ARENA Working Papers
WP 98/14

 

 


Social Construction and Integration*



Jeffrey T. Checkel
ARENA, University of Oslo



 

 


Abstract

Social construction, which has made key contributions to contemporary international relations (IR) and institutional theorizing, has yet to make significant in-roads among scholars of integration. This is unfortunate, for it has privileged methodological individualism in the study of European institutions -- either in its strict (rational-choice institutionalism) or more loose (historical institutionalist) versions. As a result, too much debate has focused on which institutions matter in the integration process, and not on how they have effects. This paper examines the latter, arguing that a sociological and social constructivist understanding of institutions as constitutive can significantly broaden the methodological tools we bring to the study of integration; it will also help us explore how, or, indeed, whether, integration is affecting fundamental actor identities, and not simply constraining strategy or behaviour.

The essay has five parts. It begins by briefly clarifying the differences among rational choice, historical and sociological institutionalism; fundamentally, it is ontology, and not epistemology, that distinguishes these schools. Second, I review the literature on European integration, demonstrating how much of it, including work of prominent historical institutionalists and proponents of multi-level governance, is premised on methodologically individualist assumptions. Third, the epistemological and ontological bases of an alternative, social constructivist understanding of European institutions are advanced. Here, I critically assess recent constructivist work in IR, demonstrating its promise to students of integration, but also warning of pitfalls to avoid. Fourth, I show how a social constructivist cut at institutions, when compared to theoretical competitors, captures different -- and, indeed, important -- aspects of the European project. The conclusion argues that Europeanists should not be content simply "to pull off the shelf" constructivist insights; rather, they should address key lacunae in contemporary social constructivism. This will require greater attention to middle-range theory development and research design, as well as a new round of interdisciplinary exchange, in this case, between constructivists and cognitive/social


Introduction

Much of the European integration process can be modelled as strategic exchange between autonomous political agents with fixed interests. At the same time, much of it cannot. Constitutive dynamics of social learning, socialization, routinization and normative diffusion, all of which address fundamental issues of agent identity and interests, are not adequately captured by strategic exchange or other models adhering to strict forms of methodological individualism. Indeed, when it comes to these processes of interest and identity formation, the empirical reality of integration is outstripping our theoretical models for understanding it. Put differently, the dominant institutionalisms in studies of integration -- rational choice and historical -- need to be supplemented by a more sociological understanding of institutions that stresses their constitutive, identity-forming roles.

After addressing definitional issues and briefly reviewing the literature on integration, I argue that social construction, a growing and vibrant literature in contemporary international relations, can help students of integration theorize and empirically explore these neglected questions of interest and identity. I demonstrate how a social constructivist cut at institutions explains key aspects of the European project -- social learning and normative diffusion -- better than its rationalist competitors, with the practical goal being to elaborate the specific methods and data requirements for such work. In the conclusion, I shift to a more conceptual level, arguing that Europeanists can and should contribute to the further theoretical development of constructivism.

Before proceeding, two comments are in order. First, the essay’s fundamental premise is that research on integration should be problem, and not method, driven; the goal is to encourage dialogue and bridge-building between rationalists and social constructivists. By itself, each school explains important elements of the integration process; working together, or at least side-by-side, they will more fully capture the range of institutional dynamics at work in contemporary Europe. Indeed, too many constructivists are themselves method-driven, ignoring the obvious empirical fact that much of every day social interaction is about strategic exchange and self-interested behaviour. [1]

Second and following on the above, the constructivism favored in this essay belongs to what has been called its modernist branch. These scholars, who combine an ontological stance critical of methodological individualism with a loosely causal epistemology, are thus well placed, within the integration literature, "to seize the middle ground" -- staking out a position between positivist and agent-centered rational choice, on the one hand, and interpretative and structure-centered postmodernism on the other. [2]


Institutions -- Thick and Thin

Of the many institutionalisms floating around these days in economics, political science and sociology, I need briefly to discuss three: rational choice institutionalism, historical institutionalism, and sociological institutionalism. For rational choice scholars, institutions are thin: At most, they are a constraint on the behaviour of self-interested actors -- be they interest groups or unitary states in IR. They are a strategic context that provides incentives or information, thus influencing the strategies agents employ to attain given ends. In this thin conception, institutions are a structure that actors run into, go "ouch," and then recalculate how, in the present of the structure, to achieve their interests; they are an intervening variable. [3]

For historical institutionalists, institutions get thicker, but only in a long-term historical perspective. In the near-term here and now, they are thin -- structuring the game of politics and providing incentives for instrumentally motivated actors to rethink their strategies; they are a constraint on behaviour. Over the longer term, however, institutions can have deeper effects on actors as strategies, initially adopted for self-interested reasons, get locked into and institutionalized in politics. Institutions thus can be both intervening and independent variables. [4]

Sociological institutionalists are unabashedly thick institutionalists. Not only in the distant future, but in the near-term, institutions constitute actors and their interests. What exactly does it mean for institutions to constitute? It is to suggest they can provide agents with understandings of their interests and identities. This occurs through interaction between agents and structures -- mutual constitution, to IR scholars. The effects of institutions thus reach much deeper; they do not simply constrain behaviour. As variables, institutions become independent -- and strongly so. [5]

Two examples help clarify this sociological logic. In international politics, sovereignty is a constitutive institution. It is the background and foundation that makes possible and legitimizes a whole range of state interests, and proscribes others. Within Europe, consider the norms embedded within its contemporary rights regime. They are an institution in the thick sense: In many policy areas regarding human rights and the rule of law, they prescribe acceptable and legitimate behaviours and interests for states. Of course, a key question is how, through what mechanisms and processes, social norms have constitutive effects -- a point addressed below. [6]

Thus, as one moves from rational choice to sociological institutionalism, the effects of institutions changes. For rational choice and historical institutionalism, the focus is on behaviour, instrumental self-interest, strategies; sociologists, in contrast, emphasize the prescriptive, proscriptive, legitimizing role of institutions. Underlying these distinctions are two radically different theories, or logics, of action.

Rational choice scholars and many historical institutionalists posit a model of human action based on utility maximization, where an agent, when confronted with various options, picks the one that best serves his/her objectives and interests. Much rational choice scholarship, as well as some historical institutionalists, also make assumptions about the content of these interests, typically, material goods such as power or wealth. Interests are therefore given a priori and exogenously. Structures and institutions constrain the choices and behaviour of pre-existing individuals/agents, who operate under a logic of consequences (means-ends calculations). [7]

Sociological institutionalists favour a very different model, one in which strategic, utility maximization is replaced by rule-governed action, where logics of appropriateness, derived from social norms, prevail. Such logics involve reasoning by analogy and metaphor, not about ends and means. Under them, agents ask "What kind of situation is this?" and "What should I do now?" -- with social institutions and norms helping provide the answers. Institutions therefore constitute agents, providing them with understandings of interests. [8]

In research and theorizing about institutions, should one of these logics be favored, serving as the baseline? The answer here is "no," for ultimately this is an empirical question. No doubt, there are many situations and aspects of European integration where agents operate under a means-end logic of consequences (meetings of the European Council or the hard-headed interstate bargaining that features prominently in intergovernmentalist accounts). At the same time, a less static perspective reminds us that much social interaction involves dynamics of learning and socialization, where the behaviour of individuals and states comes to be governed by certain logics of appropriateness (informal communication in working groups of the Council of Ministers, European-level policy networks centered on the Commission). Unfortunately, these latter logics, while equally compelling and plausible, have received little systematic theoretical attention in political science or in studies of Europeanization. [9]


Institutions and Europeanization

To students of international politics well versed in the never-ending neorealist-neoliberal controversy, the debates over Europeanization and European integration produce an eery feeling of deja vu. On the one hand, the discussion has helped advocates of opposing approaches sharpen their central arguments and claims; similar intellectual clarifications have occurred over the past decade in the debate between neorealists and neoliberals in IR.

At the same time and in a more negative sense, the debate over Europeanization, like any academic discourse, has emphasized certain methods and actors at the expense of others. To my reading, much of the discussion has been about institutions -- be they encompassing governance or federal structures, historically constructed organizational and policy legacies, or, more narrowly, bodies of the European Union (EU) such as the Commission or European Council. Moreover, in most cases, the analysis is about how such institutions structure the game of politics, provide information, facilitate side payments or create incentives for agents to choose certain strategies.

Such an emphasis, however, comes at a cost. It shortchanges the role institutions can play in politics, or, more to the point, in European integration. In particular, their constitutive role, typically stressed by sociologists, is neglected. If the neo-debate in contemporary IR can be accused of neglecting fundamental issues of identity formation, then much of the current discussion about European integration can be accused of bracketing this constitutive dimension of institutions.

I develop this critique by briefly examining the controversy between proponents of multi-level governance and historical institutionalists on the one hand, and intergovernmentalists on the other. Among the former, I include supranational institutionalists and some neofunctionalists; rational choice theorists and neorealists are grouped with the latter. The purpose is not to provide yet another review of this debate; others have done that much better than I could. Rather, my concern is the conceptualization of institutions in these analyses. [10]

Beginning with intergovernmentalists, their conception of institutions is clear, consistent and very thin. Because these scholars are self-conscious about their theoretical and methodological foundations, it is easy to comprehend immediately the brand of institutionalism on offer. Moravcsik's carefully argued work is a case in point. Institutions at the European level are an arena for strategic interaction and bargaining; by facilitating credible commitments and side payments, they help self-interested states attain efficient bargaining outcomes. Moreover, EU institutions -- typically the Council of Ministers or European Council for intergovernmentalists -- have little or no autonomy; the preferences of the large and powerful member states prevail. Thus, at most, institutions shape the strategic environment of states. [11]

In more recent work, Moravcsik has added a theory of national preference formation to his earlier intergovernmentalist approach. Given this move, an important question -- for my purposes -- is the role of domestic institutions. Do they provide actors with understandings of their interests? Hardly, for the institutions here, at the national level, are as thin as they are internationally. The preferences of social groups are "aggregated through political institutions"; the latter have no independent causal role. This should come as no surprise as Moravcsik's domestic model simply imports the rational choice microfoundations of neoliberal regime theory to the national level. [12]

The portrayal of EU institutions in game theoretic, rational choice and neorealist work differs little from that sketched above. Garrett and Weingast, for example, draw upon theories of incomplete information and incomplete contracting to offer a sophisticated analysis of the role played by institutions in the construction of the EU's internal market. The use of such approaches allows them to address puzzles left unanswered by standard rational choice models, which assume, unrealistically, that emergent cooperative arrangements represent uniquely efficient solutions to the problems faced by states.

However, the rationalist foundation that undergirds their preferred theories leads these scholars, like intergovernmentalists, to conceptualize institutions in very thin terms. Indeed, EU institutions are constructed by the most powerful member states -- in an instrumental fashion -- to promote their interests. Over time, these bodies will at most coordinate the behaviour and expectations of member states. Recent neorealist work on the Union differs little on this score, viewing EU bodies as a strategic environment constructed, in this case, by weaker states to constrain the choices and behaviour of the more powerful. [13]

How do proponents of supranational institutionalism, multi-level governance and historical institutionalists conceptualize EU institutions? One would think quite differently -- given the great amount of heat and fire in their debate with intergovernmentalists. However, this often turns out not to be the case. To begin, consider the work of Sandholtz. While he argues the need to endogenize interest formation, it is not at all clear that institutions play any meaningful role in this process. In his research on telecommunications, the Maastricht treaty and EU membership, he portrays Union organs as helping states overcome collective action problems and, more generally, affecting the calculations, behaviours and strategies of self-interested domestic actors.

Not surprisingly, given this thin conception of institutions, his three "direct influences" of the EU on "national interest formation" are all captured and explained by variants of rational choice institutionalism. In turn, this suggests Sandholtz is not so much analyzing the impact of institutions on national interests, but how they affect the strategies employed by member states in the pursuit of given interests. [14]

Proponents of multi-level governance, while adding a new and important institutional layer to neofunctionalist and intergovernmentalist arguments, do not challenge the rationalist microfoundations of either school. Consistent with such an approach, these scholars typically argue that their perspective supplements the state-centered, high-politics focus of intergovernmentalism by considering subnational actors and other points in the policy process. [15]

Through richly detailed studies, this literature has convincingly shown that multi-level governance is a reality in a variety of policy areas. However, it has not paid sufficient attention to the role of institutions in this process; moreover, the institutional analysis on offer is largely predicated on rational choice foundations. The dominant institutional metaphor and empirical focus is institutions as constraint; they induce actors to adopt certain strategies or shape their behaviour.

While early work by proponents of multi-level governance tended toward the descriptive, more recent studies are self-consciously theoretical. This theoretical base, consistent with my observations, is solidly rational choice -- theories of transaction costs and informational asymmetries, as well as principal-agent frameworks. In sum, the fight these scholars pick with intergovernmentalists is more about which institutions matter than about how they have effects. [16]

Finally, one has historical institutionalist work on the European Union. This recent addition to EU studies is best exemplified by the work of Paul Pierson. As noted earlier, this branch of institutionalism sits, somewhat uneasily, between rational choice and sociological institutionalism. However, Pierson's version moves it decisively closer to rational choice scholarship; one result is a very thin conception of institutions. His discussion, within the context of the EU, of unanticipated consequences, adaptive learning, institutional barriers to reform, sunk costs and exit costs is entirely consistent with -- and, in fact, premised on -- rational choice. EU institutions are thus all about constraint and incentives. While Pierson should be commended for providing solid rational choice microfoundations to a largely descriptive historical institutionalist literature, the costs are significant.

Indeed, he essentially severs the bridge that links parts of that same literature to a sociological or thick understanding of institutions. Put differently, Pierson is entirely correct to assert that the "path to European integration has embedded member states in a dense institutional environment"; however, he provides us with only a limited set of methodological tools for understanding exactly how that dense environment matters. [17]

In nearly all the work reviewed above, one can deduce a dominant mechanism that links European institutions to agents. Simply put, the former provide information and resources to the latter, who then alter behaviour and strategies. Even in those rare instances where the development of autonomous and potentially constitutive European-level institutions is the focus of study, they are still viewed as information-providers that at best alter agent calculations. [18]

In sum, what is wrong with all this institutional analysis on Europeanization? Nothing. Rather, the proper question is "What is missing here?" In virtually all the foregoing, institutions are at best intervening variables. Missing is a thick institutional argument, derived from sociology, that demonstrates how European institutions can constitute the identities and interests of member states and groups within them.


Social Construction and Institutional Analysis

In this section, I develop an institutional approach that addresses the above-noted gaps, and do so by drawing upon a growing and vibrant body of international relations scholarship: social constructivism. As presently elaborated, constructivism -- at least the modernist branch of concern here -- is an argument about institutions, one which builds upon the insights of sociological institutionalism. It is thus well suited, in a conceptual sense, for expanding our repertoire of institutional frameworks for explaining European integration. Below, I first define a constructivist approach to politics and, then, note several problems in recent empirical work -- problems that Europeanists will want to avoid. [19]

Social Construction -- Overview. While constructivism has been developed primarily as a response and critique of mainstream IR theorizing, there is nothing in the method that limits its application to that particular -- international political -- sphere. Constructivism is concerned not with specific IR theories or levels of analysis, but with underlying conceptions of how the social and political world works. It is not a theory, but an approach to social inquiry based on two assumptions: (1) the environment in which agents/states take action is social as well as material; and (2) this setting can provide agents/states with understandings of their interests (constitute them). Put differently, these scholars question the materialism and methodological individualism upon which much contemporary political science scholarship has been built. [20]

The first assumption reflects a view that material structures, beyond certain biological necessities, are given meaning only by the social context through which they are interpreted. Consider nuclear weapons -- the ultimate material capability. Constructivists argue that it is not such weapons themselves that matter. After all, the US worries very little about the large quantity of nuclear weapons held by the British; however, the possibility that North Korea might come to possess just one or two generates tremendous concern. [21]

The second assumption addresses the basic nature of human agents and states, in particular, their relation to broader structural environments. Constructivists emphasize a process of interaction between agents and structures; the ontology is one of mutual constitution, where neither unit of analysis -- agents or structures -- is reduced to the other and made ontologically primitive. This opens up what for most theorists is the black box of interest and identity formation; state/agent interests emerge from and are endogenous to interaction with structures. [22]

Constructivists thus question the methodological individualism that underpins much contemporary theorizing. This agent-centered view asserts that all social phenomena are explicable in ways that only involve individual agents and their goals and actions; the starting point of the analysis is actors (states) with given properties. Ontologically, the result is to reduce one unit of analysis -- structures -- to the other -- agents. [23]

It is important to note that the constructivists reviewed here do not reject explanatory analysis or causal explanation; their quarrel with mainstream theories is ontological, not epistemological. The last point is key for it suggests constructivism has the potential to bridge the still vast divide separating mainstream theorists from proponents of more discursive and interpretative methods. With the latter, constructivists share many substantive concerns (role of identity and discourse, say) and a similar ontological stance; with the former, they share a largely common epistemology. [24]

To illuminate these differences between constructivists and other schools, it is helpful to explore their understanding of central terms. Consider norms -- a concept that has gained much currency in comparative and IR scholarship over the past decade. Many theorists sees norms playing an influential role in certain issue areas. Yet, they are still a superstructure built on a material base; at best norms serve a regulative function, helping actors with given interests maximize utility. Agents (states) create structures (norms and institutions). [25]

For constructivists, norms are collective understandings that make behavioural claims on actors. Their effects reach deeper: They constitute actor identities and interests, and do not simply regulate behaviour. As explanatory variables, their status moves from intervening to independent. Norms are no longer a superstructure on a material base; rather, they are helping to create and define that base. For constructivists, agents (states) and structures (global norms) are interacting; they are mutually constituted. [26]

Taken together, these three moves by constructivists -- their questioning of methodological individualism and materialism, along with a continuing commitment to explanatory modes of analysis -- have brought a breath of fresh air to thinking about politics, and done so in ways accessible to nearly all scholars. By aspiring to more than interpretation and thick description, their particular brand of institutional analysis offers much to rational choice and historical institutionalists studying Europe: All three schools, for the most part, are talking the same language. [27]

In sum, social constructivists remind us that the study of politics -- or integration -- is not just about actors with fixed preferences who interact via a process of strategic exchange. Rather, they seek to explain theoretically both the content of actor identities/preferences and the modes of social interaction -- so evident in everyday life -- where something else aside from strategic exchange is taking place. To put the point somewhat differently, constructivism is not just the endogenous study of preferences; it also seeks to develop alternative models of action and socialization.

Social Construction -- The Challenges Ahead. However, if Europeanists are fruitfully to utilize constructivist insights, they should best avoid repeating problems extant in the current theoretical work. Here, I highlight the three most important: neglect of agency; failure to explain is a systematic way what it means for an agent to be constituted by broader social structures; and insufficient attention to theory development and research design.

First, in seeking to break with the methodological individualism of much contemporary political science, constructivists have gone too far in the opposite direction, failing to theorize in any meaningful sense the role of agency. As a result, despite their ontological commitment to a process-oriented vision of social reality ("mutual constitution"), they have largely ignored and failed to model the specific mechanisms through which social institutions such as norms connect to agents. [28]

Second and related, there is a lack of clarity on the meaning of constitute, with at least two definitions employed in the literature. In many cases, it is used in an enabling sense, where the emergence of norms and institutions makes possible new repertoires of action and behaviour, or empowers new categories of social actors. This meaning has been invoked most often in constructivist work on the global diffusion of human rights norms, where the appearance of such new knowledge structures gradually reconstitutes the interests of key agents (the US, say) or empowers and helps create new domestic and transnational groups that oppose particular regimes and policies. Used in this manner, constitute has a prolonged temporal dimension -- typically playing out over several decades. [29]

Another definition, however, suggests a more limited temporal perspective, where a social institution (norm, say) constitutes an agent by providing him/her with new understandings of interests regarding a particular policy issue. Unfortunately, these accounts are less clear on how this near-term process of constitution happens. Allusions to social learning and socialization are strong, but ultimately underspecified; as a result, constructivists have missed an opportunity to build bridges to rational choice scholars. In fact, upon closer examination, part of this constructivist socialization dynamic often involves consequential calculation by purposeful agents. Of course, from a problem-driven perspective, there is nothing wrong with that state of affairs; however, it does muddy the waters, making less clear the uniquely constructivist value-added. Nevertheless, as this second understanding of constitute is more amenable to operationalization and empirical research, I emphasize it below. [30]

Third, constructivists must pay more attention to theory development and research design if their nascent research program is to advance. On design, there has been a tendency to ignore cases where "the dog does not bark" -- that is, where the outcome of interest (identity change, say) does not occur. Methodologically, scholars have thus frequently selected on the dependent variable. Compounding this problem has been a research strategy where "N" all too often "= 1." As much of the integration literature is prone to precisely the same problems, it behooves Europeanists, as they execute their own constructivist turn, not to repeat them. [31]

On theory development, one "straw man" needs immediately to be slain. The goal of much mainstream IR theorizing is not the development of general theory of the covering-law sort. While some aspire to such goals, many do not; rather, they seek to develop so-called middle-range theory that accounts for the contingent and historical nature of international political change. The list of middle-range theories (and theorists) is large and growing: work on comparative foreign policy by Alexander George, James Rosenau and others; the literature linking domestic structures to international-political and foreign-policy change (Katzenstein, Risse); recent work on the democratic peace; research linking the global economy to processes of domestic political change (Gourevitch, Haggard, Milner), etc. These scholars have advanced complex, multi-variate arguments and, typically, have applied them to multiple (qualitative) cases. The result is middle-range theory: It is contingent, historically-bounded, but "travels," being applicable in contexts other than those for which it was originally designed. [32]

Social constructivists should also aspire to such (modest) theoretical goals, developing middle-range constructivist theories of the democratic peace or, in the European context, of intergovernmental bargaining and European institutions (see below). Nothing in constructivism precludes a focus on particular actors (states, say) or levels of analysis. While middle-range theory is not without drawbacks, its advantages are numerous. Indeed, by integrating rich empirical data with theoretically-driven arguments, it often generates the intellectual tensions and synergies that help push forward research programs. [33]


Constructivism and the Dynamics of European Integration

Constructivism has the potential to contribute to the study of integration in various areas. Below, I consider two: learning and socialization processes at the European level; and the soft or normative side of Europeanization at the national level. In each case, I explore what a constructivist approach entails, how it could be carried out empirically and its value added compared to existing work on integration. I also address and counter the argument that selection bias limits my ability to generalize. The section concludes by noting how a constructivist approach to integration can build upon and systematize theoretical arguments and descriptive insights advanced by a growing number of Europeanists; I also argue the whole exercise is not one of re-inventing the wheel.

Learning and Socialization. What does it mean for an agent to learn? Complex social learning involves a process whereby actors, through interaction with broader institutional contexts (norms or discursive structures), acquire new interests and preferences -- in the absence of obvious material incentives. Put differently, agent interests and identities are shaped through interaction. So defined, social learning involves a break with strict forms of methodological individualism. This type of learning needs to be distinguished, analytically, from the simple sort, where agents acquire new information, alter strategies, but then pursue given, fixed interests; simple learning, of course, can be captured by methodological-individualist/rationalist accounts. [34]

Consider small group settings: It is intuitively obvious that there are times when agents acquire new preferences through interaction in such contexts. This is not to deny periods of strategic exchange, where self-interested actors seek to maximize utility; yet, to emphasize the latter dynamic to the near exclusion of the former is an odd distortion of social reality. Now, the perhaps appropriate response is "so what?" In an abstract sense, it readily can be appreciated that social learning takes place at certain times, but how can one conceptualize and empirically explore whether and when it occurs? Luckily, there is a growing literature in contemporary IR -- work by constructivists, students of epistemic communities and empirically oriented learning theorists -- that performs precisely this theoretical/empirical combination. More specifically, this research suggests a number of hypotheses on when social learning occurs; these could be translated to empirical work conducted at the European level.

* Social learning is more likely in groups where individuals share common professional backgrounds -- for example, where all/most group members are lawyers or, say, European central bankers.

* Social learning is more likely where the group feels itself in a crisis or is faced with clear and incontrovertible evidence of policy failure.

* Social learning is more likely where a group meets repeatedly and there is high density of interaction among participants.

* Social learning is more likely when a group is insulated from direct political pressure and exposure. [35]

These hypotheses are just a start; they require further elaboration and operationalization. For example, can a crisis situation be specified a priori, and not in a post-hoc fashion as is typically done? When is the density of interaction among group participants sufficiently high for a switch to occur from strategic exchange to interactive learning? These are difficult issues, but they are only being raised because a first round of theoretical/empirical literature exists. Europeanists could build upon and contribute to this work -- for example, by exploring and theorizing the impact, if any, of different EU voting rules (unanimity, qualified majority voting) on these group dynamics.

The deductions also point to a powerful role for communication in constructivist research. However, in keeping with this essay’s attempted bridging function, it is a role between that of the rationalists’ cheap talk, where agents (typically) possess complete information and are (always) instrumentally motivated, and the post-modernists’ discourse analyses, where agents seem oddly powerless and without motivation. Yet, this role itself requires further unpackaging: Underlying my communication/learning arguments are implicit theories of persuasion and argumentation, which have been the subject of extensive study by cognitive/social psychologists. [36]

What are the data requirements for research based on the above hypotheses? Essentially, you need to read things and talk with people. The latter requires structured interviews with group participants; the interviews should all employ a similar protocol, asking questions that tap both individual preferences and motivations, as well as group dynamics. The former, ideally, requires access to informal minutes of meetings or, second best, the diaries or memoirs of participants. As a check on these first two data streams, one can search for local media/TV interviews with group participants. This method of triangulation is fairly standard in qualitative research; it both reduces reliance on any one data source (interviewees, after all, may often dissimulate) and increases confidence in the overall validity of your inferences. [37]

For students of integration, is this a feasible undertaking? Drawing upon my own work in progress, I suggest the answer is "yes." In a larger project, I am studying the appearance and consolidation of new European citizenship norms; an important concern is to explain, at the European level, whether and how new understandings of citizenship are emerging. To date, my focus has been on Strasbourg and the Council of Europe (CE), for this has been where the more serious, substantive work has occurred. When the CE is trying to develop new policy, it often sets up "committees of experts" under the Committee of Ministers, the intergovernmental body that sits atop the Council’s decisionmaking hierarchy. In a sense, then, these committees are the functional equivalent of the working groups of the EU’s Council of Ministers.

I have been examining the Committee of Experts on Nationality, the group that was charged with revising earlier European understandings of citizenship that dated from the 1960s. My particular interest was to describe and explain what occurred in this group as it met over a four year period: For example, why did it revise existing understandings on dual citizenship to remove the strict prohibition that had previously existed at the European level? To address such issues, I did the following. First, three rounds of field work were conducted in Strasbourg; during these trips, I interviewed various individuals who served on the Committee -- members of the Council Secretariat and experts. Second, I conducted interviews in several member-state capitals, meeting with national representatives to the committee of experts. Third, as a cross-check on interview data, more recently I was granted access to the confidential meeting summaries of the Committee.

This was a considerable amount of work, but the pay off has been high. Over time, particular individuals clearly shifted from what they viewed as a strategic bargaining game (for example, seeking side payments to advance given interests) to a process where basic preferences were rethought. This shift was particularly evident on the question of dual citizenship, where a growing number of committee members came to view the existing prohibition as simply wrong. Processes of persuasion and learning were key, and such dynamics were greatly facilitated by a growing sense of policy failure -- the number of dual nationals was climbing rapidly despite the existing prohibition -- and the committee’s insulation from publicity and overt political pressure. Indeed, the committee benefited from the public perception of Strasbourg as a quiet backwater of Europeanization -- with the real action occurring in Brussels. This allowed it to meet and work out revised understandings on citizenship prior to any overt politicization of its work. Given that the CE is a more intergovernmental body than the EU, one where member states and their interests should figure prominently, my case is a hard one for demonstrating social learning dynamics. They are thus all the more likely in Brussels.

The point of this example is not to dismiss rationalist accounts of strategic bargaining. Rather, it is to note the value added of a middle-range constructivist supplement to these more standard portrayals: It led me to ask new questions and employ a different set of research techniques. The result was to broaden our understanding of what, precisely, is going on at the European level. [38]

Socialization/Diffusion Pathways. As noted earlier, constructivists view norms as shared, collective understandings that make behavioural claims on actors. When thinking about norms in the EU context, two issues must be addressed: (1) through what process do they arise at the European level; and (2) how do such norms, once they reach the national level, interact with and socialize agents. Now, the distinction between European and national levels is false, as multiple feedback loops cut across them; at the same time, the dichotomy can be justified analytically as it helps one unpack and think through different stages in the process of European norm formation. In what follows, I am less interested in formal legal norms developed and promulgated, for example, by the European Court of Justice; a growing body of literature in both law and political science already addresses such understandings and their impact. Rather, the constructivist value added comes from its focus on the less formalized, but pervasive social norms that are always a part of social interaction. [39]

On the first issue -- the process of norm development -- constructivists have theorized about and provided empirical evidence for the importance of three dynamics. First, individual agency is central: Well-placed individuals with entrepreneurial skills can often turn their individual beliefs into broader, shared understandings. The importance of this particular factor has been documented in case studies covering nearly a 100 year period and a multitude of international organizations and other transnational movements. In the literature, these individuals are typically referred to as moral entrepreneurs. [40]

Second, such entrepreneurs are especially successful in turning individually held ideas into broader normative beliefs when so-called policy windows are open. This means the larger group, in which the entrepreneur operates, faces a puzzle/problem that has no clear answer, or is new and unknown. In this situation, fixed preferences often break down as agents engage in cognitive information searches. While the policy-window concept was first elaborated by public policy (agenda setting) and organizational theorists (garbage-can models), it was only more recently that constructivists applied its insights in the international realm to explain norm formation. [41]

Third, processes of social learning and socialization (see previous section) are crucial for furthering the norm creation process first begun by individual agents exploiting open policy windows. The basic point is that individual agency is insufficient to create durable social norms. A brief example clarifies the point. In the mid-1980s, several close advisers to Soviet leader Gorbachev played the part of entrepreneurs seeking to advance new ideas about international politics. In the near-term, such individually held beliefs, which were influential in shaping Gorbachev’s own preferences, were decisive for bringing the Cold War to a dramatic, peaceful and unexpected end. Yet, once the USSR collapsed and Gorbachev was swept from power, these ideas largely vanished, as many analysts of Russian foreign behaviour have noted. Put differently, absent social learning among a larger group of actors -- that is, the development of norms -- the particular ideas held by specific agents had no real staying power. [42]

When and if new European norms emerge, one must still theorize about the mechanisms through which they diffuse to particular national settings and (perhaps) socialize agents. Here, constructivists have identified two dominant diffusion pathways: "bottom-up" societal mobilization and "top-down" elite learning. In the first case, non-state actors and policy networks are united in their support for norms; they then mobilize and coerce decisionmakers to change state policy. Norms are not necessarily internalized by the elites. The activities of Greenpeace or any number of European non-governmental organizations (NGOs) exemplify this political pressure mechanism. [43]

The second diffusion mechanism identified by constructivists is more top-down in nature. Social learning, not political pressure, leads agents -- typically elite decisionmakers -- to adopt prescriptions embodied in norms; they become internalized and constitute a set of shared intersubjective understandings that make behavioural claims. This process is based on notions of complex learning drawn from cognitive and social psychology, where individuals, when exposed to the prescriptions embodied in norms, adopt new interests. [44]

A key challenge is to develop predictions for when one or the other of these mechanisms is likely to be at work. To date, constructivists have been silent on this issue; however, my work on European citizenship norms suggests a possibility. I hypothesize that the structure of state-society relations -- domestic structure -- predicts likely diffusion pathways, with four categories of such structures identified: liberal, corporatist, statist and state-above-society. From these, I deduce and predict cross-national variation in the mechanisms -- social mobilization and elite learning -- through which norms are empowered.

In making these arguments, I draw upon research by students of domestic structures and historical institutionalists. These scholars study how institutions (state bureaucracies, so-called intermediate associations linking state and society) structure the game of politics and, more important, policymaking within countries; their approach should be helpful in thinking more systematically about the process through which European norms are empowered in the domestic arena. [45]

A brief example highlights the utility of the approach as well as the attendant data requirements. In the project on European citizenship norms, I have explored whether and in what way they diffused to several European states, including the Federal Republic of Germany. Consider this German case. I first did research on the basic structure of state-society relations in the country; like many others, I concluded that the polity is corporatist. That is, it possesses a decentralized state and centralized society, with a dense policy network connecting the two parts; both state and society are participants in policymaking, which is consensual and incremental.

Given this coding of the German structure, I next advanced predictions on the expected process whereby norms would have constitutive effects, arguing that societal pressure would be the primary and elite learning the secondary mechanism empowering European norms in Germany. The logic is as follows. In a corporatist domestic structure, state decisionmakers play a greater role in bringing about normative change than in the liberal case, where policymakers are constantly pressured by social actors; however, this does not mean they impose their preferences on a pliant populace. A hallmark of corporatism is the policy networks connecting state and society, with the latter still accorded an important role in decisionmaking. In this setting, I thus hypothesize that it is both societal pressure (primary) and elite learning (secondary) that lead to norm empowerment.

With these predictions in hand, I then conducted extensive field work in the Federal Republic. To date, this research has confirmed my working hypotheses: Emerging European norms on citizenship are diffusing and being empowered in Germany primary via the mobilization of societal pressure from below; social learning at the elite level has been secondary. More specifically, these norms are connecting to a wide variety of social groups and individuals: NGOs favoring the integration of Germany’s large resident foreigner population; activists in the churches and trade union; and immigrant groups. At the decisionmaking level, one finds isolated evidence of elites learning new preferences from the norms (for example, a small group of Christian Democratic Bundestag deputies). [46]

Two streams of evidence are important for establishing the presence of these diffusion mechanisms , as well as their relative weighting. Most important were structured interviews with a wide range of actors -- both societal and state. As at the European level, these discussions were designed to probe the degree to which agent preferences were changing and the motivations for such change. However, as the rationalists remind us, talk is cheap. Therefore, as a cross-check on the interview data, I consulted a wide-range of primary documentation -- official summaries of Bundestag debates, media analyses, and interviews given in newspapers or on TV.

What is the value-added of all this work? It convincingly demonstrates that a rational-choice institutionalist understanding of the role norms play in social life (norms as constraint) missed an important part of the story in the Federal Republic. I indeed found instances where domestic agents simply felt constrained by the European norms (for example, a number of officials in the Federal Interior Ministry); yet, in many other cases, I uncovered evidence of non-strategic social learning where agents, in the norm’s presence, acquired new understandings of interests. Clearly, much theoretical work remains to be done -- in particular, elaborating scope conditions for when norms have constraining as opposed to constitutive effects. Addressing this latter point is crucial, for, again, the obvious empirical fact is that norms do not always constitute. And while some constructivists may not like it, such investigations will require renewed theoretical attention to agency (see concluding section).

Selection Bias? Perhaps, though, my sociological arguments about the constitutive role of institutions only work because of the particular organization and policy area from which I drew empirical examples: the Council of Europe and human rights. Such analysis is largely irrelevant for the European Union -- a special type of institution with very different policy domains. Selection bias, in other words, limits my ability to generalize. Two responses counter such a critique.

First, there are well-established theoretical reasons for suspecting that Europe, especially Western Europe, is a most likely case for international institutions to have constitutive effects. Most important, it is an institutionally dense environment, one where theorists predict high levels of transnational and international normative activity. This logic, precisely because it is a particular way of viewing the social world, is in principle equally applicable to a variety of European institutions -- whether their focus is human rights (CE) or political and economic affairs (EU). [47]

Second, assume, despite the foregoing, that differences in policy domains do matter. That is, arguments about social learning or the constitutive effects of European norms just do not work when applied to the EU. After all, the process of European integration has largely been about market integration, where national and transnational business interests have played key roles. Such groups are quite different in structure and goals from the actors of civil society -- domestic NGOs, churches -- highlighted in many of my examples. However, if the institutional (enhanced role of European Parliament) and substantive (third pillar of justice and home affairs) innovations of Maastricht and Amsterdam continue to evolve, new actors and policy issues are increasingly likely to make themselves felt within the EU. [48]

In fact, human rights pressure groups have begun utilizing the European Parliament as a means for generating precisely the sort of normative pressure-from-below documented in my CE example. Moreover, immigration, which is now on the third pillar agenda, is an issue where previous studies have documented the extensive degree to which European state interests are constituted by broader international norms. On the related issues of citizenship and racism, recent work establishes that the 1996-97 Intergovernmental Conference (IGC) saw extensive mobilization by NGOs and other transnational movements, and their qualitatively different, when compared to the past, interaction with EU institutions, as well as the IGC itself. Thus, even if differences in policy domains are important, these are presently being blurred if not erased. [49]

Summary. My purpose in the foregoing was constructive. The goal was not to dismiss rational choice or historical institutionalist work on integration; those literatures are rich and offer many insights. Yet, because of their adherence to variants of methodological individualism, certain analytic/empirical issues -- interest and identity formation, most importantly -- are bracketed. A more sociological and constructivist understanding of institutions as constitutive allows one to address such questions. Constructivism, however, need not and, indeed, should not be viewed as terra incognita to Europeanists. In fact, a constructivist cut at integration is already evident, albeit implicitly, in both theoretical and empirical/descriptive studies.

Theoretically, one has the recent work of Olsen, Kohler-Koch and Fligstein. Olsen’s writing, including that on the EU, has been concerned with broader institutional environments -- how they provide the very basis of action for political agents, how they lead to rule-governed behaviour, which may supplant instrumental, strategic calculation, and how they promote learning. Yet, he has failed to explicate, in a theoretical sense, the processes through which such institutional dynamics occur. The constructivist work reviewed above suggests a number of ways these micro-macro linkages could be developed in a specifically European context. [50]

Much of the analysis in recent work by Kohler-Koch and Knodt is also premised on sociological assumptions -- in particular, their exploration of the domestic normative impact of EU institutions, where they do not simply constrain, but constitute agents and their preferences. Unfortunately, this argument is much less clear about the process through which, and the conditions under which, EU norms have such effects. Here, constructivist hypotheses on the mechanisms through which national-level socialization and social learning occur might be relevant. [51]

Fligstein is also interested in constitutive dynamics, but in contrast to Kohler-Koch and Knodt, the focus is on Brussels. In his work on the Commission, he argues that, under certain conditions marked by crisis and uncertainty, it can play an entrepreneurial role helping culturally construct political action. Less clear, however, are the specific processes through which such construction takes place, as well as his theoretical understanding of agency’s role. All the same, this analytic move hints at rich possibilities for a dialogue with those social constructivists who theorize the role of individual agency, entrepreneurs and policy windows in their work on normative change. [52]

Empirically, the last decade has seen an explosion of work on institutional fusion, policy networks, comitology and informal communication patterns centered upon and generated by EU institutions. While this research is extraordinarily rich in a descriptive sense, it is often undertheorized. To be fair, solid empirical work is often a prerequisite for theory building. All the same, more attention to theory would help these scholars systematize their implicitly sociological view of institutions -- and constructivism has much to offer here.

Consider three examples. Wessels, Rometsch and their collaborators have made a powerful and well-documented case for institutional fusion within the EU context, where the density of interaction between European and national institutions is such that old distinctions between the two levels no longer hold. These analysts ascribe an important symbolic and identity-shaping role to institutions -- to constructivists, a constitutive role. Yet, they are silent, theoretically, on when, how and why such identity formation occurs, which leads them to advance an underspecified convergence thesis, where "the constitutional and institutional set-up of [EU] member states will converge towards one common model." Given that constructivists have already begun to specify scope conditions regarding institutions and identity change, the potential for theoretical cross-fertilization would seem significant. [53]

In a second example, recent work by Beyers and Dierickx on the EU Council and its working groups suggests that informal communication is key for understanding their operation. Yet, this research, despite its empirical richness, brackets a crucial theoretical question: Under what conditions -- if at all -- does this communication lead political agents away from situations of strategic exchange and into those marked by social learning and communicative action? For both theoretical (debates over the consequences of integration) and policy reasons (explaining when and why member-state interests change), this issue is fundamental. However, because of their reliance on a methodologically individualist ontology, Beyers and Dierickx seem simply unaware that they are in fact well placed to address it. The point is not that they get the story wrong; rather, it is incomplete. And constructivism, with its concern for modelling modes of social interaction beyond strategic exchange, could provide analytic tools for filling out the picture. [54]

Research on so-called comitology represents a third example where constructivist theorizing and empirical integration studies could profitably interact. Comitology refers to the complex set of committee rules that have evolved to implement EU policy and procedures; the system stems from a 1987 European Council decision in which member states made clear their unwillingness to lose control of the implementation process -- in particular, by ceding too much power to the Commission. These committees, by member-state dictate, are composed of government representatives and, occasionally, additional experts; yet, the growing empirical literature on them notes how these representatives must often turn elsewhere for information and, more important, interpretation. Indeed, two analysts argue that "scientific evidence" is accepted as the most valid currency for "effecting convincing arguments" in comitology. [55]

The last point suggests a link to my earlier hypotheses on small groups, communication and social learning. On the one hand, discursive argumentation, committee-level learning and its accompanying preference change do not necessarily lead to a reconstitution of member state interests, for comitology, after all, involves only policy implementation. Yet, it is well known among students of public policy that implementation is often the making of policy by other means; moreover, learning and deliberative arguing of this sort cannot be captured by standard strategic exchange models. Constructivist deductions on the role of common backgrounds, crisis, density of interaction, etc, could thus be readily exploited by Europeanists to explore more systematically the conditions under which European committees do indeed, through learning and argumentation, socialize their participants. [56]

A final issue is not so much one of new theoretical directions for analyses of integration, but, instead, a look back. Simply put, is my call for bringing constructivist insights to bear on the study of the EU a short-sighted reinventing of the neo-functionalist wheel? After all, over 30 years ago, Haas and others were writing about the identity-shaping effects of the European project. Indeed, collective identity was to emerge via a "process whereby political actors in several distinct national settings are persuaded to shift their loyalties, expectations and political activities towards a new center, whose institutions possess or demand jurisdiction over the pre-existing nation-states." [57]

While references to social learning and socialization are evident in the work of many early neofunctionalists and regional integration theorists, the differences with constructivism are significant. Most important, the latter is not a general substantive theory that predicts constant learning or a growing sense of collective identity; rather, its aspirations are more modest. As currently being developed, it is a middle-range theoretical approach seeking to elaborate scope conditions for better understanding precisely when collective identity formation occurs. Constructivism is thus agnostic as to whether the endpoint of social interaction is greater common interests and identity. Neofunctionalists, at least implicity, were not neutral on this question; there was a clear normative element to their scholarship. [58]

In addition, despite the strong allusions to identity formation and change, neofunctionalists failed to develop explicit microfoundations that moved them beyond an agent-centered view of social interaction. In fact, their is a strong element of rational choice in their research. While considerable work remains, constructivists are attempting to elaborate such alternative foundations -- their stress on logics of appropriateness and communicative action, for example. [59]


Conclusions

Instead of summarizing, this concluding section highlights two theoretical puzzles, both of which were implicit in much of the foregoing: variance in socialization/diffusion pathways linking the European and national levels (process), and variance in national-level norm-driven learning and socialization (outcome). If Europeanists structure their research to address these lacunae, they will not only demonstrate constructivism’s utility for explicating key aspects of the integration process; more important, they will contribute to its on-going consolidation as a research program in contemporary IR theory.

Process/Scope-Conditions. A first factor predicting variance in socialization pathways might go under the label bringing institutions back in. While this is sure to produce instantaneous groans among readers, the point, analytically, is an important one. Constructivists seem to have left their institutional insights "at the water’s edge," reminding us that institutions -- internationally or regionally -- matter in ways political science has too often neglected, but failing to think through their equally important role at the national level. Earlier, I suggested one such role: Historically constructed domestic political institutions shape the pattern of interaction between state and society in a systematic way across countries.

I argued that it was precisely the structure of state-society relations that predicted variance in the two dominant socialization pathways uncovered by constructivists. In a liberal polity such as the US, with strong societal inputs to policymaking, the more probable socialization mechanism is the bottom-up, mobilization one; in centralized regimes, where the gap between state and society is large -- contemporary Ukraine, say -- the more likely pathway is top-down. In countries with a more complex relation between state and society (Germany and Russia), I predict a combination of the two socialization pathways. Essentially, an argument of this type introduces domestic institutions as an intervening variable, one that structures and channels the socialization process; it tells a researcher where to look for socialization dynamics in a particular national setting. [60]

Beyond political institutions, other research suggests two additional factors explaining variance in national-level socialization pathways. The first is well-known to students of comparative foreign-policy and public policy: issue area. The argument is that certain policy areas are more likely to witness social mobilization. When it comes to human rights, say, it would be surprising not to find such mobilization at work. In contrast, normative socialization on more technocratic and obscure topics -- for example, security or the intricacies of regulatory reform -- might be expected to occur largely at the level of elites.

In fact, much of the constructivist literature supports such a hypothesis. In studies of global human rights, general environmental and racial-equality norms, these scholars uncover more evidence of bottom-up socialization. However, research on the spread of international security, welfare or specific (and hence technocratic) environmental norms suggests a much greater role for elite socialization and learning as a key mechanism linking the international and national levels. Given that much of the European project centered around the EU has been and remains a technocratic one controlled and shaped by elites, one might a priori hypothesize that the top-down, elite socialization dynamic would be more relevant in that context. [61]

A final factor highlights not a norm’s substantive content, but the degree to which it contains prescriptive guidance -- "robustness." An international or regional norm is robust if it embodies clear prescriptions, which provide guidance to agents as they develop preferences and interests on a issue; in turn, clear prescriptions imply a degree of shared consensus at the regional/systemic level. Thus, high levels of both specificity and intersubjective agreement are indicators of a robust norm. A priori, one might hypothesize that robust norms, precisely because they are held by a meaningful number of the relevant international/regional actors and would thus be carried by broader transnational networks, are more likely to generate societal mobilization at the national level. For example, elsewhere I have argued it is precisely the lack of normative robustness in the Maastricht Treaty’s European citizenship provisions that explains their inability to generate any serious social mobilization in Germany. [62]

Having identified three factors, the analytic challenge is to relate them. Note that the latter two -- norm content and robustness -- are external to the state, while the first -- structure of domestic institutions -- is internal to it. My hunch is that internal will trump external in predicting socialization pathways, with a good test being to pick a case where the external variables predict bottom-up, societal mobilization, while the structure of domestic institutions favors top-down socialization. The requisite combination would be to study the socialization of domestic agents to human rights norms in a country where the gap separating state and society is large. My Ukrainian citizenship case, which arguably fits these criteria, provides preliminary confirmation that the internal/domestic do indeed dominate. [63]

Towards a Constructivist Theory of Learning and Socialization. There is an emerging debate on the relation of constructivism to rational choice, one to which Europeanists can contribute. Indeed, when one reads across the literature, a number of different claims are advanced -- for example, that constructivism, by endogenizing interest formation, supplements rational choice; or, more ambitiously, that it represents an alternative approach to social interaction, one based on a theory of action -- a logic of appropriateness -- different from that on which rational choice is premised -- a logic of consequences. [64]

This essay has argued that constructivism is not just about the inductive study of interest/identity formation, where it would provide the raw material for the standard rationalist "two-step." Rather, it represents an attempt to theorize and empirically explore an approach to social action and socialization that moves beyond and supplements the one advanced by rational choice. Such a stand is motivated by nothing more than a simple empirical fact: In the real, here and now social world, there is often something going on that cannot meaningfully be reduced to strategic exchange among self-interested actors or rational adaptation to external constraints. [65]

The devil, however, is in the theoretical details. An empirically grounded constructivist theory of socialization and learning must model the process by which norms connect to agents, advancing hypotheses for the conditions under which they constitute or constrain them. Just because rationalists fall back on "as if" assumptions to bracket questions of process does not mean that social constructivists should do the same. [66]

A start at such a theory would specify structural pre-conditions, that is the likelihood that agents in a particular national setting will be receptive to socialization by prescriptions embodied in a norm. Both theoretical logic as well as accumulating empirical evidence suggest that the fit between international and domestic normative structures will play a key role here. To theorize this interaction, I revisit the domestic institutional variable highlighted above and extend the analysis to a deeper level, employing a synthesis of sociological and historical institutionalism. The former, with its focus on the constitutive nature of institutions, alerts one to the presence of a domestic normative dimension. The latter, with its stress on processes of institutionalization, demonstrates how these domestic norms gain political saliency and causal influence, thus acting as a filter or block to regional/systemic norms. This is one way of modeling the insights of those few analysts who highlight the potential conflict between norms at differing levels. Specifically, where there is a mismatch or lack of cultural match between regional/systemic and institutionalized domestic norms in a given policy area, I would predict a drastic slowing in the socialization process at the agent level. [67]

In turn, this prediction is premised on an analytic move from structure to process. The presence of such domestic normative barriers will lead to a break-down in (constructivist) complex learning due to cognitive dissonance or framing problems, a dynamic seemingly at work in many contemporary settings. Put more formally, in this situation of conflicting norms, complex learning is less likely than the simple sort. Since the latter can be captured by rationalist accounts, there is nothing uniquely constructivist in normative socialization under such conditions. Conversely, where European/systemic norms face lower domestic normative barriers, the likelihood of agent acceptance of and (complex) learning from the former is increased. [68]

Here, my analysis intersects and complements that of constructivists interested in longer-term normative socialization. In particular, these scholars might argue that processes of argumentation and persuasion will gradually reduce both normative barriers and breakdowns in learning. While such dynamics are theoretically possible, the empirical challenge of documenting them is daunting, particularly if one avoids the use of "as if" assumptions at the level of agents. The empirical problem is especially acute for those constructivists who invoke Habermas’ (normative) theory of communicative action to explain the (real world) behavior of agents, for the former provides one with little sense of "the various social mechanisms that might help us better to understand how social systems and individuals’ actions mesh." [69]

The foregoing is only a start at developing a distinctively constructivist theory of socialization and social learning; much work remains; indeed, the analysis remains at too aggregate a level, arguing, in essence, for a fit between regional/international and dominant domestic norms. Yet, even in those instances where the mismatch is great, one still finds evidence of constructivist agent-level learning and socialization. This suggests the (deductive) structural first cut is best viewed as a base line, which is then supplemented with more contingent factors that emerge (inductively) from on-going empirical research.

In any case, my many references to learning, role conflict, cognitive dissonance and framing suggest that as constructivists begin to model the micro-foundations of normative socialization, the benefits of an exchange with cognitive/social psychology would be great. Absent such a move, these scholars will have no real understanding of how domestic agents in the near-term here and now are socialized by European or systemic norms. The result will be to cede this promising ground to the rationalists, which would be a pity. [70]

* * *

My arguments throughout this essay were based on an obvious but too often neglected truism about our social world: The most interesting puzzles lie at the nexus where structure and agency intersect. Put differently, the real action, theoretically and empirically, is where norms, discourses, language and material capabilities interact with motivation, social learning and preferences -- be it in international or European regional politics. Research traditions such as rational choice, postmodernism and, more recently, large parts of constructivism, which occupy endpoints in the agent-structure debate, have life easy: They can ignore this messy middle ground. Yet, the true challenge for both rationalists and their opponents is to model and explore this complex interface; doing so, however, will require that social constructivists meet the other party half way, which means, as this paper has argued, paying greater attention to agency/micro-foundations and thinking more systematically about theory and research design. [71]

As one scholar recently put it, "regional integration studies could uncharitably be criticized for providing a refuge to homeless ideas." While constructivism is certainly not homeless, Europeanists should resist the temptation simply to pull it off the shelf, giving it a comfortable European home in yet another N=1, noncumulable case study. Rather, these scholars have the opportunity -- given their immensely rich data set -- to push forward one of the most exciting debates in contemporary international and political theory. [72]

References

Adler, Emanuel. 1987. The Power of Ideology: The Quest for Technological Autonomy in Argentina and Brazil. Berkeley: University of California Press.

-------. 1997. "Seizing the Middle Ground: Constructivism in World Politics." European Journal of International Relations 3 (September).

Adler, Emanuel and Michael Barnett. 1996. "Governing Anarchy: A Research Agenda for the Study of Security Communities." Ethics & International Affairs 10: 63-98.

Anderson, Jeffrey. 1995. "The State of the (European) Union: From the Single Market to Maastricht, from Singular Events to General Theories." World Politics 47 (April).

Aspinwall, Mark and Gerald Schneider. 1997. "Same Menu, Separate Tables: The Institutionalist Turn in Political Science and the Study of European Integration." Paper presented at the Research Workshop of the European Consortium for Political Research, Bergen, Norway, September.

Barnett, Michael. 1993. "Institutions, Roles and Disorder: The Case of the Arab States System." International Studies Quarterly 37 (September).

-------. 1995. "Sovereignty, Nationalism and Regional Order in the Arab States System." International Organization 49 (Summer): 479-510.

Bennett, Andrew. 1997. Condemned to Repetition? The Rise, Fall and Reprise of Soviet-Russian Military Interventionism, 1973-96 . Washington, DC: Manuscript, Department of Government, Georgetown University, September.

Beyers, Jan and Guido Dierickx. 1997. "Nationality and European Negotiations: The Working Groups of the Council of Ministers." European Journal of International Relations 3 (December).

-------. 1998. "The Working Groups of the Council of the European Union: Supranational or Intergovernmental Negotiations?" Journal of Common Market Studies 36 (??).

Blank, Kermit, Liesbet Hooghe and Gary Marks. 1996. "European Integration from the 1980s: State-Centric v. Multi-Level Governance." Journal of Common Market Studies 34 (September).

Bleich, Erik. 1998. "From International Ideas to Domestic Politics: Gatekeepers, Priors and Educational Multilateralism in England and France." Comparative Politics (forthcoming).

Bulmer, Simon. 1994. "The Governance of the European Union: A New Institutionalist Approach." Journal of Public Policy 13: 351-80.

Burley, Anne-Marie and Walter Mattli. 1993. "Europe before the Court: A Political Theory of Legal Integration." International Organization 47 (Winter).

Caporaso, James. 1998. "Regional Integration Theory: Understanding Our Past and Anticipating Our Future." Journal of European Public Policy 5 (March).

Caporaso, James, Maria Green Cowles and Thomas Risse. Editors. Forthcoming. Europeanization and Domestic Structural Change.

Charney, Jonathan. 1993. "Universal International Law." American Journal of International Law 87 (October).

Checkel, Jeffrey T. 1997a. Ideas and International Political Change: Soviet/Russian Behaviour and the End of the Cold War. New Haven: Yale University Press.

-------. 1997b. "International Norms and Domestic Politics: Bridging the Rationalist-Constructivist Divide." European Journal of International Relations 3 (December).

-------. 1998a. "The Constructivist Turn in International Relations Theory (A Review Essay)." World Politics 50 (January).

-------. 1998b. "International Norms and Domestic Agents: Probing the Dynamics of Socialization." Paper presented at the Workshop on "Ideas, Culture and Political Analysis," Center of International Studies, Princeton University, May.

-------. 1998c. "Norms, Institutions and National Identity in Contemporary Europe." International Studies Quarterly 42 (December).

Cohen, Wesley and Daniel Levinthal. 1990. "Absorptive Capacity: A New Perspective on Learning and Innovation." Administrative Science Quarterly 35 (March).

Conzelmann, Thomas. 1998. "'Europeanization' of Regional Development Policies? Linking the Multi-Level Governance Approach with Theories of Policy Learning and Policy Change," European Integration Online Papers 2, No.4 (10 June).

DiMaggio, Paul and Walter Powell. Editors. 1991. The New Institutionalism in Organizational Analysis. Chicago: University of Chicago Press.

DiMaggio, Paul. 1997. "Culture and Cognition." Annual Review of Sociology 23: 263-87.

Dobbin, Frank. 1994. "Cultural Models of Organization: The Social Construction of Rational Organizing Principles." In The Sociology of Culture: Emerging Theoretical Perspectives, edited by Diana Crane. Cambridge: Blackwell Publishers.

Dogan, Rhys. 1997. "Comitology: Little Procedures with Big Implications." West European Politics 20 (July).

Drzemczewski, Andrew. 1983. European Human Rights Convention in Domestic Law: A Comparative Study. NY: Oxford University Press.

Elster, Jon. Editor. 1986. Foundations of Social Choice Theory. Cambridge: Cambridge University Press.

Eriksen, Erik Oddvar and Jarle Weigaard. 1998. "Conceptualizing Politics: Strategic or Communicative Action?" Scandinavian Political Studies, 20 (3).

Falkner, Gerda. 1998. "How Intergovernmental are Intergovernmental Conferences? Lessons from the Maastricht Social Agreement." Paper presented at the European Consortium for Political Research 26th Joint Sessions, University of Warwick, March.

Favell, Adrian. 1998. "The European Citizenship Agenda: Emergence, Transformation and Effects of a New Political Field." Paper presented at the 11th International Conference of Europeanists, February.

Finnemore, Martha. 1996a. National Interests in International Society. Ithaca, NY: Cornell University Press.

-------. 1996b. "Norms, Culture and World Politics: Insights from Sociology's Institutionalism." International Organization 50 (Spring).

Finnemore, Martha and Kathryn Sikkink. 1998. "Norms and International Relations Theory." International Organization 52 (Autumn).

Fligstein, Neil. 1998. "Institutional Entrepreneurs and Cultural Frames: The Case of the European Union=s Single Market Program." Paper presented at the Workshop on "Ideas, Culture and Political Analysis," Center of International Studies, Princeton University, May.

Florini, Ann. 1996. "The Evolution of International Norms." International Studies Quarterly 40 (September).

Franchino, Fabio. 1998. "The Commission’s Executive Discretion, Information and Policy Preferences." Paper presented at the European Consortium for Political Research 26th Joint Sessions, University of Warwick, March.

Friedman, Jeffrey. Editor. 1996. The Rational Choice Controversy: Economic Models of Politics Reconsidered. New Haven: Yale University Press.

Garrett, Geoffrey. 1992. "International Cooperation and Institutional Choice: The EC's Internal Market." International Organization 46 (Spring).

Garrett, Geoffrey and Barry Weingast. 1993. "Ideas, Interests and Institutions: Constructing the EC's Internal Market." In Ideas and Foreign Policy: Beliefs, Institutions and Political Change, edited by Judith Goldstein and Robert Keohane. Ithaca, NY: Cornell University Press.

Goldstein, Judith. 1993. Ideas, Interests and American Trade Policy. Ithaca, NY: Cornell University Press.

Green, Donald and Ian Shapiro. 1994. Pathologies of Rational Choice Theory: A Critique of Applications in Political Science. Hew Haven: Yale University Press.

Grieco, Joseph. 1995. "The Maastricht Treaty, Economic and Monetary Union and the Neorealist Research Program." Review of International Studies 21 (January).

-------. 1996. "State Interests and Institutional Rule Trajectories: A Neorealist Interpretation of the Maastricht Treaty and European Economic and Monetary Union." Security Studies 5 (Spring).

Haas, Ernst. 1958. The Uniting of Europe: Political, Social and Economic Forces, 1950-57. Stanford: Stanford University Press.

Haas, Peter. 1990. Saving the Mediterranean: The Politics of International Environmental Cooperation. NY: Columbia University Press.

-------. Editor. 1992. "Knowledge, Power and International Policy Coordination." International Organization, Special Issue (Winter).

Hall, Peter. 1993. "Policy Paradigms, Social Learning and the State: The Case of Economic Policymaking in Britain." Comparative Politics 25 (April).

Hall, Peter and Rosemary Taylor. 1996. "Political Science and the Three New Institutionalisms." Political Studies 44 (December).

Hattam, Victoria. 1993. Labor Visions and State Power: The Origins of Business Unionism in the United States. Princeton: Princeton University Press.

Hedstroem, Peter and Richard Swedberg. Editors. 1998. Social Mechanisms: An Analytical Approach to Social Theory. Cambridge: Cambridge University Press.

Hix, Simon. 1998. "The Study of the European Union II: The 'New Governance' Agenda and its Rival." Journal of European Public Policy 5 (March).

Hooghe, Liesbet and Gary Marks. 1996. "'Europe with the Regions': Channels of Regional Representation in the European Union." Publius: The Journal of Federalism 26 (Winter).

Immergut, Ellen. 1992. Health Politics: Institutions and Interests in Western Europe. NY: Cambridge University Press.

Joergensen, Knud Erik. Editor. 1997. Reflective Approaches to European Governance. London: Macmillan Press.

Joerges, Christian and Juergen Neyer. 1997. "Transforming Strategic Interaction into Deliberative Problem-Solving: European Comitology in the Foodstuffs Sector." Journal of European Public Policy 4 (December).

Johnson, James. 1993. "Is Talk Really Cheap: Prompting Conversation between Critical Theory and Rational Choice." American Political Science Review 87 (March).

Johnston, Alastair Iain. 1998. "Socialization in International Institutions: The ASEAN Regional Forum and IR Theory." Paper presented at the Workshop on "The Emerging International Relations of the Asia-Pacific Region," University of Pennsylvania, May.

Kato, Junko. 1996. "Review Article: Institutions and Rationality in Politics - Three Varieties of Neo-Institutionalists." British Journal of Political Science 26: 553-82.

Katzenstein, Peter. 1985. Small States in World Markets: Industrial Policy in Europe. Ithaca, NY: Cornell University Press.

-------. 1993. "Coping with Terrorism: Norms and Internal Security in Germany and Japan." In Ideas and Foreign Policy: Beliefs, Institutions and Political Change, edited by Judith Goldstein and Robert Keohane. Ithaca, NY: Cornell University Press.

-------. Editor. 1996. The Culture of National Security: Norms and Identity in World Politics. NY: Columbia University Press.

Keck, Margaret and Kathryn Sikkink. 1998. Activists Beyond Borders: Transnational Advocacy Networks in International Politics. Ithaca: Cornell University Press.

Klotz, Audie. 1995a. "Norms Reconstituting Interests: Global Racial Equality and US Sanctions against South Africa." International Organization 49 (Summer).

-------. 1995b. Norms in International Relations: The Struggle Against Apartheid. Ithaca, NY: Cornell University Press.

Koelble, Thomas. 1995. "The New Institutionalism in Political Science and Sociology." Comparative Politics 27 (January).

Kohler-Koch, Beate and Michele Knodt. 1997. "Multi-Level Governance: The Joy of Theorizing and the Anguish of Empirical Research." Paper presented at the European Consortium for Political Research 25th Joint Sessions, Universitaet Bern, February.

Krasner, Stephen. Editor. 1983. International Regimes. Ithaca, NY: Cornell University Press.

Legro, Jeffrey. 1996. "Culture and Preferences in the International Cooperation Two-Step." American Political Science Review 90 (March).

-------. 1997. "Which Norms Matter? Revisiting the 'Failure' of Internationalism." International Organization 51 (Winter).

Leibfried, Stephan and Paul Pierson. Editors. 1995. European Social Policy: Between Fragmentation and Integration. Washington, DC: Brookings Institution.

Levy, Jack. 1994. "Learning and Foreign Policy: Sweeping a Conceptual Minefield (Review Article)." International Organization 48 (Spring).

-------. 1997. "Prospect Theory, Rational Choice and International Relations." International Studies Quarterly 41 (March).

Longstreth, Frank, et al. Editors. 1992. Structuring Politics: Historical Institutionalism in Comparative Analysis. NY: Cambridge University Press.

March, James and Johan Olsen. 1989. Rediscovering Institutions: The Organizational Basis of Politics. NY: Free Press.

-------. 1998. "The Institutional Dynamics of International Political Orders." International Organization 52 (Autumn).

Marks, Gary. 1993. "Structural Policy and Multilevel Governance in the EC." In The State of the European Community II: The Maastricht Debates and Beyond, edited by Alan Cafruny and Glenda Rosenthal. Boulder: Lynne Rienner.

Mattli, Walter and Anne-Marie Slaughter. 1998. "Revisiting the European Court of Justice (Review Essay)." International Organization 52 (Winter).

Meyer, John and David Strang. 1993. "Institutional Conditions for Diffusion." Theory and Society 22 (August).

Moravcsik, Andrew. 1991. Negotiating the Single European Act: National Interests and Conventional Statecraft in the European Community." International Organization 45 (Winter).

-------. 1993. "Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach." Journal of Common Market Studies 31 (December).

-------. 1994. "Why the European Community Strengthens the State: Domestic Politics and International Cooperation." Working Paper #52, Center for European Studies, Harvard University.

-------. 1995. "Explaining International Human Rights Regimes: Liberal Theory and Western Europe." European Journal of International Relations 1 (June).

-------. 1997. "Taking Preferences Seriously: A Liberal Theory of International Politics." International Organization 51 (Autumn).

Morrow, James. 1994. Game Theory for Political Scientists. Princeton: Princeton University Press.

Nadelmann, Ethan. 1990. "Global Prohibition Regimes: The Evolution of Norms in International Society." International Organization 44 (Autumn).

North, Douglass. 1990. Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press.

Obradovic, Daniela. 1997. "Eligibility of Non-State Actors to Participate in European Union Policy Formation." Paper presented at the Workshop on "Non-State Actors and Authority in the Global System," University of Warwick, October.

Olsen, Johan. 1995. "European Challenges to the Nation State." Oslo: ARENA Working Paper 95/14, September.

-------. 1996. "Europeanization and Nation-State Dynamics." In The Future of the Nation State: Essays on Cultural Pluralism and Political Integration, edited by Sverker Gustavsson and Leif Lewin. Stockholm: Nerenius and Santerus Publishers.

Pedler, Robin and Guenther Schaefer. Editors. 1996. Shaping European Law and Policy: The Role of Committees and Comitology in the Political Process. Maastricht: European Institute of Public Administration.

Peterson, M.J. 1997. "The Use of Analogies in Developing Outer Space Law." International Organization 51 (Spring).

Pierson, Paul. 1994. Dismantling the Welfare State? Reagan, Thatcher and the Politics of Retrenchment. NY: Cambridge University Press.

-------. 1996. "The Path to European Integration: A Historical Institutionalist Analysis." Comparative Political Studies 29 (April).

Pollack, Mark. 1997. "Delegation, Agency and Agenda Setting in the European Community." International Organization 51 (Winter).

-------. 1998. "Constructivism, Social Psychology and Elite Attitude Change: Lessons from an Exhausted Research Program." Paper presented at the 11th International Conference of Europeanists, February.

Powell, Robert. 1994. "Anarchy in International Relations Theory: The Neorealist-Neoliberal Debate (Review Article)." International Organization 48 (Spring).

Reus-Smit, Christian. 1997. AThe Constitutional Structure of International Society and the Nature of Fundamental Institutions.@ International Organization 51 (Autumn).

Risse, Thomas. 1997. "Let’s Talk! Insights from the German Debate over Communicative Action." Paper presented at the Annual Convention of the American Political Science Association, September.

-------. 1998. ATo Euro or Not to Euro: The EMU and Identity Politics in the European Union.@ Oslo: ARENA Working Paper 98/1, January.

Risse, Thomas and Kathryn Sikkink (1999) "The Power of Principles: The Socialization of Human Rights Norms into Domestic Practices." Chapter 1 in The Power of Principles: International Human Rights Norms and Domestic Change, edited by Thomas Risse, Stephen Ropp and Kathryn Sikkink. Cambridge: Cambridge University Press.

Risse-Kappen, Thomas. 1991. "Public Opinion, Domestic Structures and Foreign Policy in Liberal Democracies." World Politics 43 (July): 479-512.

-------. 1994. "Ideas Do Not Float Freely: Transnational Coalitions, Domestic Structures and the End of the Cold War." International Organization 48 (Spring).

-------. Editor. 1995a. Bringing Transnational Relations Back In: Non-State Actors, Domestic Structures and International Institutions. Cambridge: Cambridge University Press.

-------. 1995b. "Democratic Peace - Warlike Democracies? A Social Constructivist Interpretation of the Liberal Argument." European Journal of International Relations 1 (December).

-------. 1996. "Exploring the Nature of the Beast: International Relations Theory and Comparative Policy Analysis Meet the European Union." Journal of Common Market Studies 34 (March).

Rittberger, Volker, Andreas Hasenclever and Peter Mayer. 1996. "Interests, Power, Knowledge: The Study of International Regimes." Mershon International Studies Review 40 (October).

Rometsch, Dietrich and Wolfgang Wessels. Editors. 1996. The European Union and Member States: Towards Institutional Fusion? Manchester: Manchester University Press.

Ron, James. 1997. "Varying Methods of State Violence." International Organization 51 (Spring).

Sandholtz, Wayne. 1993a. "Choosing Union: Monetary Politics and Maastricht." International Organization 47 (Winter).

-------. 1993b. "Institutions and Collective Action: The New Telecommunications in Europe." World Politics 45 (January).

-------. 1996. "Membership Matters: Limits of the Functional Approach to European Institutions." Journal of Common Market Studies 34 (September).

Schimmelfennig, Frank. 1996. "Idealistic Liberalism and International Community: Liberal Values, Rhetorical Action and International Integration in Europe." Paper presented at the Annual Convention of the International Studies Association, April.

Schmidt, Susanne. 1997. "Sterile Debates and Dubious Generalizations: European Integration Theory Tested by Telecommunications and Electricity." Journal of Public Policy 16: 233-71.

Sikkink, Kathryn. 1991. Ideas and Institutions: Developmentalism in Brazil and Argentina. Ithaca: Cornell University Press.

-------. 1993a. "Human Rights, Principled Issue-Networks and Sovereignty in Latin America." International Organization 47 (Summer).

-------. 1993b. "The Power of Principled Ideas: Human Rights Policies in the US and Western Europe." In Ideas and Foreign Policy: Beliefs, Institutions and Political Change, edited by Judith Goldstein and Robert Keohane. Ithaca, NY: Cornell University Press.

Soysal, Yasemin. 1994. Limits of Citizenship: Migrants and Postnational Membership in Europe. Chicago: University of Chicago Press.

Stein, Janice. 1994. "Political Learning by Doing: Gorbachev as Uncommitted Thinker and Motivated Learner." International Organization 48 (Spring).

Strang, David and Patricia Mei Yin Chang. 1993. "The International Labor Organization and the Welfare State: Institutional Effects on National Welfare Spending, 1960-80." International Organization 47 (Spring).

Wapner, Paul. 1995. "Politics Beyond the State: Environmental Activism and World Civic Politics." World Politics 47 (April).

Weber, Steven. 1994. "Origins of the European Bank for Reconstruction and Development." International Organization 48 (Winter).

Weingast, Barry. 1995. "A Rational Choice Perspective on the Role of Ideas: Shared Belief Systems and State Sovereignty in International Cooperation." Politics & Society 23 (December).

Wendt, Alexander. 1987. "The Agent-Structure Problem in International Relations Theory." International Organization 41 (Summer).

-------. 1992. "Anarchy is What States Make of It: The Social Construction of Power Politics." International Organization 46 (Spring): 391-426.

-------. 1995. "Constructing International Politics." International Security 20 (Summer).

-------. 1996. Social Theory of International Politics. Book Manuscript (August).

Zuern, Michael. 1997. "Assessing State Preferences and Explaining Institutional Choice: The Case of Intra-German Trade." International Studies Quarterly 41 (June).


Footnotes

* Paper presented at the workshop on "Social Constructivism in European Studies," University of Aarhus, June 1998. The financial support of the Alexander von Humboldt-Stiftung, German Marshall Fund of the United States and Norwegian Research Council is gratefully acknowledged. An earlier version of this essay was presented at the European Consortium for Political Research (ECPR) 26th Joint Sessions, University of Warwick, March 1998.

[1]. For example, Thomas Christiansen, "Reconstructing European Space: From Territorial Politics to Multilevel Governance," in Joergensen 1997, 54-55.

[2]. On the different schools within constructivism, see Adler 1997, 335-37. Checkel 1998a provides a critical overview of the modernist branch. The phrase "seizing the middle ground" comes from Adler.

[3]. My analysis here and in the following paragraphs draws upon DiMaggio and Powell 1991, chapter 1; Longstreth 1992, chapter 1; Koelble 1995; Kato 1996; Katzenstein 1996, chapter 2; Finnemore 1996b; and Hall and Taylor 1996.

[4]. For historical institutionalists employing a thin conception of institutions, see Immergut 1992; and Pierson 1994. Thicker conceptualizations are found in Hattam 1993; and Goldstein 1993. Consistent with my near/long-term distinction, the analysis in Immergut and Pierson is contemporary, while that in Hattam and Goldstein spans decades.

[5]. Students of organization theory should recognize these arguments: They have roots in sociological work on organizations. See DiMaggio and Powell 1991, passim; Dobbin 1994; and March and Olsen 1998, 11-27.

[6]. On the European rights regime and its constitutive role, see Sikkink 1993b; and Drzemczewski 1983.

[7]. On these points, see, especially, Jon Elster, "The Market and the Forum," in Elster 1986; Morrow 1994, chapter 2; Green and Shapiro 1994, chapter 2; Weingast 1995; and Dennis Chong, "Rational Choice Theory's Mysterious Rivals," in Friedman 1996.

[8]. On logics of appropriateness more generally, see March and Olsen 1989.

[9]. See, however, Joergensen 1997.

[10]. Useful overviews include Anderson 1995, 443-51; Risse-Kappen 1996, 55-57; and Blank, Hooghe and Marks 1996, 343-47.

[11]. Moravcsik 1991, 25-27: Idem 1993, 496-517.

[12]. See, especially, Moravcsik 1994, 4-6, 14, 24-25, 43, 49, 59, 63; the quote comes from Moravcsik 1993, 481. For a full elaboration of Moravcsik’s theoretical logic, which makes clear his strong ontological commitment to methodological individualism, see Idem 1997.

[13]. For the game theoretic, rational-choice work, see Garrett 1992; and Garrett and Weingast 1993. Grieco 1995, 1996 develops a neorealist theory of international institutions as applied to the European Union. More recently, principal-agent analysis, a subset of rational choice with roots in the American politics literature, has been applied to institutions such as the Commission. While this research is rigorous and systematic in ways typically not seen in the EU literature, it portrays institutions in the same, thin manner criticized here. See Pollack 1997; and Franchino 1998.

[14]. Sandholtz 1993a, 4, and passim. Also see Sandholtz 1993b; and Idem 1996. In the latter, it is especially clear that Sandholtz views EU institutions in a behavioral, constraining sense only.

[15]. See Marks 1993; Leibfried and Pierson, 1995, chapter 1; Hooghe and Marks 1996; and Blank, Hooghe and Marks 1996. On the rational choice foundations of intergovernmentalism, including its liberal variant, see Moravcsik 1993; and Idem 1994. Risse-Kappen 1996, 56, discusses the rationalist underpinnings of neofunctionalism.

[16]. Blank, Hooghe and Marks, 1996, 347-56. Also see Leibfried and Pierson, 1995, 28, 30-31, 38, 40. I am indebted to Gary Marks and Liesbet Hooghe for discussion on these points. Hix’s otherwise excellent critique of the multi-level and new governance schools misses the rational-choice foundation of much of this literature. Hix 1998, passim.

[17]. Pierson 1996, 158-59, passim. In a separate monograph, he has elaborated this particular synthesis of rational choice and historical institutionalism: Pierson 1994, chapters 1-2. On historical institutionalism and the EU, also see Bulmer 1994.

[18]. Moravcsik 1995, for example.

[19]. The following summarizes an argument made at greater length in Checkel 1998a, 325-28, 338-47.

[20]. Wendt 1992; Katzenstein 1996, chapter 2; and Adler 1997 provide useful overviews of these core constructivist assumptions.

[21]. Wendt 1995, 73.

[22]. For an excellent discussion of this black box for mainstream theorists, written by an analyst sympathetic to their enterprise, see Powell 1994, 317-24.

[23]. Reliance on methodological individualism is widespread throughout both the IR and integration literatures. For IR critiques, see Wendt 1987, 340-44; and Rittberger, et al. 1996, 183-87. On the integration literature, see the previous section.

[24]. See Wendt 1996, chapters 1-2. Within the integration literature, there is widespread confusion on this last point, with many seeing an unbridgeable epistemological divide separating rationalist/mainstream work from constructivism. Among others, see Schmidt 1997, 236-37, 240; and Hix 1998, 47-48.

[25]. For example, Krasner 1983; and North 1990, chapter 5.

[26]. Finnemore 1996a, chapters 3, 4; and Klotz 1995b, chapter 6, for example.

[27]. For a somewhat different interpretation, see Aspinwall and Schneider 1997.

[28]. This problem stems from the reliance of many constructivists on the theoretical insights of sociological institutionalists, whose systematic bracketing of agency is almost pathological. See Checkel 1998b, 2-5, for details.

[29]. For example, Sikkink 1993b; and Risse and Sikkink 1999.

[30]. For constructivist accounts that emphasize dynamics of learning and socialization, see Klotz 1995a; and Finnemore 1996a. In Checkel 1997b, I demonstrate that once the role of agency is unpacked in constructivist accounts, parts of the analysis are better captured by standard rational-choice arguments.

[31]. On these research design problems within constructivism, see Jeffrey Legro and Paul Kowert, "Norms, Identity and Their Limits: A Theoretical Reprise," in Katzenstein 1996, chapter 12; and Checkel 1998a, 339-40. For identification of similar problems within the non-rationalist integration literature, see Schmidt 1997, 242, 250; Joergensen 1997, 7-8; and Hix 1998, 43-46.

[32]. Straw manning of the type noted here occurs all too often among critics of mainstream theory -- for example, Marlene Wind, "Rediscovering Institutions: A Reflectivist Critique of Rational Institutionalism," in Joergensen 1997, 24-26, 32-34. A quick look at the IR works published by Cornell University Press confirms my point on the middle-range theoretical aspirations of much scholarship. Cornell runs two IR series -- one on political economy and the other on international security. Combined, the two lists have published over 100 books, the vast majority of which advance middle-range theoretical arguments.

[33]. Good examples of middle-range theory, its use and development, are Haas 1992 (epistemic communities); and Risse-Kappen 1995a (new transnationalism). In the comparative literature, see Immergut (health-politics/historical institutionalism); and Pierson 1994 (welfare-state politics), among others. On the importance of middle-range theory in the EU context, see Hix 1998, 46. Hedstroem and Swedberg 1998, chapter 1, passim, persuasively argue the more general case for mid-range approaches in social theory.

[34]. Levy 1994 is an excellent introduction to the learning literature.

[35]. These hypotheses derive from a number of sources. See DiMaggio and Powell 1991, passim; Haas 1990, 1992; Hall 1993; Risse-Kappen 1996; and Checkel 1997a, chapters 1, 5.

[36]. On the last point, see Johnston 1998. Johnson 1993 provides an excellent and balanced discussion of the theoretically incomplete role accorded communication in rational-choice analyses.

[37]. Also see the excellent discussion in Zuern 1997, 300-302.

[38]. Within the integration literature, both Joergensen and Falkner have recently highlighted, from a constructivist perspective, the role of socialization and social learning dynamics at the European level. Both arguments, while important, are underspecified: They advance no (middle-range) propositions and scope variables that clarify the conditions under which such dynamics might be expected to occur. Joergensen, "PoCo: The Diplomatic Republic of Europe," in Joergensen 1997, 174-75; and Falkner 1998, 6-7, 12, 17, passim.

[39]. Of course, initially informal norms may, at some later point in their life history, become incorporated into formal legal codes and doctrines. Mattli and Slaughter 1998 provide a detailed review and critique of the literature on the Court of Justice.

[40]. On such entrepreneurs and the role, more generally, of individual agency in processes of norm development, see Nadelmann 1990; Finnemore 1996a; Florini 1996; and Finnemore and Sikkink 1998.

[41]. The epistemic and ideational branches of constructivism are especially helpful here. See Haas 1992; and Checkel 1997a, chapter 1.

[42]. Checkel 1997a, chapters 5-6.

[43]. Nadelmann 1990; Sikkink 1993a; Moravcsik 1995; Klotz 1995a, 1995b; Wapner 1995; Ron 1997; Keck and Sikkink 1998, chapter 1, passim; and Risse and Sikkink 1999. In the international law literature, Charney 1993, 543-550, hints at the growing importance of such pressure in creating new international legal norms.

[44]. Soysal 1994; Stein 1994; Risse-Kappen 1995b; Finnemore 1996a; Robert Herman, "Identity, Norms and National Security: The Soviet Foreign Policy Revolution and the End of the Cold War," in Katzenstein 1996, chapter 8; and Wendt 1996, chapters 7-8.

[45]. For details on the following, see Checkel 1998c, 5-9. Katzenstein 1985; Longstreth 1992; Risse-Kappen 1991, 1994; and Bleich 1998 provide excellent introductions to the domestic structures and historical institutionalist literatures.

[46]. For extensive documentation of these points, see Checkel 1998c, 15-26, where I also consider alternative explanations for the results presented here.

[47]. See Weber 1994; Risse-Kappen 1995a, chapter 1; and Adler and Barnett 1996, 97, passim.

[48]. Such issue and actor expansion has already occurred. See Hooghe and Marks 1996, and, more generally, the entire literature on multi-level governance. Indeed, the June 1997 Amsterdam Treaty, by incorporating the social policy articles of Maastricht directly into the Treaty on European Union, codified the access of various non-state actors to EU decisionmaking in that area. Obradovic 1997.

[49]. See Soysal 1994; Emma Tucker, "UK Balks at European Union Moves on Criminal, Civil Law," Financial Times, February 7, 1997; Roula Khalaf, "Call to Strengthen Rules on Human Rights and Democracy," Financial Times, February 11, 1997; Emma Tucker, AEuropean Union Asylum Seeker Policies Slammed, Financial Times, December 4, 1997; and Favell 1998, 3-7, 17.

[50]. Olsen 1995; and Idem 1996. Thanks to Johan Olsen for discussion on these points.

[51]. Kohler-Koch and Knodt 1997. Similarly, Conzelmann 1998, Part IV, passim, while making a convincing empirical case for national-level policy learning in the EU, fails to explicate testable scope conditions for when such dynamics are more or less probable.

[52]. Fligstein 1998, passim. Other theorists hinting at a constructivist-EU research agenda include Risse-Kappen 1996, 68-72; Schimmelfennig 1996; Knud Erik Joergensen, "Introduction: Approaching European Governance," in Joergensen 1997, chapter 1; and Risse 1998. The need for just such an agenda was also the focus of two well-attended sessions at the 11th International Conference of Europeanists (Baltimore, February 1998), as well as a workshop on "Social Constructivism in European Studies," University of Aarhus, June 1998.

[53]. Rometsch and Wessels 1996, preface, chapters 1-2, 14 -- quote at p.36. Caporaso, Cowles and Risse forthcoming, chapter 1, advance an historical institutionalist thesis to explain the variance in member-state convergence that so puzzles Wessels and Rometsch.

[54]. See Beyers and Dierickx 1997; and Idem 1998. Comments by Beyers at an ARENA seminar on 10 March 1998 confirm both the value of his study and its theoretical limitations.

[55]. Joerges and Neyer 1997, 617. Useful introductions to the comitology system are Pedler and Schaefer 1996; and Dogan 1997.

[56]. While several students of comitology have hinted at the importance of such factors for generating learning dynamics within EU committees, to my knowledge they have not been operationalized and empirically tested. See Pedler and Schaefer 1997, 47; and, especially, Joerges and Neyer 1997, 618.

[57]. Haas 1958, 16.

[58]. Neofunctionalism itself, despite its normative claims, offered no theory to explain when, how and why political agents might shape a collective identity. Risse-Kappen 1996, 56. On neofunctionalism’s in-built bias toward viewing integration as a "perpetual [forward] motion machine," see Caporaso 1998, 6-7.

[59]. On the rational-choice foundations of neofunctionalism, see Burley and Mattli 1993, 54-55. An excellent review and comparison of the neofunctionalist, integration and constructivist literatures, one which reinforces the points made in the preceding paragraphs, is Pollack 1998, passim.

[60]. For full details on these Ukrainian, Russian and German cases, see Checkel 1997b.

[61]. On human rights, general environmental and racial-equality norms, see Wapner 1995; Klotz 1995b; Ron 1997; and Risse and Sikkink 1999. For security, welfare and technocratic environmental norms, see Haas 1990; Strang and Chang 1993; and Thomas Risse-Kappen, "Collective Identity in a Democratic Community: The Case of Nato," in Katzenstein 1996, chapter 10.

[62]. Checkel, "The Europeanization of Citizenship?," in Caporaso, Cowles, Risse forthcoming. Of course, my phrasing begs the question of how one operationalizes "a meaningful number of the relevant international/regional actors." Legro 1997 is an excellent methodological discussion on gauging norm robustness independent of national-level effects.

[63]. Checkel 1998b, 19-24.

[64]. For differing views on constructivism, compare Adler 1997; and Checkel 1998a.

[65]. On the rationalist two-step, see Legro 1996. I thus disagree with those who implicitly endorse a temporal division of labor between rational choice and constructivism, with the study of the near-term ceded to the former. Finnemore and Sikkink 1998, 9, 15.

[66].To be fair, constructivists are not the only ones who, in critiquing rational choice, have fallen back on implicit as if assumptions of their own, failing to explicate alternative microtheories of process; similar problems have bedeviled work on prospect theory, learning theory and analogical reasoning. See Levy 1997, 101; and Peterson 1997, 266-67.

[67]. On the theoretical logic, see Barnett’s (1993, 1995) excellent discussion of role conflict and the related analysis by Meyer and Strang (1993) of cultural matches. For constructivist-ideational empirical research that documents the importance of domestic normative barriers, see Adler 1987; Sikkink 1991; Klotz 1995b, chapter 7; and Legro 1997. On institutionalization and the political influence of norms and other ideational variables, see Longstreth 1992; Katzenstein 1993; and Goldstein 1993. For one effort at operationalizing cultural matches in a manner amenable to empirical testing and replication, see Checkel 1998c, 4-5, 11.

[68]. On the cognitive logic, see Cohen and Levinthal 1990; and Bennett 1997, chapter 3.

[69]. Axel Van Den Berg, "Is Sociological Theory too Grand for Social Mechanisms?," in Hedstroem and Swedberg 1998, 206-12 (quote at 212). Risse 1997; and Eriksen and Weigaard 1998 provide general background on Habermasian theory and its possible application in IR. For an example of a constructivist study that invokes notions of communicative action, see Reus-Smit 1997, 564, 569-70, passim, where the theoretical-empirical disconnect highlighted here is particularly evident.

[70]. On the last point, see the trenchant rationalist critique in Moravcsik 1997, 538-41, which takes constructivists to task precisely for failing to elaborate a distinctive theory of socialization. On cognitive psychology, recent work by sociologists also points to its utility for bridging this macro (norm) - micro (agent) gap. DiMaggio 1997.

[71]. On the need to explore this interface in the context of debates over European integration, see Hix 1998, 55-56.

[72]. For the quote, see Caporaso 1998, 7.








[Date of publication in the ARENA Working Paper series: 15.08.1998]