the bonds of solidarity constructed by performing civil society can become effective in neutralizing the bonds created by national citizenship, by religious belief, by lineage, and finally by ethnicity, and make another social bond, the bond of solidarity, the organizing principle of its own story.

6 The outcome of practicing civil society

The final question I want to address is the issue of the outcome of practicing civil society. There are different ways of addressing this problem:

1.	Civil society creates a network of agents which influence public agenda setting—this is civil society as a watchdog.
2.	Civil society enters as a participant in governance thus giving substantive input (knowledge or counter-knowledge) to the policy process—this is civil society as a policy actor.23
3.	Civil society becomes a collective actor in its own right, replacing religious institutions, state institutions and family institutions in their role of repairing the social bond—this is civil society as a civilizing force for other spheres of social life.
4.	Civil society is a self-rectifying mode of creating civic and solidary bonds among people—this is civil society as a learning process.24

These outcomes are hypotheses on possible paths of development of civil society. The first is the classic role of civil society as a public sphere which fights against any form of censoring the voice of a people. It is civil society ties to the rule of law guaranteed by the nation-state. Civil society is the politically mobilized society conceived as the other side of the state. The risk of this outcome is that society is identified with the state, leaving to civil society the option of defending in an either affirmative or critical way the nation organized by the state. The stage is the nation-state.

The second is a civil society conceived as a cooperative partner to political institutions, be they local, national or supranational. It is no longer tied to the nation-state since political institutions no longer coincide with a society—they rather address particular people that vary in time and space. This outcome is typical for phenomena such as a European civil society. Since this outcome relies on equal partnership power relations easily creep into this solution, turning civil society into a “useful idiot” for stabilizing complex policy processes. The stage is institutions of governance.

The third outcome is civil society embedded in a global public sphere that civilizes society via diffusion to other spheres of life. The most important effect and impact would be to stabilize a secular bond against the permanent intrusion of citizenship bonds, religious bonds, family bonds or ethnic bonds into its own practices. This outcome no longer focuses on the state but rather at society which is already politically organized. Here civil society is a mechanism of transforming established forms of state–society relations by applying universalistic criteria to institutionalized forms of state–society relations. Civil society acts for inclusion of those excluded in politically institutionalized societies by reference to human rights, global justice or minimizing environmental risks. It emerges as force in polities facing plural societies, often called “states with divided societies” while these partial societies crosscut the boundaries of the state. The stage of civil society are emerging transnational societies where transnational NGOs provide a forum for redirecting political decisions making processes on the national as well as supranational level (such as Europe).

The last outcome is self-referentiality of civil society: civil society as a mechanism of collective learning, mediated by the permanent force of continuing a narration once started. The stage is society which uses civil society as a rectifying mechanism. In normative terms this outcome is built into the normative idea of a cosmopolitan civil society. The success of civil society depends on the permanent performance of a narrative that simultaneously tells the boundaries of the civil sphere and rectifies its internal practices according to the universalizing principle of disregard of differences. Whether it does so in space and time is to be looked at empirically. Civil society as such, its script, is no guarantee for the good.

And what does this theoretical perspective mean for Europe? It tells us that European civil society is a possible outcome which is neither cosmopolitan nor transnational and no longer national. It is a civil society bound to governance institutions which limit its performance to the governance script. It is simply a special case of staging civil society.

Notes

1 An earlier version was presented at “The Third Sector and Sustainable Social Change: New Frontiers for Research”, the ISTR Eighth International Conference and 2nd EMES-ISTR European Conference, in partnership with CINEFOGO, held at Universitat de Barcelona, Barcelona, Spain, 9–12 July 2008.

2 The notion of site is taken here in a formal sense: it is a space where people produce events by acting in a site that is spatially identifiable. See, for this notion, Tilly (2000). See also Sassen (2002) who talks in this concrete sense of micro-sites of civil society.

3 This argument also guides the conceptualization of civil society offered by Somers (1995). This still leaves open the conceptual issue of relating argumentative and narrative elements in civil society discourses.

4 This mode of justification is well described in Boltanski and Chiapello (1999), defining its basic principle of grandeur as caring for the common good for all.

5 This refers to the classic debate in political theory and political philosophy which also has shaped the contemporary debate on civil society (Bobbio, 1989; Cohen & Arato, 1992, and Offe, 2000).

6 This processual view of not having an image or idea, but of engaging in practices of appropriation of such images or ideas (which can be long or short, successful and unsuccessful) has been proposed by McAdam, Tarrow, and Tilly (2001).

7 This has also been called the “performative turn”, a slight exaggeration since this is what sociologists normally do all the time. See Alexander, Mast, and Giesen (2006).

8 For such a view see, among many others, Kohler-Koch (1997), Kohler-Koch and Eising (1999), the contributions to Jobert and Kohler-Koch (2008), Kohler-Koch and Finke (2007) and finally Eising (2008). This notion has often been criticized as an affirmative and uncritical conception of civil society. Yet it reflects well what civil society is doing empirically in Europe. A recent and balanced review of approaches to EU-civil society relations is found in Greenwood (2007).

9 This difference can be described as a difference between traditionalists who defend the old parliamentary tradition of linking democracy and parliamentary representation and modernists who argue for a loss of parliamentary organizations in favour of representation by public interest groups. Kohler-Koch and Finke (2007) frame this difference in terms of a principled versus a functional conception of democratic participation.

10 As an interesting case, see the analysis of the role of national Parliaments in the European Union by Benz (2004).

11 This is a question less of linking normative issues to empirical social science as Moravcsik has called for (Moravcsik, 2006). It is rather a question of embedding normative claims in social science theory and using such embedded models for understanding and explaining empirical events and processes.

12 In Europe it is hard to relate to society in terms of a compound whole such as the nation. Europe is sociologically speaking a meandering society which produces a public without clear boundaries. It shrinks at times, it extends at times even beyond the citizens of the EU toward all “Europeans” and it even extends to all cosmopolitans finding their common root in Europe's enlightenment tradition. Thus Europe is a network of people with changing boundaries. This marks a decisive difference from national societies.

13 As a substitute the term “ethnic” has gained momentum simply because anything can be defined as ethnic that claims some people to be different from others.

14 What counts as civil society therefore depends upon the construction of a list of such actors by institutions. The EU in fact has such official lists. In this sense civil society is constructed in time and space by institutions, here the political institutions of the EU.

15 For a formal discussion of this idea of turning points in narratively organized sequence of events, see Abbott (2001).

16 This focus is contested, at least in terms of the background assumptions to which it refers. John Keane argues that the market has been neglected as a central ingredient of civil society and fights the anti-market script of civil society today. His solution, however, ends up in saying that the market is embedded in civil society, thus extending civil society to society at large. See Keane (2005).

17 This effect has been discussed as “the dark side” of civil society. See Heins (2004), Mazlish (2005) and Kopecky and Mudde (2002). I leave aside moral judgement and speak of real civil societies.

18 This argument is old and has been central to the analysis of the new social movements in the 1980s (McCarthy & Zald, 1977).

19 Civil society actors perform like participants in a market where actors occupy niches and relate to other actors through narratives which tell each actor about his role to play in a collective performance. This image of a narrative organization of social relations in a market has been proposed by Harrison White (White, 1981, 2002).

20 This problem is obvious in the way Habermas has used these terms, thereby confounding both terms and using them interchangeably. This is especially clear when dealing with the relationship between religion and the public sphere (Habermas, 2006).

21 The three criteria are analytically distinct yet are always realized in conjunction. Representation is a formal principle that is further specified by requirements which clarify the relationship between representatives and their public. These principles regulate the “fusion” of actors on the stage and the public.

22 This obviously is the model which Habermas has privileged, a competent public sphere, engaged in deliberative practices (Habermas, 1996).

23 This is the governance perspective preferred by Jobert and Kohler-Koch (2008), and Kohler-Koch and Finke (2007), among many others.

24 An analogous proposition has been made in social movement research by Diani (1997).