The uneven but manifest emergence of a more coherent EU Common Foreign and Security Policy, which now includes significant elements of defence and military cooperation, makes it even more imperative to refine a theory and a set of analytical tools for studying the role of the EU as an international actor. As has often been noted, however, CFSP and its predecessor, European Political Cooperation, have not been well served by theory. ‘Like the debate over a common Community foreign policy itself, there is no agreement among academics on the most useful theoretical approach for comprehending this activity’ (Holland 1994: 129). Consequently, much of the literature on European foreign policy defines itself as ‘pre-theoretical’ (Hill 1993a), while limited importance is attached to the EU as an actor vis-à-vis its member states. In part, this reflects the dominance of the realist paradigm in international relations and traditional foreign policy analysis, which accords analytical primacy to states. In addition, it is a consequence of the multi-level and institutionally polyphonous character of EU decision-making in the second pillar (CFSP), and the continuing impact of national rather than communitarian approaches to foreign policy and conflict management – as the experience of the wars of Yugoslav succession amply demonstrates.

Nonetheless, as other contributors to this volume have argued, there are a number of non-realist approaches to international relations and foreign policy analysis that provide useful insights relevant to the study of the EU as an international actor. This chapter provides an overview and analysis of some of these approaches, and proposes an analytical framework with which to explore the complex interplay of factors affecting European foreign policy. This framework is based on a synthesis of elements of social constructivism, the new institutionalism and neo-classical realism. Foreign policy, it has been argued, ‘is the result of a complex interplay of stimuli from the external environment and domestic-level cognitive, institutional and political variables’ (Checkel 1993: 297). The analytical framework outlined below is based on three main concepts – interests, institutions and identities – and departs in two crucial respects from neo-realist and rationalist models of foreign policy analysis. First, it stresses that ‘institutions matter’ and, second, it views interests not as a given but as contingent on norms, beliefs and values (Schaber and Ulbert 1994). It also draws attention to the importance of foreign policy culture as ‘a set of attitudes, beliefs, and sentiments which give order to the foreign policy process and which provide the underlying assumptions and rules that govern behaviour in the international system’ (Kirste and Maull 1996: 284). Only by taking cognisance of the reflexivity inherent in the relationship between interests, institutions and identity, this chapter suggests, is it possible to understand the nature and dynamics of the European Union as an international actor.

Foreign policy analysis and the end of the Cold War

The pealing back of Cold War bipolarity has revealed the extent to which underlying processes of societal transformation have changed the structural dynamics of international society in late modern Europe. These transformations have been concentrated in Western Europe but their impact has also been felt across Central and Eastern Europe. Of particular significance has been the emergence of a zone of stable peace (Singer and Wildavsky 1996), with the transatlantic security community at its core. This has helped erode some of the main pillars of the Westphalian state system, and reduced the prevalence of balance-of-power and Realpolitik considerations in European order (Hyde-Price 2000: 70–101). Given that much traditional analysis of foreign policy has been grounded on realist assumptions about international anarchy and the state as ‘coherent units’ (Keohane and Nye 1977: 24), there is a pressing need for conceptual and theoretical innovation in this field.

New conceptual tools are particularly needed for analysing the external relations of the European Union, given its sui generis nature. Neo-realism offers little of value in this endeavour. For neo-realists like Kenneth Waltz (1979) and John Mearsheimer (1990), the behaviour of an international actor – in their case, states – is determined, first and foremost, by its relative power capabilities and its structural position in the anarchical states’ system.¹ This dictates an endless struggle for power and survival, in which states seek to cope with an intractable security dilemma through a balance of power. The decisions of states ‘are shaped by the very presence of other states as well as by interactions with them’. It is the ‘situation in which they act and interact’ which ‘constrains them from some actions, disposes them towards others, and affects the outcomes of their interactions’ (Waltz 1979: 65).

The problems with neo-realist explanations of state behaviour in late modern Europe are threefold. First, neo-realism’s paradigmatic assumptions are no longer appropriate to twenty-first-century Europe. The pursuit of parsimonious theory leads neo-realists to ignore the impact of historical, political and societal change on the structural dynamics of European order. Key neo-realist concepts such as ‘anarchy’ and ‘multipolarity’ are too ahistorical to offer much insight into contemporary European international politics. Neo-realist theory is unable to account for change or the enormous variation in behaviour between states with comparable power capabilities within ‘anarchic’ and ‘multipolar’ systems,² and overlooks the socially textured nature of European international society in the late modern era.

Second, ‘the refusal to consider what goes on within states is perhaps the most serious flaw of neorealism’ (Hoffmann 1995: 283). Because of the overarching explanatory weight neo-realists attach to structural factors such as ‘anarchy’, they overlook the importance of domestic factors on foreign policy behaviour. Finally, neo-realism underestimates the effects of economic interdependence and the impact of international organisations on state behaviour.³ In addition to its more general weaknesses, Thomas Pedersen notes (1998: 30), ‘structural realism has some particularly weak spots, the case of European integration being one’. Neorealism’s pessimistic assumptions about the possibilities of cooperation means that it is hard-pressed to account for the extent to which multilateral integration has developed in Europe. Neo-realism’s crucial weakness is thus ‘its inability to provide an explanation of the high degree of institutionalisation and notably its durability in the European region’ (ibid.).

In contrast to neo-realism’s privileging of the structural determinants of state behaviour (which is shared by some versions of social constructivism⁴), the analytical framework outlined below considers both actor-level and structural determinants of foreign policy behaviour. The analysis itself draws on a number of sources. First: some of the more innovative approaches within the field of foreign policy analysis, particularly those associated with so-called ‘second generation’ theorists (Neak, Hey and Haney 1995). Work in this field has focused, inter alia, on the institutional processes involved in foreign policy decision-making, and on the cultural and societal context within which policy options are considered.

Second: the insights of social constructivists such as Emmanuel Adler (1997b) and Jeff Checkel (1997), 1998a), who have sought to carve out an analytically robust ‘middle ground’ between rationalists and reflectivists. Constructivism’s most important insight is that ‘the manner in which the material world shapes and is shaped by human action and interaction depends on dynamic normative and epistemic interpretations of the material world’ (Adler 1997b: 322).

Third: what has been termed ‘neo-classical realism’ (Schweller 1996; Snyder 1996; Rose 1998). Scholars working in the classical realist tradition continue to place emphasis on the importance of material power capabilities of a state in relation to the wider international system as a dominant factor shaping broad patterns of foreign policy over time. However, they also recognise that there is no direct transition belt linking material capabilities to foreign policy behaviour, and acknowledge the importance of both elite concepts of national role and identity, and the institutional policy-making process. In this way, neo-classical realists ‘open spaces for conversations, perhaps even mutually profitable sharing of insights’, involving ‘constructivists of various stripes (including the English School)’ (Donnelly 1998: 403; see also Dunne 1995; Williams 1997: 300).

Interests

Defining the ‘national interest’

Understanding the behaviour of actors in international society is impossible without some notion of their interests; that is, their preferences and concerns. The concept of ‘interests’ has long been central to political analysis, yet substantial methodological and epistemological difficulties remain in defining collective ‘interests’. This is particularly the case with the concept of ‘national interests’.⁵ However, if ‘society as a whole is granted a reality other than that of the sum of its contending parts, then it becomes difficult to deny that the group defined by the society has its “interests” just as the smaller, more particular groups have theirs’ (Clinton 1991: 49). Similarly, if states have their national interests, the EU as an international actor must also have its own interests, however diffuse and amorphous they may be at times.

The central problem in defining the ‘European interests’ of the EU is how to distinguish between the interests of the EU as a whole and those of the specific interests of its individual member states and other influential non-state actors within it (such as the agricultural lobby). Once one discards neo-realist assumptions about states and international actors as rational utility maximisers in favour of a conception of international actors as complex institutional ensembles, then ‘European’ interests can only be seen as the outcome of a discrete political process. The main actors involved in this political process are the member states, the Commission, powerful lobby groups and public opinion articulated partly through the medium of the European Parliament. The analysis of the EU’s interests thus involves consideration of both the political discourse surrounding its external role conceptions and of the institutional policy-making process within the EU.

The limits of materialism

How a political community defines its interests depends on objective, material factors (such as geography, size and wealth), but also on a range of subjective, normative considerations. These include the identity of a community, its political culture, dominant moral and ethical values, sense of justice and conception of the common good, and its belief in what makes it distinctive as a political community.

Stanley Hoffmann has argued that the foreign policy of a state can only be understood by considering its ‘national situation’. This refers to both the domestic characteristics of a state and its people, and to its position in international society. ‘It is a composite of objective data (social structure and political system, geography, formal commitments to other nations) and subjective factors (values, prejudices, opinions, reflexes, traditions toward and assessments of others, and others’ attitudes and approaches).’ This national situation is not a ‘given’ that dictates policy, but it does ‘set up complicated limits that affect freedom of choice’ (1995: 75–6).

Although Hoffmann’s concept of the ‘national situation’ was developed with a view to analysing the foreign policy behaviour of individual nation-states, it can be utilised to understand the EU’s role as an international actor. It certainly provides a much richer approach to the study of international politics than that of neo-realism, which focuses almost exclusively on material power capabilities. John Mearsheimer, for example, emphasises that state behaviour ‘is largely shaped by the material structure of the international system’, and it is the ‘distribution of material capabilities among states’ that provides ‘the key factor for understanding world politics’ (1995: 91; original emphasis). This assumption of the ‘objectivity’ of interests is shared by structural Marxists, who emphasise the centrality of economically determined class interests, and by neo-liberal institutionalists, who ‘continue to treat actor identities and interests themselves as preexisting and fixed’ (Kowert and Legro 1996: 458).

However, the problem with rationalist approaches is that ‘[m]aterial capabilities as such explain nothing; their effects presuppose structures of shared knowledge, which vary and are not reducible to capabilities’ (Wendt 1995: 73; original emphasis). ‘The inescapable fact seems to be that this type of analysis stressing strategic and geopolitical factors is badly flawed’ (Hilsman 1990: 30). The importance of normative and epistemic interpretations of the material world was evident to an earlier generation of classical realists. Hans Morgenthau, for example (1948: 11), quoted Weber to the effect that ‘Interests (material and ideal), not ideas, dominate directly the actions of men.’ However, he went on to observe that ‘the “images of the world” created by these ideas have very often served as switches determining the tracks on which the dynamism of interests kept actions moving’. Morgenthau thus concluded that ‘the kind of interest determining political action in a particular period of history depends on the political and cultural context within which foreign policy is formulated’. Consequently, ‘[t]he goals that might be pursued by nations in their foreign policy can run the whole gamut of objectives any nation has ever pursued or might possibly pursue’ (1948: 11).

A central concern of contemporary foreign policy analysis is thus to explore the ‘national situation’ within which international actors are located. Material factors are important in setting the broad parameters within which the EU operates. However, the study of these material factors is not sufficient to indicate the course and objectives of the EU’s Common Foreign and Security Policy. Understanding this involves considering two other factors: institutions and identities.

Institutions

Institutional analysis provides the second leg of the foreign policy analysis model presented here. As we have seen, while the systemic distribution of material capabilities establishes the broad parameters of foreign policy behaviour over time, specific foreign policy decisions result from a complex process of institutional policy-making. Institutional structures are therefore a decisive factor in determining what has been called ‘the capacity for collective action’; in other words, the ability of a state to extract and utilise resources for foreign policy goals (Hoffmann 1995: 268).

The assumption underpinning this leg of the conceptual model is ‘that institutions sometimes matter, and that it is a worthy task of social science to discover how, and under what conditions, this is the case’ (Keohane and Martin 1995: 40). Institutions ‘do not merely reflect the preferences and power of the units constituting them; the institutions themselves shape those preferences and that power’ (Keohane 1988: 382). They can do this by altering ‘the calculations of interest by assigning property rights, providing information, and altering patterns of transaction costs’ (Keohane 1993a: 29). Over time, commonly agreed and jointly observed principles and norms become internationalised by the actors involved, thereby reshaping the perception of interests (Rittberger 1993: 19).

The important point to note about institutions is that they do not simply facilitate the bargaining process between political actors by reducing transaction costs and reducing uncertainty, as liberal institutionalists suggest. More importantly, institutions play a key role in the process of interest and identity formation. This has been one of the most important insights of both social constructivism and the ‘new institutionalism’. Institutions ‘offer a normative context that constitutes actors and provides a set of norms in which the reputation of actors acquires meaning and value’ (Katzenstein 1997: 12–13). International institutions and multilateral structures thus facilitate the emergence of a sense of Gemeinschaft (community) based on shared interests, trust and a common identity.

The constitutive and transformationalist impact of institutions on the process of interest and identity formation is clearly evident from the operation of EPC/CFSP over the last two decades. The institutionalisation of regular consultation, negotiation and cooperation between diplomats involved in shaping the EU’s external relations has transformed the working practices of foreign ministers and ministries, and generated new styles of operating and communicating. As Hill and Wallace (1996): 6) note,

Organisational process and bureaucratic politics

The seminal text for understanding the role of institutional actors in the foreign policy-making process is Graham Allison’s classic work, Essence of Decision – ‘one of the few genuine classics of modern International Relations’ (Brown 1997: 75). Allison outlined three models of foreign policy decision-making: the rational actor model (RAM), the organisational process model and the bureaucratic politics model. The rational actor model assumes that foreign policy decisions are rational responses to a particular situation, formulated by a single unitary state actor operating on the basis of perceived national interests. The problem with the RAM approach is that states are not unitary actors and that interests are not endogenously given.

Allison’s other two models are of more use in understanding the institutional dimension of EU foreign and security policy. His second model (organisational process) takes as its point of departure the institutional pluralism of modern states and focuses on intra-organisational factors. It posits that policy outcomes result from the interaction of multiple organisations, each of which has distinctive ways of making policy decisions – organisational routines and standard operating procedures. These organisations are highly resistant to attempts at centralised control and coordination. While Allison’s model was based on the US experience, it is also relevant to the EU, given the institutional polyphony that characterises its complex, multi-level decision-making processes.

Allison’s third model (bureaucratic politics) focuses on inter-organisational factors. While the two previous models suggest that foreign policy decisions will be taken on foreign policy grounds, the bureaucratic politics model suggests that this is not necessarily so. Political, economic and other factors external to the specific foreign policy issue being considered may affect decision-making. In particular, bureaucracies have their own specific interests and concerns to defend (budgets, resources, influence), and may therefore propose policy options that enhance their position within the overall policy-making process. Allison’s third model thus stresses that bureaucracies make decisions on the basis of their own specific organisational interests – in a nut-shell, ‘where you sit determines where you stand’.

As Allison argued, the three different models each have a certain explanatory power and need to be integrated rather than segregated. Allison’s analytical framework is certainly of relevance when it comes to investigating the role and impact of the key institutional actors involved in policy-making in pillar two. These include the European Council, which determines the CFSP’s principles and general guidelines, and its common strategies; the foreign affairs ministers in the General Affairs Council, who can recommend common strategies to the European Council and who are responsible for implementing them, notably through the adoption of common positions and joint actions; COREPER (the Permanent Representatives Committee), which prepares the proceedings of the General Affairs Council; the Political Committee, which monitors the international situation, provides opinions for the Council and oversees the implementation of policy; the Presidency of the European Council, which represents the Union in CFSP matters and is responsible for the implementation of CFSP decisions; the High Representative of the CFSP, who assists the Presidency and the Council by contributing to the formulation, drawing up and implementation of political decisions; and the European Commission, which, according to the Maastricht Treaty on European Union, is to be ‘fully associated’ with the work carried out in the CFSP field. In addition, a ‘policy planning and early warning unit’ has been established in the General Secretariat of the Council under the responsibility of the High Representative.

With the launch of the Union’s European Security and Defence Policy at the Cologne Council of June 1999 and the adoption of the ambitious ‘Headline Goal’ by the Helsinki Council of December 1999, the second pillar has acquired a number of new institutional decision-making structures. In addition to ad hoc meetings of defence ministers within the framework of the General Affairs Council, a Political and Security Committee, a Military Committee and a Military Staff (including a Situation Centre) have been established. With the exception of its Article V mutual security guarantee, all of the WEU’s key functions and responsibilities have been folded into the EU. The rapid development of the ESDP means that the institutional structures and decision-making procedures in the second pillar are still very much in flux, and substantial study and analysis is required before their effectiveness can be assessed. Indeed, the development of the institutional complexity of the second pillar underlines the necessity for detailed institutional analysis in order to understand the behaviour of the EU as a foreign and security policy actor. For such an analytical undertaking, Allison’s work remains an essential reference point.

However, the biggest lacuna in Allison’s work, and in the literature on ‘institutional pluralism’ more generally, is an insufficient account of the cognitive and ideational dimension of decision-making. This has been an area of considerable theoretical innovation over recent years, although many questions remain to be addressed. The ideational or socio-cognitive dimension thus constitutes the third leg of the analytical framework developed in this chapter.

Identities

The ‘cognitive revolution’

While an analysis of institutional policy-making processes can provide valuable insights into the making and execution of European foreign policy, it can only reveal part of the picture. One of the important developments in the field of foreign policy analysis since Allison’s path-breaking study has been the increased awareness accorded to cognitive and psychological influences on policy-making. This has been part and parcel of what has come to be known as the ‘cognitive revolution’. The central insight of the cognitive revolution is that ‘structural factors – such as institutions, bureaucracies, international regimes, the state of the economy, geopolitical emplacement, etc. – are cognitively mediated by the actors in question rather than affecting policy actions directly’ (Carlsnaes 1994: 284; original emphasis). ‘Thinking about foreign affairs’, one British diplomat has written, ‘ – like any other kind of thinking – requires a conceptual map which, as maps do, simplifies the landscape and focuses on the main features’ (Cooper 1996: 8).

The cognitive revolution has further eroded rationalist assumptions about the pre-given and objectively determined character of interests. The central argument of the cognitive revolution has been that rationality is bounded. Actors ‘satisfice’ rather than optimise. They do not have access to all relevant information, nor do they seek it. They process information differently in times of stress than under routine conditions. They act on the basis of hidden agendas, such as their specific institutional or bureaucratic concerns. Above all, historical precedents, cultural values, normative beliefs and other socio-psychological factors influence the perception of interests. Consequently, ‘images of other states are difficult to alter. Perceptions are not responsive to new information about the other side; small changes are not likely to be detected. Once a statesman thinks he knows whether the other needs to be deterred and what kind of strategy is appropriate, only the most dramatic events will shake him’ (Jervis 1983: 25–6).

In an important study of the impact of ideas on foreign policy, Judith Goldstein and Robert Keohane (1993) have proposed a threefold categorisation of ideas or beliefs. First, ‘world views’: these are embedded in the symbolism of a culture and entwined with identity conceptions. Second, ‘principled beliefs’: these are normative ideas for distinguishing between right and wrong. Third, ‘causal beliefs’: these are beliefs about cause–effect relationships which provide guides as to how to achieve a given policy objective.

They go on to suggest that ideas can influence foreign policy behaviour in three ways. First, they can serve as ‘road maps’, guiding actors’ preferences and indicating ways of achieving them. Second, they can act as a ‘focal point’ for actors in the absence of compromise or cooperation. Third, they can become ‘institutionalised’; that is, embedded in the operation of institutions and social practices. Their argument is that ‘ideas influence policy when the principled or casual beliefs they embody provide road maps that increase actors’ clarity about goals or end–means relationships, when they affect outcomes of strategic situations in which there is no unique equilibrium, and when they become embedded in political institutions’. In contrast to many social constructivists, however, they do not argue that it is ideas rather than interests that ‘move the world’. Rather, they follow a Weberian approach in arguing that ‘ideas as well as interests have causal weight in explanations of human action’ (Goldstein and Keohane 1993: 3–4; original emphasis).

Goldstein and Keohane’s Weberian assumptions that both ideas and material interests shape foreign policy behaviour are of considerable utility in understanding European foreign policy, even if one does not accept their rationalist approach to empirical research. Understanding and explaining the behaviour of the EU as an international actor necessitates a research strategy that examines different national discourses on Europe (see for example Carlsnaes and Smith 1994). The purpose of this should be to uncover the ideas and cultural assumptions that influence policy debates (particularly in the larger and most influential member states) and to explore the impact on EU foreign policy of an underlying set of cultural norms and values, understood as ‘social knowledge structures that define collective expectations of appropriate action’ (K. Aggestam 1999: 36). This approach assumes that international actors behave according to the ‘logic of appropriateness’ rather than the ‘logic of expected consequences’ (March and Olsen 1998b: 8–11).⁶ The EU’s role as a foreign and security policy actor can thus only be fully understood by considering the cultural determinants which shape it, and which provide ‘a framework for organizing the world, for locating the self and others in it, for making sense of the actions and interpreting the motives of others, for grounding an analysis of interests, for linking collective identities to political action, and for motivating people and groups toward some actions and away from others’ (Ross 1997: 42).

Identity and foreign policy

One of the most important areas of theoretical innovation in foreign policy analysis over recent years has been the growing awareness that identities matter in international relations. Identities shape the definition of national and European interests and thereby constitute an important influence on foreign policy behaviour. National identity ‘serves not only as the primary link between the individual and society, but between a society and the world’ (Pritzel 1998: 19). In the case of the EU, its foreign and security policy rests on a shared European ‘identity’ which has evolved gradually and, at times, fitfully, over the last few decades. This emerging sense of European identity is important in defining Europe’s role in international society, its friends and enemies, and its common interests and preferences. This sense of an emerging European identity has been, and is, continually contested and redefined, in response to the shifting domestic elite and mass attitudes to the European integration process. Nonetheless, however indistinct and contested Europe’s identity may at times seem, it does have some core elements which have evolved since the early 1970s, if not before. These core elements are associated with the values of liberal-democracy, social market economies and the peaceful resolution of disputes, and have been further defined and reinforced by the experience of 1989 and the post-communist democracies’ ‘return to Europe’.

Europe’s evolving political identity helps shape the broad directions of its Common Foreign and Security Policy by providing cognitive lenses for policy-makers, based on political judgements about the ‘national interest’. Analyses of official discourses on European identity can only explain the grand schemes of European diplomacy. They cannot provide explanations of discrete foreign policy behaviour or the nuances of negotiating positions on specific policy issues. They cannot, for example, shed significant light on the EU’s negotiations on bananas, or on specific policies with the framework of the Northern Dimension Initiative. These are determined by a political process involving relevant officials within the Commission in association with national ministries and governments, other political actors and private sector interests. Conceptions of the European interest only affect this political process indirectly, by establishing the broad ideational parameters within which competing policy options are considered. European identity thus provides the cognitive framework within which the EU’s foreign and security policy is formulated.

Role theory

As a result of the cognitive revolution, it was increasingly recognised in foreign policy analysis circles that perceptions were vital in the policy-making process. Perceptions provided a means of intercepting, classifying and interpreting information in terms of pre-established beliefs. These beliefs, it was suggested, were organised into structured systems that set cognitive limits to rational decision-making. However, ‘most of the studies which focus on decision-makers’ perceptions include only perceptions of the external environment, especially enemy characteristics and actions, and very few investigate decision-makers’ perceptions of their own nations’ (Wish quoted in Jönsson 1984: 3). One notable exception to this was the research tradition on national role conceptions.

Role theory was first introduced to foreign policy analysis by Kalevi Holsti in a seminal article originally published in 1970. He drew on sociological interpretations of role in order to suggest how perceptions may structure and guide foreign policy-making. He argued that a state’s foreign policy is influenced by its ‘national role conception’. This is a product of a nation’s socialisation process and is influenced by its history, culture and societal characteristics. ‘A national role conception’, he argued, ‘includes the policymakers’ own definitions of the general kinds of decisions, commitments, rules, and actions suitable to their state, and of the functions, if any, their state should perform on a continuing basis in the international system or in subordinate regional systems’ (Holsti 1987b: 12; original emphasis). His methodology involved analysing elite perceptions of national role, and initiated a research tradition focused on role conceptions (see for example Walker 1979; Wish 1980; Jönsson and Westerlund 1982). This approach has been relatively neglected for many years, although more recently it has received growing academic interest (see for example. Aggestam 1999; Kirste and Maull 1996; Tewes 1998).

‘Why the new interest in role conceptions?’, Hudson and Vore ask (1995): 226). The answer, they suggest, is that national role conceptions provide ‘one of the few conceptual tools we have for the study of how society and culture serve as a context for a nation’s foreign policy’. Role theory thus ‘allows one to bridge the conceptual gap between the general beliefs held in a society and the beliefs of foreign policy decision makers’. Defining a foreign policy role and having it accepted by others is one of the basic objectives of a state. As Le Prestre (1997): 5–6) notes, a role concept ‘reflects a claim on the international system, a recognition by international actors, and a conception of national identity’. Consequently, foreign policy change ‘must rest on a redefinition of a role and on the role’s congruence with politics’. Moreover, ‘contrary to what structural realists would assert, capacities alone do not define a role’. Thus he argues that role theory ‘can help explain the general direction of foreign policy choices’.

The articulation of national role betrays preferences, operationalises an image of the world, triggers expectations, and influences the definition of the situation and the available options. It imposes obligations and affects the definition of risks. Focusing on this concept, therefore, allows one to go beyond the traditional explanation of foreign policy, which is based on security or on the national interest defined as the prudent search for power. Roles help define national interests and divorce them from power (Le Prestre 1997: 5–6).

THE EU’S EVOLVING ROLE-SET

The importance and analytical utility of role theory for an understanding of European foreign policy are discussed at greater length elsewhere in this volume (see chapter 6 by Lisbeth Aggestam). In terms of the discussion here, therefore, only three points will be made.

First, the EU’s evolving role-set in the 1990s has been defined primarily by the end of the Cold War and the desire of the new democracies of East Central Europe to ‘return to Europe’. This role-set involves both the role perceptions ascribed to the EU by outsiders and role conceptions expressed by the role-holders themselves (see chapter 6). In the early 1990s, these role perceptions and conceptions coalesced around a view of the EU as the institutional embodiment of the ‘European ideal’. The EU was seen as performing the roles of an anchor of stability in the new Europe; a magnet for the new democracies of Central and Eastern Europe; and a focus for European cooperation and integration. Above all, the EU was seen as a bastion of liberal democratic values and practices, involving a respect for human rights and the rule of law, and the peaceful resolution of disputes.

Second, the EU’s foreign and security policy role-set has been steadily evolving over the last decade or so, but is likely to undergo a more profound redefinition in the coming years. The catalyst for this will be the development of the European Security and Defence Policy (Missiroli 2000: 2–3). The EU has traditionally regarded itself, and has been seen by others, as a ‘civilian power’ (see Duchêne 1973; Bull 1982). This self-image as a ‘civilian power’ will, however, be hard to sustain as the EU moves to acquire what the Cologne Council defined as ‘the capability for autonomous action, backed up by credible military forces, the means to decide to use them, and a readiness to do so, in order to respond to international crises without prejudice to actions by NATO’ (for details see Van Ham 2000; Heisbourg 2000; Howorth 2001). The policy implications of this shift in role perceptions of the EU by Russia or countries in the Middle East and the Maghreb may be quite profound, especially as regards attitudes towards EU enlargement or the role of the Union in the Arab–Israeli peace process.

Third, the EU’s political identity and its role conceptions are highly contested, given the different national foreign and security policy traditions of its member states, and the EU’s relatively short history as a foreign policy actor. As suggested above, the experience of working closely together in common institutions within the framework of the second pillar may in time lead to a process of engrenage, in which habits of working together gradually generate shared perceptions of common interests. However, as Lisbeth Aggestam concludes, foreign policy objectives are still conceptualised by the three largest EU member states ‘first and foremost in a national mind-set’ (see p. 97 of this volume). This illustrates all too clearly the problems facing the EU as an international actor. The success of its CFSP – especially in the area of preventive diplomacy and crisis management in regions such as the Eastern Europe, the Middle East and the Balkans – will depend above all on the willingness of member states to reach a broadly acceptable consensus on the aspirations, goals and means of EU foreign policy. This will be a long-term task, and one that is likely to be complicated by the enlargement process (Sjursen 1999).

Conclusion

The aim of this chapter has been to outline a conceptual framework for understanding the role of the EU as an international actor. This analytical model rests on three ‘legs’ – interests, institutions and identities. A constant theme throughout has been the limitations of the dominant neo-realist approach to foreign policy analysis, and the need to consider both the material and ideational factors defining Europe’s Common Foreign and Security Policy. Attention has been drawn both to the role of institutional politics in shaping policy outcomes, and to the importance of culture and identity to foreign policy behaviour. In terms of this last point, the utility of role theory as a conceptual tool for operationalising the study of identity and socio-cognitive factors on foreign policy has been underlined.

As was noted in the introduction to this chapter, the study of EPC/CFSP has been under-theorised. It is hoped that the ideas presented above will contribute to the growing debate within international relations and European studies on the need to think more conceptually and theoretically about the EU as an international actor. It does so in the spirit of theoretical pluralism. Although much of the analytical framework presented above draws primarily on social constructivism and the new institutionalism, it also incorporates some insights from classical realism, which has been enjoying something of a revival in recent years. This is because exclusivist and partisan claims to knowledge, either from neo-realists or post-positivists, should be viewed with a high degree of scepticism. As Michael Clarke has argued, if we wish to ‘understand the working of political power in our contemporary world’ we need to draw on a wide range of disciplinary and theoretical perspectives. ‘The world into which we are moving offers prima facie evidence that nothing less than such an ambitious attempt at eclecticism will do’ (Clarke 1993: xvi). Understanding and explaining the foreign and security policy of an actor as complex and multi-faceted as the EU thus involves utilising an eclectic analytical framework that draws from a plurality of different traditions and approaches. Only in this way can we open up spaces for theoretical debate and discussion on the EU as an international actor.