Mare Europaeum: Baltic Sea Region Security and Cooperation from post-Wall to post-Enlargement Europe

Fabrizio Tassinari Mare Europaeum Baltic Sea Region Security and Cooperation from post-Wall to post-Enlargement Europe Department of Political Science University of Copenhagen i ii Preface and Acknowledgements The title of this dissertation reflects two considerations that, by the time I ended this virtual and real journey across the Baltic Sea, had become of crucial importance. The first consideration results from a rather peculiar, and to me revealing, sequence of occurrences. A few years ago, I learned from historian Uffe Østergaard that one European geographical centre – measured from the North Cap to Sicily and not including Russia on the east-west axis – is located in the Danish mainland. This notion – coming at the very beginning of this research – sparked my curiosity and I later came to find another geographical centre of Europe – this time measured from the Atlantic to the Urals, and therefore including part of Russia – in Lithuania, some 25 km north of the capital Vilnius. I have not discovered other ‘centres of Europe’ in the Baltic Sea area since. However, I recently came across a motto composed by the Municipality of Stockholm: “The Baltic Sea Area, the Centre of Europe” (UBC Bulletin, 1997). That motto brought my memory back to those geographical coordinates, and most importantly to the all-embracing presence of the wider European dimension in the shaping of my argument, all of which convinced me that Europe had to be emphasised in the title of this study. The second consideration refers to the very wording of the title. Admittedly, the evocative usage of Latin terminologies with reference to issues related to seas is rather common (e.g. ‘Mare Nostrum’, ‘Mare Clausum’ or, more obviously, ‘Mare Balticum’). But the Latin title here is also justified by more conceptual reasons. It is meant to acknowledge the influence of liberalist thinking in part of my argument, particularly in the formulation of the final discussion regarding the post-enlargement scenario for the Baltic region. This discussion, we shall see, pushes the liberalist argument somewhat. And in doing so, it makes reference to seventeenth-century Dutch jurist Hugo Grotius, who, quite aptly, entitled one of his major works “Mare Liberum”, “The Freedom of the Sea” (Grotius 1916 [1608]). Hence, Mare Europaeum: the European Sea. The completion of this dissertation would not have been possible without the help and support of a number of individuals and iii institutions that I wish to acknowledge here. First of all, I would like to thank my supervisor at the University of Copenhagen, Uffe Jakobsen, for believing in this project, and in me in the first place. His encouragement and optimism have been a most important source of strength during the course of this study, especially during the final demanding stages. The EU-project BaltSeaNet: “Baltic Sea Area Studies: Northern Dimension of Europe” has supported this work financially, and I wish to acknowledge the help of all its members. In particular, project administrators Kazimierz Musial, Leena-Kaarina Williams and Sigita Baronaite have provided a most valuable source of professionalism and enthusiasm for a very large and diverse group of individuals. Moreover, I am grateful to Project Director Bernd Henningsen. Whether or not a Baltic Sea identity really exists, he truly managed to make me – a ‘Mediterranean’ – feel a part of this great community. May the EU Commission be satisfied with our work. The financial and academic support I received from the Department of Political Science at the University of Copenhagen has been equally precious. I would like to thank the two directors of the PhD programme I had the opportunity to work with: Hanne Foss Hansen and Jens Bartelson. I am also grateful to my fellow PhD Students, and particularly to Henrik Breitenbauch, Olaf Corry and Peter Toft for reading and commenting upon preliminary drafts of this dissertation. Thanks also to Ole Nørgaard from the Department of Political Science at the University of Aarhus, who attended my very first internal seminar at our Department and encouraged me to deepen research in the environmental field. Moreover, I am indebted to Morten Kelstrup and Anders Wivel, who acted as opponents in the final seminar at the Department and commented on the entire manuscript. The insights I received from them considerably strengthened the ‘big picture’ behind this thesis. I also wish to thank the members of the EUR project at the Copenhagen Peace Research Institute (COPRI) – the ‘cradle’ of the Copenhagen School – and particularly Vibeke Schou Pedersen, for managing to secure my involvement in the project. I am also grateful to Project Director Barry Buzan for allowing me to attend the monthly seminars. Even though I accessed the project in its ‘dying days’, as he put it, the discussions at the EUR seminars were more than ‘alive’. iv They represented a very loud ‘wake-up call’ to bring the writing of this dissertation up to speed. Speaking of COPRI and of the EUR group, I also wish to express my gratitude to Ole Wæver of the Department of Political Science, University of Copenhagen, to Pertti Joenniemi of the Danish Institute of International Studies (DIIS) in Copenhagen, and especially to Chris Browning of the University of Birmingham, UK. They all commented on earlier drafts of this thesis and helped me to ‘stretch’ the theoretical continuum upon which this work is based. My knowledge of Russian foreign policy has been challenged by Alexander A. Sergounin of the Nizhny Novgorod Linguistic University, Russia; by Paul Holtom of the Centre for Border Studies, University of Glamorgan, UK; and by Lars Poulsen-Hansen of DIIS. They commented on some of my papers and provided depth and perspective to my arguments. My special interest for the Russian side of this story is also a result of their valuable advice, for which I am grateful. My insights regarding ‘real’ security and foreign policies have been enriched by a number of diplomats and officials from the Baltic Sea states and in Brussels, people who took time away from their busy schedules to share their experiences with me. I thank them all, and particularly Lars Grønbjerg of the Royal Danish Ministry of Foreign Affairs. Lars not only managed to find a number of early documents from the Council of the Baltic Sea States for me – he was also kind enough to share his impressions on the present and future of EU-Russia relations on several occasions. Above all, I would like to thank Marie, my fiancée, who has endured both the virtual and real parts of this journey across the Baltic. I could not have written about security, peace, and stability of this or any other region, without the love we managed to create within our little family. For this, and for many many other reasons, this dissertation is dedicated to her. Of course, none of these people are responsible for that which follows. The responsibility for this dissertation, and for any possible mistake in it, is entirely mine. Fabrizio Tassinari, Copenhagen, April 2004 v vi List of Abbreviations BSR: Baltic Sea Region BSSSC: Baltic Sea States Subregional Cooperation CBSS: Council of the Baltic Sea States CFSP: Common Foreign and Security Policy CIS: Commonwealth of Independent States CSR: Common Strategy on Russia CSS: Critical Security Studies ED: Eastern Dimension EEC: European Economic Community e-PINE: Enhanced Partnership in Northern Europe EU: European Union FEZ/SEZ: Free Economic Zone and Special Economic Zone (in Kaliningrad) IPE: International Political Economy IR: International Relations NATO: North Atlantic Treaty Organisation NDEP: Northern Dimension Environmental Partnership NDI: Northern Dimension Initiative NEI: Northern European Initiative NGO: Non-Governmental Organisation OECD: Organisation for Economic Cooperation and Development OSCE: Organisation for Security and Cooperation in Europe PCA: Partnership and Cooperation Agreement (with Russia) PfP: Partnership for Peace PHARE: Poland and Hungary: Action for the Restructuring of the Economy RBA: Region Building Approach RSCT: Regional Security Complex Theory TACIS: Technical Assistance for the Commonwealth of Independent States UBC: Union of the Baltic Cities WNIS: Western Newly Independent States (Ukraine, Moldova, Belarus) WTO: World Trade Organisation vii Contents Preface and Acknowledgements .…………………………………….. iii Abbreviations …………………………………………………………vii Introduction…………………………………………………………. 1 PART I: The Framework 1: Regionalisms in International Relations Theory …………….… 15 1 Framing the Debate…………………………………………………………... 15 2 ‘Old’ Regionalisms……………………………………………………..……. 20 3 ‘New’ Regionalisms ………………………….……………………………... 25 4 Back to the Matrix …..………………………………………………………. 32 2: A Critical Region Building Approach ………………………….. 36 1 Security Concepts ……………….……….………………………………….. 37 2 Regional Security …………….…………………………….……………….. 47 3 Suitability of the Critical Regional Security Complex Theory …………...… 52 4 The Region Building Approach …..………………………………………….55 5 Three ‘Critical’ Arguments …………………………………………………. 58 6 Summing up ……………………………………………………….………... 75 3: A Methodological Trait D’Union ……………………………….. 81 1 Constructivist Epistemology...………………………………………………. 82 2 Methods and Sources ……….………………………………………………. 91 3 Tempo and Modus of the Analysis ……….………………………………… 98 Part II: The Analysis 4: The Rise …………………………………………………………. 109 1 Prologue: Space and Time ………………………………………………..…109 2 The Remaking of Norden …………………………………………………...115 3 ‘Baltism’ ……………………………………………………………….…... 119 4 Institutionalisation………………………………………………………….. 127 5 The Council of the Baltic Sea States ………………………………………. 131 6 Analysis: ‘Regionality’ Undone …………………………………………… 133 viii 5: The Framing: States ……………………………………………. 143 1 ‘Old’ Europe 1.1 The Nordics …………………………………………………….….…… 145 1.2 Germany …………………………………………………………………155 2 ‘New’ Europe 2.1 The Baltics ………………………………………………………………159 2.2 Poland ………………………………………………………………….. 166 3 Russia …………………………………………………………………….…169 4 Analysis: The Litmus Test of Regionalism ………………………………...179 6: The Framing: Europeanisation ………………………………... 189 1 The Enlargements ……………………...…………………….…………….. 190 2 The EU Northern Dimension Initiative ………...…………………………... 201 3 Analysis: Between Circles and Rings …..………………………………….. 215 4 From the Second Phase Back to the Future ………………………………... 220 7: After 2004 …………………………………………………….…. 227 1 At the Dawn of the Enlargements ……..…………………………………… 227 2 The Model of Environmental Security …………………………………….. 230 3 Analysis ……………………………………………………………………. 238 4 Applicability ……………………………………………………………….. 251 Conclusions ………………………………………………………... 266 1 Conclusions ………………………………………………………………... 266 2 Whither and Whether to Widen: towards a ‘Wider Europe’ ………………. 282 Bibliography………………………………………………………... 301 Abstract ……………………………………………………………. 343 ix Introduction which political association is viewed as resulting from fear; to David Mitrany’s functionalist integration, in which the existence of common concerns justifies cooperation; to Karl Deutsch’s security community, in which cooperation is associated with ‘peaceful change’, just to mention a few (cf. Joenniemi and Browning 2004, 4). These reflections do indeed go to the roots of the different theoretical traditions within IR. The analytical framework proposed in this study will refer to these reflections in order to analyse regions. It will build on these arguments correlating security and cooperation in order to understand the correlation between regional security and regional cooperation. This statement introduces one central feature that this framework will contain, in fact the most important one. In view of the aforementioned richness in the theoretical spectrum and the diversity of the Baltic Sea milieu, I have chosen not to embrace any theoretical viewpoint a priori, which would link this analysis to one specific theoretical interpretation. On the contrary, I aim to develop a framework that can allow me to construct the analysis according to how the actual correlation between security and cooperation in the Baltic Sea area has unfolded. The fact that there are numerous interpretations of the study of regions, and indeed of regional security, does not entail that some of them are right and others are wrong. As Peter Katzenstein persuasively argued, “regional designations are no more “real” in terms of geography than they are “natural” in terms of culture” (Katzenstein 1996, 3). Thus, I will refrain from selecting a theoretical interpretation of reality before reality has unfolded. Instead, I will wage and construct my interpretation of reality upon it. Hence, as far as the first research question is concerned, the theoretical aim of this study is to develop an analytical framework to explain the construction of regional cooperation by actors through the analytical construction of the security interdependence existing at the regional level. The framework that this study proposes to achieve this objective builds on the Region Building Approach (RBA) originally elaborated by Iver B. Neumann (1992, 1994), which I refer to as the ‘classical’ RBA. This approach is suitable because of its eminently constructivist core, and, not least, because of its relatively wide application to the case of the Baltic region. However, in order to insert the security component, and thereby explore the relationship between regionalism 6 Introduction and security, the framework that I put forward here represents a development of Neumann’s approach. I will do so by correlating his classical RBA to the reflections on security by the so-called ‘Copenhagen School’, particularly in relation to its ‘critical’ Regional Security Complex Theory (RSCT). The critical RSCT is a major outcome of the collegial work performed by Ole Wæver concerning the conceptual aspects of security (see Wæver 1995) and by Barry Buzan, particularly as regards the regional aspects of international security (see Buzan 1991, 186-229). The result (Buzan et. al 1998; Buzan and Wæver 2003) is a comprehensive framework for the analysis of regional security interdependence. The critical RSCT, as the classical RBA, is radically constructivist. This affinity constitutes the crucial point of contact to put forward my framework, the critical Region Building Approach. As the name suggests, this is not a framework in its own right; rather, it proposes a combination under certain conditions of Neumann’s classical RBA with some critical reflections on regional security by the ‘Copenhagen School’. As far as the second research question of this dissertation is concerned, the primary aim of the empirical analysis is rather straightforward. There is an explanatory element in it, in the sense that it examines the regional cooperation in the BSR through the analysis of the dynamics of regional security interdependence from 1989 to 2004. There is also an exploratory element, which is the post-enlargement scenario. Hence, I will first interpret the construction of the regional cooperation through the analytical construction of the regional security from the post-Cold War to the 2004 EU and NATO enlargements. I will then envisage a likely post-enlargement scenario. This is where I will not be able to construct my analysis upon facts, since reality has yet to display itself. Hence, I will develop this part of the analysis on the basis of a case study, namely the case of environmental security and cooperation in the Baltic region. This case, I will demonstrate, is taken as the model upon which I will tailor my projection vis-à-vis the future of security and cooperation in the BSR at large. The third research question also refers to the more explorative empirical aims of this study. It deals with the lessons to be learned from such analysis with respect to the wider European scenario. As noted, the Baltic Sea experience will be analysed in close correlation with European developments. In this respect, I will focus on the 7 Introduction relevance of the Baltic Sea Region for EU external relations with its neighbours, which is the aspect of this analysis that is considered most topical, for reasons that will become clearer as the narrative unfolds. For now it is sufficient to note that, geographically speaking, the BSR is not located at the centre of the enlarged EU, a position that is common to other areas such as the Mediterranean Sea and the southeastern areas of the enlarged EU. Cooperation in these areas responds to different needs and is characterised by different features. However, the regionalisation of security in these areas and their possible coordination are increasingly viewed as a sound tool for managing EU cooperation with its neighbours (see Commission of the European Communities, 2003b). The Baltic Sea, in this respect, has experienced developments that can serve as inspiration. This is particularly relevant vis-à-vis Russia, the main EU neighbour and prominent strategic partner. Since the Baltic Sea region is – and is bound to remain in the foreseeable future – the only spot where the EU physically meets Russia, these fifteen years of regional experience will provide some indications with regards to EU-Russia relations. Hence, the empirical aims of this study, as formulated in the second and third research questions, are to discuss the development of the regional cooperation through security analysis in the Baltic Sea Region from post-Wall to post-enlargement, and to draw lessons from this experience for Europe and EU-Russia relations in particular. The Relevance These objectives mainly make such study relevant for students of the Baltic region. In this sense, the main asset of this study is that it has been carried out shortly prior to the 2004 enlargements, which is the point of reference that existing sources and discourses have referred to. This timing will enable me to produce a factual account of that which for most other students of this region was an exploratory exercise of predicting likely scenarios. In this respect, my own exploratory exercise regards the post-2004 scenario. In view of the arguments presented in this introductory section, however, there are also two other more far-reaching elements, which make the present study relevant. The first element relates to the theoretical insights provided by the framework of analysis. The critical RBA draws upon the rich theoretical literature on both regionalism and 8 Introduction security, and is a useful tool for framing the correlation between security and regionalism. Much of the theoretical work focusing on international regions has either focused more explicitly on one of the two elements, i.e. regional security (e.g. Buzan and Wæver 2003; Lake and Morgan 1997) or regionalism (e.g. Hettne 1994; Hurrell 1995), or has more or less consciously combined the two (e.g. Söderbaum 2002), the result being that the analysis of regional security becomes regionalism, and vice-versa. This study aims to cover this blind spot of the existing theory, first by keeping them analytically separated and second by suggesting a way to correlate the two aspects. The second far-reaching element of relevance is for students and practitioners of EU foreign policy. The Baltic Sea Region, it was argued, is positioned in a strategically peculiar location within the framework of European geo-politics. It is on the periphery of the EU. Nonetheless, it includes one of its major players, Germany, and its most important neighbour, Russia. The EU, I noted, has valued this peculiarity and contributed to the regional initiatives taking place in the Baltic. In a sense, Brussels has actually regarded the regional experience in Northern Europe as a pilot project for its external relations vis-à-vis its neighbours in the East (cf. Commission of the European Communities 2003a, 8). The developments brought about by the 2004 enlargement allow to draw some conclusion out of this engagement, and to design prospects for future policies in view of this experience. The Structure of the Dissertation This study is divided into two parts. The first part sets out the framework of the analysis and aims to answer my first research question. The first chapter reviews and systematises different approaches to regionalism, i.e. regional cooperation by actors, in IR theory. A bi-dimensional matrix will be designed to categorise the various theoretical approaches to regionalism. Towards the end of the chapter, the bi-dimensional matrix will be reduced and simplified into a one-dimensional ‘continuum of approaches’ (following but slightly revising a categorisation originally developed by Iver Neumann) in order to put forward my analytical framework. The second chapter introduces my analytical ‘prism’: security analysis. Here, different approaches to security are reviewed and 9 Introduction categorised. The analysis of regional security is then introduced, with particular reference to the RSCT in both its classical and critical versions. After this review, the correlation of the two analytical pillars characterising the analytical framework of this dissertation is put forward. I will begin by completing the discussion of Iver Neumann’s classical RBA (Neumann 1992; 1994), which constitutes the most direct inspiration for my framework. I then unfold my own critical version, which, as the classical model, is deeply rooted in constructivism, but develops three arguments, which are inspired by the critical RSCT of the Copenhagen School, and focus on the correlation between security analysis and the regionalist debate. The third chapter lays out the methodological implications of this analytical framework. First, its radically constructivist epistemology is discussed and problematised in view of its two ‘ancestors’: the classical RBA and the Copenhagen School framework. The centrality of discourse analysis as the main method and source of the analysis is then accounted for. The chapter is concluded by highlighting the methodological aspects of the empirical study and thus paves the way for the analysis itself. The second part of the study constitutes the actual analysis. The narrative is divided into three chronological phases, whose rationale and limits will be more thoroughly explained at the close of the methodological discussion. The first phase (Chapter 4) will assess the rise of regional cooperation in the BSR on the basis of the discourses on regional security from the late 1980s to the early 1990s. The second phase (Chapters 5-6) discusses that which I have termed the framing of the regional cooperation. In Chapter 5 the discourses characterising the foreign and security policies of the littoral countries vis-à-vis the BSR will be in focus. This discussion is followed in Chapter 6 by an analysis of the security discourses at the European level vis-à-vis this region. The chapter is concluded by discussing how these security discourses concerning the BSR emerging from the countries’ foreign policies together with European developments explain the regional cooperation until 2004. The third phase (Chapter 7) discusses the future of regional cooperation after 2004. As noted, here the case of environmental security is introduced and unfolded as a model. This model is then discussed and applied to other sectors, i.e. civil security 10 Introduction and economic cooperation, in the light of current discourses on the future of the region. In the Conclusions of the dissertation, I will first summarise my answers to the first two research questions. I will then complete and systematise my answer to the third question by putting forward a number of conclusive arguments on the wider relevance of fifteen years of Baltic regional cooperation for EU external relations with its neighbours and Russia in particular. 11 Introduction 12 Regionalisms in International Relations Theory Part I The Framework 13 Regionalisms in International Relations Theory 14 Regionalisms in International Relations Theory 1 Regionalisms in International Relations Theory 1: Framing the Debate The study of regions has occupied students of International Relations (IR) for a relatively long time (Russett 1967; Nye 1968; Cantori and Spiegel 1970; Lindberg and Scheingold 1971; and Thompson 1973 are among the earliest and more comprehensive contributions). Terms such as regionalism, regionalisation, regional cooperation and regional integration have broadly entered the theoretical and empirical jargon in analytical works of various kinds. More often than not, there is not perfect harmony in terms of the manner in which these terms are employed in the different sources. As a result, the attempt to put forward a comprehensive review may easily fall short as a rather sterile struggle focusing on attaching and detaching different labels to different concepts (see Hurrell 1995; Söderbaum 2002; Neumann 1994; Hettne 1994 for different categorisations). Hence, mapping different regionalist approaches in IR theory is not merely an ambitious task; it is also highly contestable. The overview of regionalist theories presented in the following therefore makes no pretence of providing an exhaustive analysis; rather, it offers a general picture of the defining trends, bearing in mind that its foremost objective is to introduce a number of concepts to be utilised in the analytical framework of this study. As already noted in the Introduction, Iver B. Neumann’s Region Building Approach (RBA) (Neumann 1992 and 1994) provides the fundamental inspiration for my analytical framework. Hence, his way of categorising regionalism is devoted particular attention. Notably, the categorisation of regions itself is to some extent sidelined in the actual RBA, because the focus of the actual framework is on the more radical constructivist aspects of regionalism; however, I will return to this in the next chapter. For the moment, I will refer to the principles guiding the categorisation of regionalism in that approach. I therefore introduce the idea of the so-called continuum of approaches, which characterises Neumann’s proposal (Neumann 1994, 53-57). Different approaches to 15 Regionalisms in International Relations Theory disquisition regarding their external goals. This then brings us to the next approach. Liberal Theories The inter-paradigm debate between neo-realism and liberal theories does not pertain to the scope of this study and goes beyond the rationale of this critical appraisal of regionalisms in IR theory. However, the fact that “the major developments in the Liberal tradition of international relations theory in the post-1945 period occurred in studies of regional integration” (Nye 1988, 239) mainly has to do with the theorisation of the formation of interests at the heart of liberalism which, as noted, is a blind spot in neo-realist thinking. Neofunctionalism and liberal institutionalism are the two most popular ways for explaining how liberalism has filled this gap. Neo-functionalism challenges classical functionalist theories (Mitrany 1966) inasmuch as it regards cooperation in a rather utilitarian manner, i.e. ‘function follows interests’, rather than, as in the case of functionalism, as a strategy to achieve peace, i.e. ‘form follows function’ (see Söderbaum 2002, 6). Ernst Haas was the primary figure behind the development of neo-functionalism in the 1960s. This approach was regarded as a rather successful analytical tool for explaining regional integration4 at the time, and it was most successfully applied to the case of Western Europe. This empirical case, I shall argue, simultaneously constitutes a point of strength and weakness in this approach. Neo-functionalists also regard the state as the primary and most significant actor in regional phenomena – including those elites that influence state behaviour. Haas, in this respect, argued: “The study of regional integration is concerned with explaining how and why states cease to be wholly sovereign, how and why they voluntarily mingle, merge and mix with their neighbours so as to lose the factual attributes of sovereignty while acquiring new techniques for resolving conflict between themselves.” (Haas 1970, 610) This quote also suggests that states balance the partial loss of sovereignty following regionalisation by means of some sort of payoffs 22 Regionalisms in International Relations Theory deriving from this collective action. In relation to the latter, Haas also argues that regional integration is enhanced when these collective actions are “made incrementally, based often on consequences not initially intended by the actors (governments and important interest groups)” (Haas 1970, 617). These unintended but beneficial consequences enhance regional integration itself and introduce the theorisation of the spillover effects, which is one of the neo-functionalist building blocks. Spillover effects can be distinguished in terms of ‘scope’ and ‘level’. Scope, in Haas’ words: “is confined to decisions and objectives relating to the realisation of full benefits” (ibid.), which is directly connected to the nature of the interests of state actors involved in the regional integration process and to their ability to pursue these interests. Spillover in terms of ‘level’, instead, refers to the “little progressive penetration from supranational institutions into the lower reaches of decisionmaking at the national and local levels” (ibid.). The result is that the regional ‘outcome’ of neo-functionalist-like integration provides ‘feedback’ to the various stakeholders involved in the region in order to maximise their benefits (the case of the European Economic Community in the 1970s is the classic example in neo-functionalist literature). Neo-functionalism tends to have a short-term orientation regarding the maximisation of results by actors benefiting from regional integration. This relates strictly to the assumptions that some of the regional outcomes might be unexpected (the spillover), and that states (and national elites) are incapable of interpreting the processes leading to a certain regional result. This implies that states are neither certain to repeat that regional outcome, nor willing to take risks. Hence, even in its most sophisticated versions, neo-functionalism “avoids normative assertion” (Haas 1970, 627). This makes this theoretical formulation particularly flexible and applicable in various empirical cases. By not posing precise hypotheses, neo-functionalism offers a highly empirically oriented formulation and rejects the option that regional integration can be explained in terms of “systemic generalizations” (ibid.). This peculiarity strongly contrasts with the neo-realist framework that, as was noted, “tells us a small number of big and important things” (quoted in Nye 1988, 242). 23 Regionalisms in International Relations Theory One of the most evident shortcomings of neo-functionalism then relates to the contingency of its predictions. The absence of epistemological foundations renders the neo-functionalist claim somewhat deficient as far as its actual ability to explain empirical phenomena is concerned. In other words, by solely studying the nature and the degree that determine spillover effects, neo-functionalism rejects theoretical generalisations as inapplicable by generalising empirical outcomes which are equally inapplicable in the long term. Theoretical generalisations can be granted a certain degree of vacuity, and a modicum of empirical inconsistencies do not render a theory inconsistent; as Waltz put it: “a few false predictions do not falsify a theory” (in Nye 1988, 243). Yet, a whole theorisation based mainly (if not only) upon its applicability to empirical cases jeopardises its potential for resisting empirical inconsistencies, thus leaving it vulnerable. Most notably, neo-functionalism lacks dependent variables with which to frame the virtuous circle of the spillover effects (see Haas 1970, 628). As a result, the neo-functionalist architecture is considered durable because its foundations are strongly anchored to the empirical ground. Nevertheless, the empirical grounds may vary considerably and render the framework inapplicable, as demonstrated by application to regions other than Western Europe. Within liberal theories, liberal institutionalism “has become the dominant approach […] (that is, as far as regionalism is concerned)” (Söderbaum 2002, 6-7). As is the case with other traditionalist approaches, liberal institutionalism is also strongly centred on the state, the latter being understood as a selfish actor seeking to maximise its gains; this is reminiscent of neo-realist claims. Liberal institutionalism differs from the neo-realist argument, however, inasmuch as it has retained liberalist concern regarding the formation of interests. Hence, it is argued that states, in order to ensure their own survival, may pursue interdependence with other states where common interests are envisaged. Interdependence therefore constitutes an important element of the “changing conceptions of self-interest” (Keohane 1993, 271). Moreover, the existence of common interests is a premise for 24 Regionalisms in International Relations Theory cooperation, and thus regionalisation, which represents another essential point of the liberal institutionalist claim. Liberal institutionalists argue that states promote the creation of formal and informal institutions in order to facilitate the solution of common problems and to coordinate action. This international (regional) cooperation is defined by the creation of international regimes (Krasner 1983) that are functionalised, issue-specific and concerned with the maximisation of the benefits attainable via regional institutionalised transactions. These institutional regimes do not therefore have “mythical positions of authority over the state” (Keohane 1993, 274); rather, they are drawn upon to advance state interests “by facilitating the making and keeping of agreements through the provision of information and reductions in transaction costs” (ibid.). It is argued that as a greater number of common problems are solved through cooperation, it becomes increasingly structured and can achieve a vast range of goals spanning from growing economic transactions to the diffusion of democracy. Liberal institutionalists claim that participating states can obtain absolute gains by means of (regional) cooperation (as opposed to the relative gains postulated by the neo-realists, cf. Waltz 1979, 105). Yet, they also argue that states may move beyond the mere preservation of a specific equilibrium, as in the neo-realist argument as regards alliance formation. Actually, the benefits arising from issue-specific cooperation may lead states to intensify interdependence and institutionalisation, thereby enhancing a degree of regional cohesion (see Hurrell 1995, 64). 3: ‘New’ Regionalisms Hettne’s definition of ‘new’ regionalism frames regional phenomena from ‘within’ and from ‘below’. Considering phenomena such as globalisation ‘from within’ appears debatable, as I will argue in the following. On the other hand, regionalism ‘from below’ is a useful image for introducing critical theories on regionalism. Also in this case, the ‘from below’ label might be perceived as a simplified schematisation for very diverse paradigms. However, it can serve as a useful starting point to mark the difference from the ‘old’, traditionalist and state-centred regionalism ‘from above’. 25 Regionalisms in International Relations Theory Globalisation Globalisation is traditionally viewed as a systemic approach (see Cox [1992] 1996). Hence, globalists, like neo-realists, view regions resulting from the international system. Globalisation should thus be regarded as an ‘outside-in’ approach rather than ‘from within’. As mentioned above, this possible analogy between globalisation and neorealism is one of the arguments for framing this review in terms of ‘old’ and ‘new’ regionalisms. The incommensurable ontological contrast between these approaches, particularly as to how actorness in regional phenomena is framed, presents another reason to do so. While rationalist approaches consider regionalism as the result of a ‘problem to be solved’, globalisation, and ‘new’ regionalisms in general, focus on the nature and evolution of the participation of various subjects in the international arena, which may result in the formation of regions. Let us explore how. The concept of globalisation is built upon a loose image; in fact, it has developed more as a metaphor than as an actual theory. It focuses on the borderless mobility of people, on the flow of capital, and on the movement of values and ideas taking shaping in the global arena (see Hurrell 1995, 54). One can argue that globalisation, with its borderless dimensions (Ohmae 1991), determines a sort of ‘end of Geography’ (O’Brien 1992), i.e. it is connected to territory in a rather unconventional manner. This territorial aspect actually constitutes the cue to understanding regional processes within globalisation. Concepts such as the ‘end of geography’ are associated with the end of the Cold War, since bipolarity determined a territorial (for not to mention ideological) demarcation limiting actors to a binary choice. In this sense, the end of the Cold War is an ‘end of Geography’ inasmuch as it broke bipolar territoriality and thereby deterritorialised space (see Forsberg 1996, 357). Nevertheless, this ‘deterritorialisation’ is also, and most notably, connected to the progressive weakening of the modern Westphalian conceptualisation of space, which identifies states as the primary holders of sovereignty. Since politics, to quote Anthony Giddens, is “any system of rule as comprising legitimate dominion over a spatial extension” (quoted in Ruggie 1993, 148), some globalisation theorists have associated ‘deterritorialisation’ with the reorganisation of the post-statist space and therefore to ‘post-sovereignty’ (Forsberg 1996, 26 Regionalisms in International Relations Theory 364). Hence, post-sovereignty is regarded in the task at hand as a synonym of post-Westphalianism, i.e. a redistribution of power structures with respect to post-statist territory, which suggests that ‘deterritorrialisation’ is followed by reterritorialisation (see Wæver 1997, 298). This dynamic of reterritorialisation is most evident in multinational financing – which is commonly referred to as the globalisation phenomenon par excellence, since global trade, it is argued, “cannot work entirely without a territorial basis and practice” (Forsberg 1996, 368). Less explicitly, reterritorialisation also applies to the alleged emergence of a global ‘community of values’, which is another defining pillar of the ‘loose image’ of globalisation. In effect, ideas may move and spread globally over space and time, but they must also permeate society and individuals in order to not weaken and ultimately vanish. This concept is often correlated with the notion of ‘glocalisation’, i.e. the link between the local and the global dimensions (see Albert 2000), which, in fact, may be regarded as a form of reterritorialisation. These brief items of reflection concerning the relationship between territory and globalisation lead us to consider the role played by regions in globalist approaches. Globalisation stands in opposition to regions in many senses, because it opposes the notion of ‘modernist’ territorial sovereignty. Hurrell, in this respect, argues that: “globalisation works against the emergence of regionalism” (Hurrell 1995, 55). Yet, as noted, the post-sovereign space also follows some sort of reterritorialisation. Hence, actors (and especially the nonterritorial/non-Westphalian ones) may reterritorialise the global space, and it is here that a different way of framing regions enters the globalist debate. Within this context, globalisation encourages regionalism rather than opposing it (see Katzenstein 1996), and regions emerge as a result of the natural reshuffling of the international arena operated by actors who reterritorialise the global space. This shift is most easily observed in the fields of the environment and economic cooperation. The environmental field is often regarded as challenging the supremacy of the state power, because the very nature of environmental problems is usually perceived as being global (Forsberg 1996, 368-369). In order to tackle environmental problems, however, the global picture is often broken down into dimensions that may cope more effectively with the specific issues. Within the context 27 Regionalisms in International Relations Theory of this study, this defines a reterritorialisation of global environmental threats from which the formation of regions derives. With respect to economic cooperation, the leitmotif of debates focusing on global economy is that an unregulated expansion of economic interdependence disregards and possibly exacerbates social disparities (Gamble and Paine 1996; Hurrell 1995, 55). Hence the world order, which Robert Cox defines as “neutral as regards the nature of the entities that constitute power” (Cox [1992] 1996, 494), may end up fostering disproportions and inequalities in the distribution of wealth and resources. Here regionalism intervenes by reterritorialising the global economy and taking back some of the dispersed political control (see Katzenstein 1996, 4-8; Forsberg 1996, 368). Two observations may follow from these examples: first, it may be argued that the formation and development of regions appears to be rather function-driven, i.e. to manage a certain field more efficiently. As Andrew Gamble and Anthony Paine write, “regionalist projects emerge as a means to help achieve the globalist project in a world where there is no longer a single state with the authority and capacity to impose its leadership” (Gamble and Paine 1996, 252-253 and quoted in Söderbaum 2002, 12). The quote suggests that, in a sense, regionalism within the framework of globalisation may itself turn into a rather Westphalian formation and formalisation, a ‘fortress’, a sort of ‘region-state’ (see Ohmae 1993). The second, more general, observation is that the rationale coupling globalisation and regionalism focuses on the complementarity of the two dynamics. This complementarity is based on a rather physiological and natural readjustment of territory and sovereignty intended to cope efficiently with global challenges. This approach is persuasively synthesised by Peter Katzenstein, who argues, “globalisation and regionalism are not antithetical. […] Globalisation and regionalism are complementary processes. They occur simultaneously and feed each other” (Katzenstein 1996, 4-5). In this respect, Held and McGrew are, if possible, even more explicit: “regionalism has not been a barrier to contemporary political globalisation – involving the shifting reach of political power, authority and forms of rule – but, on the contrary, has been largely compatible with it” (Held and McGrew 2000, 12). 28 Regionalisms in International Relations Theory Critical Theories The main conceptual contribution provided to regionalism by critical theories focuses on the roots and features of social participation in the regional phenomena. According to critical interpretations, regions are about inclusiveness and participation (Paasi 1991, 239-240). Any subject can be part of constructing a political community, and this community can result in the construction of a region. According to critical authors, this also implies that subjectivity in international relations, and ultimately power, can rest anywhere and thus are no longer necessarily state-based. Hence, regional actorness is characterised by the ‘commonality’ of the social groups taking part in the region, i.e. cultural, historical and social affinities (Paasi 1991, 240). As Hurrell explicitly writes, “Constructivist5 theories focus on regional awareness and regional identity, on the shared sense of belonging to a particular regional community, on what has been called ‘cognitive regionalism’. They stress the extent to which regional cohesion depends on a sustained and durable sense of community based on mutual responsiveness, trust, and high levels of what might be called ‘cognitive interdependence’.” (Hurrell 1995, 64) Hence, the core of critical theories to regionalism resides in the identity-building process that originates from it. Regional identity is achieved when ‘we-ness’ is detectable among the participants of the regional project and to the sense of belonging shared by regional actors. These initial considerations make it somewhat natural to introduce the debate on security communities, which is of central importance when theorising critical approaches to regionalism and, as the term suggests, will be present in the discussion on regional security in the next chapter. With respect to regionalism, it may in fact be argued that the entire debate on identity building and the sharing of social values that was introduced as one of the building blocks of critical approaches is also applicable to the framework of security communities and its further elaborations. The two classics that have characterised the study of security community are the works by Karl Deutsch (1957) and Adler & Barnett 29 Regionalisms in International Relations Theory (1998). The former work actually represents the very core of the study on security communities, while the extensive research conducted within the framework of the latter in the mid-1990s, about forty years after the original, contributed to revitalise the framework. In an historical and scientific milieu in which IR theory was virtually monopolised by the realist paradigm, Karl Deutsch elaborated the concept and framework for security community based on three major pillars: integration, sense of community and peaceful change. In Deutsch’s words, the first refers to the “attainment […] of institutions and practices strong enough and widespread enough to assure for a long time, dependable expectations of peaceful change among its population” (Deutsch 1957, 5). ‘Sense of community’ implies that “common social problems must and can be resolved by process of peaceful change” (ibid.). ‘Peaceful change’ is invoked in the two previous definitions and is defined as “the resolution of social problems, normally by institutionalised procedures, without resort to large-scale physical force” (ibid.). Without excessively delving into the reasoning behind the virtual abandonment of this approach during the Cold War decades, one aspect deserves particular attention: the downplaying of the conceptualisation of state actors. It is not ignored, since Deutsch refers to that which he labels as ‘pluralistic’ security communities, i.e. composed of sovereign states. Nevertheless, “transnational and interstate interaction” (Adler and Barnett 1998, 30) is the object of analysis and represents that which security communities are primarily about. Obviously, this represented a powerful conceptual breakthrough in the late 1950s. Deutsch’s reflections therefore anticipated and founded the crucial shift from the ‘state’ to ‘individuals’, which, I will argue, has also been central in the security debate. Naturally, this then provides the most obvious point of departure from which to attack his statement. The downplaying of the state constituted a frontal attack on the dominant realist paradigm. Moreover, Deutsch’s attempt to apply security community theories to an imagined Atlantic community was viewed at the time as a rather hypothetical experiment “not in tune with the overall situation of power politics and spheres of influence during the cold war” (Stadsholt 2001, 3). Given these basic notions, a reformulation of security community theories appears to be naturally well-rooted in these critical 30 Regionalisms in International Relations Theory formulations of regionalism. Post-sovereign paradigms could well fit into Deutsch’s initial hints and fill its vacuums. Moreover, the post1989 political setting could also support and perhaps inspire this reconceptualisation. Hence, Adler and Barnett’s work is valuable for the further development of Deutsch’s contribution although, arguably, it also sensibly modifies his original idea in a more radically constructivist light. First, the role of the state required retheorisation. As the authors claim, “After all, states usually maintain that their actions are performed in the interest of the wider community when, in fact, they derive from much more selfish concerns” (Adler and Barnett 1998, 31). Therefore, the new paradigm on security communities does indeed follow the original inputs by Deutsch, but has derived from it a new reconceptualisation of state (which may spur a longer discussion regarding Deutsch’s limits6). Second, according to Adler and Barnett, communities evolve around three basic pillars: “shared identity, values and meanings […] face-to-face encounters and relations in numerous settings […] long term interests and perhaps even altruism” (Adler and Barnett 1998, 31). These features follow Deutsch’s conceptual claim rather closely; however, one may note that geographical contiguity does not reflect a necessary condition for the construction of a security community and therefore a region. Regions, as the entire reality, represent a discursive practice that may vary according to the meanings granted to them and the contexts in which they are used. For instance, since ‘shared identity values and meanings’ are a crucial characteristic of communities “Australia is a member of the Western security community even though it resides thousands of miles from the “core” members” (Adler and Barnett 1998, 33). Finally, Adler and Barnett note that a community satisfying these three conditions does not necessarily generate a security community. As the two scholars maintain, “Sometimes a community of states will establish pacific relations, sometimes a community will not. But those that do have formed a security community” (Adler and Barnett 1998, 37). This final argument can actually be considered as a sort of line of demarcation in the debate on security communities. To put it bluntly, it separates the ‘security’ from the ‘community’ dimensions of the debate. It is a very porous border and, actually, the two elements should not be separated at all. Within this overview, however, I have 31 A Critical Region Building Approach extraordinary proliferation and a conceptual breakthrough in the discipline of IR. Cold War security studies in particular were driven by a focus on military and political security concerns and thereby on confrontation and deterrence: from the Western perspective, the Soviet Union was the threat to be contained and around which the entire dynamic of international security revolved. The end of the Cold War brought about major conceptual developments which, according to Buzan (1997), can be categorised in four broad trends: traditionalists, wideners, critical schools and the socalled ‘Copenhagen School’. This overview will proceed to discuss the concepts of security according to Buzan’s categorisation. I will employ the figure above for the purpose of introducing this discussion, as it maps the four theoretical trends in terms of security actors (state/individuals) and security agendas (narrow/wide). The traditionalist trend (see Gray 1993) deriving from rationalist paradigms represents the first rather homogeneous category within security studies. Their argument was particularly apt to function in a Cold War scenario. In such a framework, security is concerned with the ‘narrow’ agenda of the military and political spheres. Such logic delimits rather strictly the range of that which is referred to as ‘referent objects’ in this overview, i.e. “things that are seen to be existentially threatened and that have a legitimate claim to survival” (Buzan et al. 1998, 36). Moreover, realist security also limits the range of ‘securitising actors’, which in this discussion of security studies represent “actors who securitise issues by declaring something – a referent object – existentially threatened” (ibid.). According to traditionalist (or ‘narrow’) security paradigms, the state is the sole referent object simultaneously serving as a securitising actor. The international arena is anarchic and generates fear. Since the purpose of mankind is to strive for survival, this very fear constitutes the core of security. Mankind is willing to sacrifice its natural rights and delegate them to the one actor capable of achieving security and thereby guaranteeing survival. That actor is the state, which is supposed to be the main guarantor of security and whose survival is the main guarantee for security. The second trend characterising security studies is liberalist thought (see Nye and Lynn-Jones 1988). Such a paradigm advocates the need to widen the spectrum of the security agenda. This can be considered to 38 A Critical Region Building Approach be the most important input to security studies provided by liberal thought. While the political–military realm does indeed offer a very precise and objective understanding of what security is about, as well as about its actors and referent objects, it is also clearly reductive as regards the observation of the complex reality of security interdependence. Liberalists, as was discussed in the previous chapter, claim that international relations are characterised by cooperation and interdependence, instead of confrontation and competition (see Joenniemi and Browning 2004, 8-10). This also means that the security agenda cannot be narrowly construed in terms of military and political security alone; rather, it must also be made to include a wider variety of issues – issues which one may refer to as ‘non-military’ – that are not included in this ‘narrow’ categorisation, but do continue to produce significant threats to survival; they thus merit being regarded as ‘security’ issues. Such ‘non-military’ threats can be categorised differently and thus refer to fields as diverse as the environment, economy, migration, crime, etc. Traditionalists challenge this widening operation of the security focus, as the most diverse range of issues could potentially be viewed as belonging to the field of security. As Stephen Walt powerfully notes, “Defining the field in this way would destroy its intellectual coherence and make it more difficult to devise solutions for any of these problems” (quoted in Buzan 1997, 10). One may indeed ask what it is that ties these different issues together conceptually, i.e. what it is that they have in common beyond the negative attribute of being ‘non’- military. Liberalists have not provided a clear answer in this regard, although they have provided some input. Most importantly, the liberalist ‘widening’ refers to both the securitising actors and referent objects of the security move. Traditionalists, it was noted, monolithically place states at the centre of the security move. The narrow focus of the military and political realms indicates that the state is the only unit generating – and benefiting from – security. The liberalist ‘widening’ implies that actors that contribute to, and are affected by, the security move are inevitably ‘wider’ as well. Hence, as the security agenda widens, the circle of involved actors also expands. This suggests that although liberals remain anchored to a rather statecentric view of international relations, they emphasise the importance of dynamics capable of transforming states’ perception of security. 39 A Critical Region Building Approach Given the important contribution of liberalists to the widening of the security agenda, the next, rather obvious, steps are thus: 1) to observe ‘to what extent’ it can actually be widened, i.e. the extent to which an issue can indeed be regarded as a security issue, and 2) to qualify the non-state actors involved in the security move. Both aspects are first and foremost a point of ‘coherence’. Buzan (Buzan 1997, 9) remains sceptical about the prospects for achieving coherence when it comes to widening the security discourse in both directions (agendas and actors). On the other hand, he also argues, such coherence does not necessarily have to be an objective of the process. These observations contribute to the sketching of the third trend within security thinking, Critical Security Studies (CSS), in which Buzan’s considerations on the ‘extent of the widening’ of security agendas and actors are particularly relevant (see Krause and Williams 1997). CSS “is meant to imply more an orientation toward the discipline than a precise theoretical label” (Krause and Williams 1997a, x-xi). This also suggests that CSS does not in fact represent a theoretical trend unto itself, and that some degree of oversimplification is unavoidable. The various currents referring to CSS are all marked by strong constructivist connotations. Reality, and in this case the security move, does not respond to fixed rules. Security is the result of an individual understanding and can be represented in multiple ways and with multiple combinations. In this sense, CSS postulates the ultimate widening, since security can be generated and practiced virtually anywhere. An interesting aspect related to this construction of the security move is the representation of securitising actors and referent objects. The security move can potentially involve all possible subjects present in the international arena, from the most structured ones, e.g. states, to the least structured, e.g. individuals. This aspect marks a clear shift from traditionalist epistemology, and is instrumental in introducing the fourth category within security studies, the so-called ‘Copenhagen School’. The ‘Copenhagen School’ The Copenhagen School framework primarily refers to work carried out by Ole Wæver and Barry Buzan together with a group of scholars 40 A Critical Region Building Approach affiliated with the Copenhagen Peace Research Institute. It has produced an alternative perspective on security studies, from which a relevant application to regional security follows. The framework can be included within the ‘widening’ category of security studies and, as in CSS, is inspired by a radically constructivist epistemology, although in a slightly different manner. I will begin to sketch these features by observing the ontological status of the concept of security. According to the proponents of the Copenhagen School, the incoherence emerging from the ‘widening’ operation cannot be totally eliminated, and certainly not by once again retrenching to a narrow sense of security. The logic ought to be one of maintaining a degree of unavoidable incoherence but then to attempt to re-examine the ontological foundations of security. For Ole Wæver, the member of the ‘School’ who has been more prominently engaged in the conceptual aspects of the issue, the core of the dispute rests in the conceptualisation of security (Wæver 1995). The focus here is on the importance of discourses in shaping the concept of security, which is at the heart of that which is defined as the ‘speech act approach’: “In this usage, security is not of interest as a sign that refers to something more real; the utterance itself is the act” (Wæver 1995, 55). An issue is security issue if it is talked as a security issue. As in the CSS, this conceptualisation opens up the security practice to actors beyond the state. More importantly, as in the case of CSS, it opens the conception of security to subjectivity rather than accepting the objective reality of the realists and liberalists. One may then wonder about the differences between the two approaches. The Copenhagen School is even more radically constructivist than CSS (cf. Buzan 1997, 20). The core of the discursive practice of the Copenhagen School is not merely that security is to be ‘spoken into existence’. More importantly, the security move must also be ‘endorsed’ (Buzan 1997, 13). It is not sufficient for an issue to transcend mere politicisation to become a security issue. Prior to this elevation to security, and even more importantly, the issue must be endorsed by leaders. Once politicians, intellectuals and experts talk about an issue and refer to it as a ‘security’ issue, they turn it into security on the strength of their leadership. The core of the speech act approach therefore rests on the fact that by talking about it, leaders ‘endorse’ the issue and elevate it to a security issue. This ‘dramatic’ 41 A Critical Region Building Approach move, by which leaders construct security “as a more extreme version of politicization” (Buzan 1997, 14), is referred to as securitisation (see Wæver 1995). Therefore, as in CSS, the security move cannot be objectified, since some actors can consider it as a threat, while others may not: “‘Security’ is thus a self-referential practice, because it is in this practice that the issue becomes a security issue” (Buzan et al. 1998, 24). The use of textual analysis is meant to unfold this subjectivity; however, this is also an important argument whereby the Copenhagen School distances itself from the CSS. In effect, the speech act of ‘endorsement’ by leaders implies that the constructivist nature of the reality cannot be totally subjective in the sense that individuals are not the only unit of analysis. The perception of security must be recognised by a group in order to become ‘real’. A number of isolated individuals cannot transform a threat into a security issue merely by saying so. In this respect, while the Copenhagen School remains constructivist in philosophical terms, it is closer to traditionalist thinking when it comes to limiting the excesses of the widening emanating from CSS. As Buzan noted: “While our philosophical position is in some sense more radically constructivist in holding security to be always a political construction, not something the analyst can tell about how it ‘really’ is, we are in our purposes closer to Traditional Security Studies (TSS), which at its best attempted to grasp security constellations and thereby steer them into benign interactions. This is in contrast to the ‘critical’ purposes of CSS, which point towards more wholesale refutation of current power wielders.” (Buzan 1997, 20) This means that the speech act entails an act of responsibility. Elevating an issue to the level of a security issue is the result of a choice, a choice made especially by the leaders practicing the speech act. Thus, while the ‘widening’ also opens up for non-state actors, the constructivist analyst must restrain from pursuing a path of radical individualism upon which the security move would lose itself. This explains the Copenhagen School’s focus on collectivities. The security move remains socially constructed but must be legitimised (in various ways) beyond the mere individual recognition and thereby reinforced: 42 A Critical Region Building Approach “the issue is securitized only if and when the audience accepts it as such” (Buzan et al. 1998, 25). This also implies that the legitimisation of the security move and its potentials to become effective securitisation do not merely depend upon the securitising leaders or upon the audience. The effectiveness of the securitisation depends upon a sort of ‘balance’ between the securitising actor and the referent object. This tells about the extent to which the ‘widening’ can take place, i.e. the “reasonableness of a securitization” (Buzan et al. 1998, 30), i.e. whether securitisation actors (securitisers and securitised) will be able/willing/interested to respond to it. Against this background, one ought to consider that, in principle, securitisation is to be seen as a ‘negative’ rather than a ‘positive’ development inasmuch as it implies that political issues are elevated to the level of security threats. This means that the objective of a polity in the speech act approach would be, on the contrary, to desecuritise an issue, i.e., to return it to the sphere of ordinary politicisation (see Wæver 1995). This is what (ideally) takes place in a security community sort of model, to which I will return below. On the other hand, such a claim raises the question as to what should be preferable in empirical terms. Securitisation should be regarded as a negative development; however, transforming an issue into a security issue also means that society can respond to threats; urgency can trigger public attention, mobilisation, and ultimately action (Buzan et. al 1998, 29). This possible disagreement as concerns the finalité of security takes me to one last conceptual aspect of the speech act approach: the distinction between the actor and analyst (whose more epistemological aspects will be dealt with in the methodological chapter below). This is yet another point of difference between the CSS and the Copenhagen School approach, and it is central in view of my framework. The analyst in the speech act approach cannot make a definite assessment of the security move. The analyst employs his/her own criteria to observe reality, thereby constructing reality in his/her own way (which, by the way, can mean that an issue is not necessarily perceived as security). More importantly, Wæver argues, the analytical operation carried out by the analyst should not be confused with actors’ discourses on the security move and with their actions: “Actors who securitize do not necessarily say ‘security’, nor does their use of the term security necessarily always constitute a security act” (Buzan et al. 43 A Critical Region Building Approach 1998, 33). This is to say that analysts decode and interpret the security move performed by actors. Thus, actors act while analysts observe and judge the security move. Moreover, given the constructivist nature of the security move, “no one conclusively ‘holds’ the power of securitization” (Buzan et al. 1998, 31). The speech act approach can provide an indication concerning who is more likely to securitise and about possible ‘facilitating conditions’1. The main point, however, is that the focus of the analysis of security interdependence should not be limited to the performer of the security act, but on the practice of (de)securitisation itself. In relation to the discussion at hand, this aspect returns to the important differences between the individualism of CSS and the focus on collectivities of the Copenhagen School. CSS, by not distinguishing actors from analysts, fails to underline the importance of the work of evaluating whether a securitising move is ‘appropriate’ or not2. I shall now discuss the choice and determination of sectors and levels of analysis. This represents the outcome of the Copenhagen School where the influence of the Barry Buzan’s earlier writings is stronger. The discussion concerning the subdivision into sectors and levels of analysis is introduced in his path-breaking People, States and Fear (1983; 1991). The following discussion begins with sectors and then continues with the levels of analysis in order to introduce regional security. In the first edition of the work, security analysis is divided into four sectors: military, political, economic and ecological (Buzan 1983, 7583). This opens up for an interesting characterisation of security as economic and environmental, which supplements the traditionalist formulation of military and political security. While the discussion concerning the characterisation of each sector will be briefly sketched in the following, and to some extent in Chapter 7 when applied to the case of the Baltic region, here it is important to make two observations. The first observation is that the very choice and characterisation of the security sectors, especially when it comes to the non-military fields, may vary substantially; Buzan’s categorisation merely represents a single possibility. Holger Moroff, for one, notes: “After all, nonmilitary security is a negative definition and could theoretically encompass everything in the realm of internal state security” (Moroff 44 A Critical Region Building Approach 2002, 26). A UNDP report published in 1994 identifies non-military security in the “destruction of environment, scarcity of resources, population growth, poverty, lack of rule of law.” Bertel Heurlin connects it to nationalism, ethnicity, human rights and international crime and migration patterns (quoted in Moroff, ibidem). Secondly, Buzan’s own conceptual development between the first and the second edition of People, States and Fear should not be disregarded. The influence of the early works on the speech-act approach begins to emerge in the 1991 edition, wherein the defining importance of the end of the Cold War is also stressed. With respect to Buzan’s definition of sectors, this means that the concept of societal security adds up to the other four. The formulation of military and political security remains anchored rather firmly in the realist/traditionalist paradigm. The focus remains on the state as the primary securitising actor and reference object in these two sectors. That which also remains central is the rather objectivist way of identifying military and political threats: “Generally speaking, military security concerns the two-level interplay of the armed offensive and defensive capabilities of states, and states’ perceptions of each other’s intentions. Political security concerns the organisational stability of states, systems of government and the ideologies that give them legitimacy” (Buzan 1991, 19). The state also remains at the centre of this discussion in more recent formulations (e.g. Buzan 1997, 6-7; Buzan et al. 1998, 52-57 and 145-150), though leaving greater room for possible reconstructions of securitisers and referent objects. As far as non-military sectors are concerned, I will focus directly on how they have been defined by the Copenhagen School. The major difference in relation to Buzan’s original articulation of them3 is that the ‘widening’ operation stresses on inclusiveness of actors other than the state. The formulation of economic security derives from the progressive ‘triadisation’ of the world market, i.e. the passage from the dominance of the US economy after the end of the Second World War to the progressive recovery and emergence of Japan and Europe as powerful and competitive trading players. The boom of economic security formulations following the Cold War is to be attributed precisely to the fact that the economic rivalry between East and West was 45 A Critical Region Building Approach overshadowed by military and political considerations. Having said this, the notion of economic security is in many ways controversial. Given the importance of the notion of risks in liberal economies, it is to be assumed that securitisers and securitised units ‘should’ feel insecure in order to pursue an efficient level of economic interdependence. However, economic security may apply to the need or the attempt to not “generate more losers than winners” (Buzan 1997, 8) in the economic system. Environmental security, which is of particular importance because it constitutes the case study of this dissertation, focuses on the protection of issues as diverse as climate, nature, resources and biodiversity. The broadening of actors involved in this security practice is somewhat more obvious. The referent object of this securitising move can actually shift considerably. The environment can be perceived as a threat to national security. This was somewhat the case in the Soviet bloc during the late Cold War period, when the environment was articulated as a security problem in order to distract public attention from the military and the political fields (see Chapter 7). More obviously, the natural environment can also be considered the very object at risk of survival. Last but not least, humans can also be perceived as the referent objects of this securitising move, since mankind as such can be jeopardised in the long run by environmental changes and disasters. Societal security is a notion that touches upon identity. The need of the people to secure their language, religion, tradition, i.e. the very core of a ‘nation’ and its peculiar identity, is at the heart of this security move. In this respect, it ought to be distinguished from a notion such as ‘social security’. The difference is precisely in terms of the identity component rendering the securitisation of the societal an international security issue, as opposed to a securitisation of the social (e.g. welfare, health etc.), which more typically has to do with the domestic dimension. The notion of societal security therefore confirms the focus of the Copenhagen School on collectivities rather than on individuals and sparked an interesting debate among scholars regarding the nature of the societal sector (see Mc Sweeney 1996, 1998; Williams 1998; Buzan and Wæver 1997). Sectors, regardless the criterion by which they are subdivided, are to be considered analytical devices, i.e. an analytical tool. Once an 46 A Critical Region Building Approach analyst has been examining the security interdependence in a given milieu and focusing on certain sectors, the objective must necessarily become that of providing a more comprehensive picture, i.e. synthesising the sectors. In order to provide a sufficiently homogeneous overview of the security interdependence, an analyst must then overcome analytical divisions: “items identified by sectors lack the quality of independent existence […] Sectors might identify distinctive patterns, but they remain inseparable parts of the complex wholes […] The analytical method of sectors thus starts with disaggregation, but must end with reassembly” (Buzan et al 1998, 8). Given the constructivist inspiration of the Copenhagen School model, work with reassembling the sectors might produce a result in which sectors do not ‘line up’. When analysing security interdependence in the various sectors, it may well then occur that while relevant military security interdependence affects the state level more substantially, the environmental one may take place most significantly at the regional or global levels. For the purposes of this overview, this note on the fact that the different sectors may not ‘line up’ leads the discussion towards one last crucial element of the Copenhagen School: levels of analysis, particularly as regards the regional level. 2: Regional Security Levels of analysis are defined as “objects for analysis that are defined by a range of spatial scales […] where both outcomes and sources of explanation can be located” (Buzan et al. 1998, 5). From this basic definition, two considerations follow: first, the definition confirms the constructivist orientation of the Copenhagen framework. Levels are not intrinsically defined and do not refer to specific theories. On the other hand, one ought to point out that the discussion regarding levels of analysis is usually associated to neo-realism, since “levels of analysis scheme reinforce state-centric thinking” (Buzan et al. 1998, 7). This is instead in contrast with the Copenhagen School. The tension between these two components could be clearly spotted in the earlier formulations of Buzan’s approach to security. Wæver notes how this problem was evident in the first edition of Buzan’s People, States and Fear where 47 A Critical Region Building Approach arguments, the latter in particular, hint at the ethical/methodological question and at the type of constructivist epistemology envisaged by the Copenhagen School, to which I will return in the next chapter. The second point of relevance of the Copenhagen School framework is the amity/enmity pattern. The spectrum of the amity/enmity pattern was already shaped within the framework of Buzan’s classical reflections. This is where the Copenhagen framework proves to constitute a comprehensive tool of analysis in which the various theoretical traditions of international relations can be included. The security spectrum, which the proponents of the Copenhagen School have also referred to as ‘variations’ (Buzan and Wæver 2003, 82), can in fact contain the different interpretations of reality in international relations. Amity and enmity are notions that epitomise IR’s theoretical spectrum. The utmost expression of enmity is incarnated by those traditionalist approaches viewing international relations as conflictual and confrontational, centred on the states and characterised by a narrow understanding of security. On the other side of the spectrum, the ultimate form of amity pattern is security community. In between these two extremes, there are all possible combinations of security interdependence, These can be labelled security regimes (ibid. and Buzan 1991, 218), which are defined as “those principles, rules and norms that permit nations to be restrained in their behaviour in the belief that others will reciprocate” (Jervis 1983, 173). Within this rather loose category, one can place those liberal theories focusing on cooperation, the widening of the security agenda to non-military issues, and that are more or less markedly state-centric, although attributing importance to interest formation and changes7. Figure 8: The complete spectrum of the security analysis Confrontation Security Regime Security Community Enmity Amity Buzan and Wæver argued that the three stages of confrontation, security regime, and security community can be placed along the same lines as Alexander Wendt’s (Wendt 1999, 259-308) distinction of three philosophical traditions: Hobbesian, and Lockean and Kantian (Buzan 54 A Critical Region Building Approach and Wæver 2003, 82). Thus, the notions deriving from realism and Hobbes’ Leviathan refer to ‘conflict formation’; Locke’s liberalism portrays different stages of a security regime8; and idealism and Kant’s perpetual peace reflects a security community pattern. Moreover, one could argue it is within this framework that the (de)securitisation of the speech act approach applies. In fact, depending on where within the spectrum a certain security interdependence is placed, the securitising actors either escalate a given issue to the level of ‘panic politics’, or choose to bring it back (or to maintain it) to the business-as-usual level of the politicisation. In other words, by making that choice, the securitising actor moves the pendulum of the status quo of security interdependence, either towards a conflict formation pattern (in the case of extreme securitisation) or towards a security community scenario (as the ultimate desecuritisation). This suggests that the amity/enmity spectrum can also be clearly connected to the constructivist inspiration of the Copenhagen School. And in this respect, as I am about to discuss, it constitutes an aspect of central relevance in view of its correlation to regionalism. 4: The Region Building Approach On the basis of this overview of the conceptual and analytical aspects of security, I can now introduce the core of the classical RegionBuilding Approach (RBA), which constitutes the fundamental inspiration to bring the two overviews on regionalism and security together and to understand the relationship between them, which is at the centre of the first research question of this study. Some actually regard this RBA as an actual ‘school’ (Wæver 1997, 293), and Iver B. Neumann (1992, 1994) is the one member of this ‘school’ who elaborated somewhat more systematically on the theoretical and conceptual aspects of the classical RBA. The RBA focuses on the genealogy of regions. According to Neumann, the keyword of the approach is ‘whose region’ (Neumann 1994, 57). This means that the RBA investigates the origins of regions, where they ontologically come from, and attempts to understand the meaning of the region that is talked about. The RBA draws consistently from the literature on nation-building, since the nation state “simulates the archetypal principle for political organization” (ibid. 58), and on the genealogy of nations as ‘imagined 55 A Critical Region Building Approach communities’ (i.e. ibid., 58 and Anderson 1983). Neumann argues that while the spatiality of nations is “beyond doubt” (ibid.), its temporal existence is more controversial, since it is constructed upon historical materials of various kinds, which the political actors are responsible for evaluating and tailoring to the nation-building project. His argument is that regions can also represent imagined communities of such kind, and their region builders are also capable of creating a regional history and mythology. Certainly, Neumann continues, the traditional linkage between nation-building and sovereignty cannot apply to regionbuilding, because regionalism does not deal with power in a traditional sense. On the other hand, region builders are not interested in regionalism as an objective, because they do not necessarily aim to establish a political entity that would transform regionalism into a variation of nationalism. Quite simply, region building is a practice, a platform, a meeting place. It is based on participation and it is constantly in fieri. In these features, ‘genealogical’ approaches of nation building can apply also to regions. As a result, regions are essentially a product of political actors and are created by actor-generated discourses on the region. As for the discourse on security, “regions are defined in term of speech acts” (Neumann 1994, 59), which confirms the radically constructivist nature of the approach. A region exists and enhances itself via the discourses of actors relating to it. In a sense, the genealogy of a certain region can be traced in language practices as much as in the political act itself. One note should be devoted to recalling the review of regionalist approaches carried out in Chapter 1. Neumann observes that the RBA is not to be perceived as a theory as such. As he put it: “the region building approach is not offered as an attempt to place the study of regions in international relations on a new footing. […] Rather, it is a tool with which to dot the margin of the ongoing debate” (Neumann 1992, 6). The RBA does not reject the different approaches to regionalism. Actually, prior to introducing the constructivist core of the argument, Neumann discusses the different approaches to regionalism. As discussed in Chapter 1, the numerous approaches to regionalism can be rearticulated along a ‘continuum of approaches’, which I took the liberty to categorise in terms of ‘old’ and ‘new’ regionalisms. The point here is that the RBA is not designed to replace these theories but to actually go beyond (or beneath) them. The various regional 56 A Critical Region Building Approach approaches can be “rich in insights and diverse in assumptions” (Neumann 1994, 57), but they all neglect one crucial point: they all assume that the region, exists and possesses a range of specific features, which obviously vary depending on the different theoretical inspiration of each approach. Conversely, the constructivist inspiration of the RBA rests precisely on the fact that the origins of the region are to be created and unfolded, which is something that does not exist a priori. It is within this context that Buzan’s classical RSCT, and thereby regional security, is referred to. According to Neumann, the classical RSCT is the “most useful” (Neumann 1992, 12) of all approaches, since it proposes a “combination” (Neumann 1992, 13 emphasis added) of inside-out and outside-in approaches. Figure 9: Continuum of regionalist approaches ‘combined’ by the classical RSCT Enmity Amity Confrontation Security community ‘Old’ Regionalisms ‘New’ Regionalisms (Outside-in) (Inside-out) My rearticulation of the inside-out/outside-in continuum in terms of a ‘new’/‘old’ continuum facilitates a clearer understanding of the ontological foundations of this claim. Buzan’s amity/enmity spectrum of regional security is, in fact, laid out according to different theoretical orientations, as is the discussion concerning ‘old’ and ‘new’ regionalisms. With respect to regionalisms, I noted that traditionalist approaches view regions as an outcome of the anarchical international arena. It was observed that the leftmost end of the regionalist continuum can be studied as alliance-building, and that one of the principal reasons why states choose to pursue regional cooperation is to relate to or contrast with a hegemonic power. Alternatively, actors may choose to pursue functional cooperation in order to maximise their gains. Hence, security interdependence resting on the enmity side of the security continuum shares the theoretical pillars of traditionalist IR paradigms with ‘old’ regionalism. Likewise, ‘new’ regionalism was 57 A Critical Region Building Approach defined as a result of globalist dynamics on the one hand, and cohesion, awareness, and identity on the other. This is indicative of an idealist framework of analysis that could be perceived as common with the amity side of the continuum. Thus, according to the classical RBA, the amity/enmity spectrum introduced by the classical RSCT is the most useful of all frameworks, because it combines approaches to regionalism. In view of the founding constructivist claim of the RBA, however, Buzan’s RSCT is introduced by Neumann not only because of its usefulness, but most importantly because it exemplifies the major “oversight” (Neumann 1994, 57) of all regionalist approaches. Neumann argued, “all the authors concerned seem to assume that there exists a neutral and analytical ground, a spot above the regional fray to which the sovereign author can retreat. They never acknowledge the inevitable political dimension of their analytical endeavours.” (Neumann 1992, 5) This is to say that the classical RSCT is comprehensive, but it is also the one approach that best epitomises the most important omission of all regional theories, i.e. the assumption that regions exist a priori. Classical RSCT “makes an assumption about what a region is” (Neumann 1994, 58) or, even worse, transforms the ontological matter of the region’s existence into a “technical question” (Neumann 1994, 57). This is where RBA’s constructivist epistemology intervenes by posing questions such as: “who draws the line between the inside and outside? Who takes it upon themselves to include and exclude, with what intentions, and with what consequences?” (Neumann 1994, 58). These questions can be answered by examining the construction of the actors. On the basis of this claim, I am now able to put forward my three critical arguments. 5: Three Critical Arguments The classical RBA remains the predominant analytical reference of this study, mainly because of the comprehensiveness of its argument; nevertheless, it is also inspirational on the grounds of this reference to the security debate sketched immediately above – and in particular the security ‘spectrum’ of the classical RSCT. This is a slightly more 58 A Critical Region Building Approach indirect argument, since Neumann introduces this reference in order to criticise it. It is the one cue, however, that assists in the framing of the main argument of the critical RBA regarding the correlation between security and regionalism. My First Argument is based on the fact that the radically constructivist core of the critical RSCT put forward by the ‘Copenhagen School’ can assist in overcoming Neumann’s legitimate (from a constructivist point of view) critique of Buzan’s classical RSCT. The critical RSCT can explain how the ‘useful’ scheme put forward by Buzan can be freed of the charges of ‘oversight’ raised by the RBA, i.e. defining what a region is a priori. While providing possible response to Neumann’s critique, however, the critical RSCT will itself bring a critique to the classical RBA, in relation to how the use of RSCT is framed. After this framework is laid out, I will put forward my Second Argument, in which the concept of ‘regionality’ is introduced in order to explain this correlation between security and regionalism in the case of ‘security community’ and thereby to underline the non-exclusive nature of this correlation. Finally, I must introduce a Third Argument, which relates to the applicability of the constructivist features of the RBA, vis-à-vis both the conceptual and territorial characterisation of the regional security complexes. First Argument: Constructing a Region and Constructing Regional Security The radical constructivist core of the critical RSCT is suitable to overcoming some crucial objections raised by the classical RBA. The securitisation approach, which is at the core of the critical RSCT, shares the importance of discourses in understanding reality with the RBA and, therefore, the methodology of the ‘speech act’. Even more importantly, the critical RSCT, like the RBA, also stresses the crucial distinction between analyst and actor. Construction of the reality and analysis of reality are two different operations, and both the RBA and the critical RSCT clearly draw a line between the two. Hence, the constructivist inspiration of the critical RSCT can well fill the gap between Neumann’s criticism of the classical RSCT, and his judgement that Buzan’s framework is in fact ‘useful’ because it combines different regionalist approaches9. Both the RBA and the critical RSCT share a constructivist perception of reality; no claims are 59 A Critical Region Building Approach therefore made regarding the existence of the region a priori. Once this main epistemological ‘oversight’ of the classical RSCT is overcome, the analyst can explain the speech acts of the actors and interpret them along the spectrum/continuum of approaches. Nonetheless, this argument also introduces critique of the classical RBA, which is what characterises my First Argument. What is not specified when referring to Buzan’s security complex theory is that that framework is devoted to the analysis of security interdependence, and it is not necessarily concerned with, and interested in, the regionalist outcomes, i.e. the nature of the regions the actors are constructing. Admittedly, Neumann does note that “the construct [i.e. the RSC] does not assert its authority as an ‘imagined community’, a cognitive construct shared by persons in the region themselves” (Neumann 1994, 57), and that “the delineation of regions as security complexes is a technical question” (ibid.). Nevertheless, the ontological difference between the analysis of regional security and of regionalism is not problematised to a great degree in the classical RBA. In other words, Buzan’s RSCT becomes one of the regionalist approaches and notably the most useful of them. Regional security analysis becomes a framework of regionalism. To the contrary, notwithstanding that both the RBA and the critical RSCT are radically constructivist and focus on discursive practices, the RBA and RSCT perform two different operations. The RBA focuses on the region-building speech act, and critical RSCT focuses on the security speech act. Hence, even though the RSCT is the most ‘useful’ in terms of introducing the concept of a ‘spectrum’ combining multiple theoretical approaches, the two different focuses, security and regionalist, cannot be confused and ought to remain separated. The critical RSCT does not ignore this distinction (see in particular Buzan et al. 1998, 34 and 46) and explicitly focuses on the former10: “We are not (in the context) interested in whether the actors define themselves as a region or whether they claim that their true region is something larger or smaller. Security complexes do not require that their members think in terms of the concept security complex […]. Analysts apply the term security complex (and therefore designate a region) based upon the contingent, historically specific, and possibly 60 A Critical Region Building Approach changing constellation generated by the interdependent security practices of the actors.” (Buzan et al. 1998, 20) The quote suggests that the security analyst limits his/herself to providing the security move with a constructivist character and to observing how it plays out through the lenses of the speech act approach. The security analyst does not necessarily consider the regionalist aspect, which is at the centre of the RBA. The point here is that the critical RBA assumes its name from the fact that it refers to the critical RSCT elaborated by the Copenhagen School rather than Buzan’s classical version. This implies that the critical RBA shares the radically constructivist perception of the regional reality with Neumann’s classical version, which also constitutes the heart of the critical RSCT. However, the critical RBA rejects the argument that the RSCT, even in the Copenhagen School’s constructivist version, aims to ‘combine’ different regionalist approaches, for the simple reason that the RSCT and the RBA are created for two different analytical operations. These observations clarify my First Argument concerning the correlation between (the construction of) regional security and (the construction of) regionalism. The critical RBA retains the spectrum of analysis proposed by the RSCT to describe different forms of regional security interdependence – spanning from confrontation to security community. It also retains the regionalist ‘continuum’ – ‘old’ and ‘new’ regionalisms – inspired by Neumann. While, in principle, these two frameworks should not merely be combined, and thus confused, the critical RBA argues that, under one crucial condition, the two frameworks can be correlated and that, in particular, the analytical construction of regional security is capable of indicating the nature of regionalism created by actors. Allow me to first explain how this ‘correlation’ is supposed to work, after which I will explain what this ‘one crucial condition’ under which the correlation is supposed to work should be. The very idea that the radically constructivist analysis of regional security interdependence aims to analyse power relationships within the regional space suggests that the result of this analysis is a state-ofthe-art of the type of cooperation or conflict (or anything in between) that is taking place in that region; and thus the type of regionalism. The 61 A Critical Region Building Approach two continuums discussed within the two constructivist frameworks of the classical RBA and the critical RSCT both span from traditionalist approaches to idealist ones. They both distinguish state-centrism from multiple inclusiveness; conflict from cooperation; ‘outside-in’ approaches from ‘inside-out’ approaches. They both encompass the entire spectrum/continuum of IR. This correlation is not exclusive, which is an aspect that will be dealt with in the next section. The argument here is that the construction of regional security can explain what kind of outcome is generated in terms of regionalism. One may, for instance, assume (as in the leftmost end of the figure below) that regional security interdependence is constructed in a traditionalist manner, e.g. discourses are state-centred, security is narrowly defined and the focus is on the balance of power. Then, according to the critical RBA, these discourses can be correlated to a construction of the region by actors which is akin to ‘old’ approaches, i.e. alliance building, functionalist cooperation. Conversely, the more security discourses move across the security continuum to the right, the more the regional security interdependence will be about cooperation, widening its agenda, and possibly including non-state actors. These features can correspond to a construction of the region by actors that is characterised by identity building and social inclusiveness, i.e. a ‘new’ regionalist pattern. Figure 10: The Critical Region Building Approach Security Analysis Regionalisms ’Old’ Regionalisms ‘New’ Regionalisms Confrontation Security Regime Security Community Re g ion In other words, my critical RBA shares Neumann’s positive judgement of the usefulness and comprehensiveness of the RSCT. However, it revises the relationship between regional security and 62 A Critical Region Building Approach regionalism. Following the Copenhagen School, it argues that regionalism and regional security are two distinct operations. On the other hand, it retains the argument that both analyses are ontologically alike in that they unfold along the ‘continuum’ spanning from traditionalist/confrontational approaches to idealist/reflectivist ones. This ontological analogy inspires the correlation of the two continuums. Nevertheless, this correlation is only possible if one specific condition is satisfied. The argument about the separation of the two analyses was based on the fact that the Copenhagen School is not really ‘interested’ in the regionalist outcome. Furthermore, I noted that this non-interest owes to the circumstance that actors may or may not ‘define themselves as a region’ in the security analysis. This constitutes the core of the correlation between the two analytical frameworks. The condition rendering it possible for the analysis of security interdependence to explain regionalist outcome is that actors perceive themselves as being part of a region. The awareness of the actor is the crucial factor for explaining how the correlation between the Copenhagen School and the RBA unfolds. The analytical prism of security can explain what happens in terms of the construction of a region (or lack thereof) if the actors constructed by the analyst as a regional security complex are aware of existing as a (possible) region. This is made explicit by the latest formulation of the RSCT: “The existence of an RSC is not in terms of the discursive ‘construction of regions’ […] RSC is an analytical concept defined and applied by us, but these regions (RSCs) are socially constructed in the sense that they are contingent on the security practice of the actors. […] We study the security discourses and security practices of actors, not primarily their regional(ist) discourse and practices. The latter is an interesting an important question […] and is an element of our analysis but not the basis of it.” (Buzan and Wæver 2003, 75 emphasis added) Hence, this one element of the security analysis referred to by Buzan and Wæver is precisely that which the critical RBA focuses on: region building constitutes an element of the security analysis on the condition that the security analyst selects conscious actors, i.e. actors 63 A Critical Region Building Approach This is because, according to the Copenhagen School, this functional delimitation of the territory of a security complex indicates whether certain type of security interdependence in one specific spatial focus are ‘relevant’ or not. If not, the analysis will draw its conclusions regarding the degree and nature of the cooperation in a certain area based on the fact that that a certain degree of spatial focus is not ‘relevant’ in terms of regional security interdependence. However, in order to refine their functional delimitation of the territory of security regions, Buzan and Wæver introduce one further spatial category, which is particularly important for the purpose of this study: “Since the earliest development of the RSCT we have also allowed the idea of subcomplexes as a ‘half-level’ within the regional one, and we stick with that here. Subcomplexes have essentially the same definition as RSCs, the difference being that a subcomplex is firmly embedded within a larger RSC. Subcomplexes represent distinctive patterns of security interdependence that are nonetheless caught up in wider pattern that defines the RSC as a whole […] Subcomplexes are not a necessary feature of RSCs, but they are not uncommon either, especially where the number of states in an RSC is relatively large.” (Buzan and Wæver 2003, 79-80) This reference to sub-complexes refines the functional delimitation of the regional territory in the critical RBA. Here I am working with both complexes and sub-complexes. It then follows that the linkage between regional security and regionalism can also be refined territorially. Regional security complexes refer to that which in terms of regional cooperation are defined as macro-regions. The security terminology of sub-complexes can instead refer to that which I have defined as meso-regions in terms of regional cooperation. Indeed, talking of ‘regions’ instead of macro-regions when referring to a complex, i.e. the EU, means to assume that the complex level is by definition in a position of superiority, and that all of the other levels included within it are in a subordinate position and can thus be defined as sub-regions, sub-sub-regions and so forth (cf. Møller 2000, 48-49). In other words, the (macro-)region is the analytical level that defines the others, and the different subor trans-regional dimensions are 70 A Critical Region Building Approach automatically tailored to it. This ‘submission’ may well prove correct in the empirical analysis, in which case one will indeed talk of a subor trans-regions, but the answer must nevertheless be found in reality. Instead, providing such definitions a priori while constructing a framework for application to the case of the Baltic Sea Region (BSR) would result in the establishment of a sort of hierarchy in which the identification of security interdependence at the BSR level is perceived to be ‘inferior’ to the EU (macro-) regional security level, which is the region. The prefixes mesoand macroare more neutral inasmuch as they have the advantage of only defining the territorial dimension of a region while allowing for the subjectivity of one regional phenomenon to unfold regardless of whether a region is territorially smaller, and possibly contained within a larger one. Figure 12: Complexes/sub-complexes; macro-regions/meso-regions. Regional Security Complex/macro-region Subcomplex/ Meso-region The macroand mesolabels could also apply to the security implications, and thus the complex/sub-complex distinction; however, I prefer not also modifying this latter distinction. The reason for not doing so is that these semantic differentiations ought not to draw me too far away from the mainstream terminology on international security, which does indeed treat macro-regions as ‘regions’. Moreover, maintaining the differentiation of these categorisations (macro/meso vs. complex/subcomplex) can be viewed as a sort of ‘reminder’ concerning the partition between the security analysis and regional cooperation. The Copenhagen School does not preclude a more territorially limited interdependence being constructed as a complex17, or be on its way to becoming one – in which case one would talk of a ‘pre-’ or ‘proto-complex’ (see Buzan and Wæver 2003, 71 A Critical Region Building Approach 96). For my purposes, however, the notion of sub-complex serves as a ‘reminder’ that while a security analyst can view a certain region as part of the larger complex, actors in that region can continue to perceive the region as having a (‘meso’)-regional subjectivity of its own. This quarrel concerning the labelling of the regional levels leads to a couple of more substantial observations on the relationship between complexes and sub-complexes. The existence, formation and geographical positioning of a subcomplex within the security complex depend upon the features and evolution of both the sub-complex and of the complex itself; they can therefore vary considerably. Likewise, the role that sub-complexes are capable of playing within the complex may vary, depending on the type of security interdependence characterising the two formations. The sub-complex will constitute a source of concern and instability for the complex if its security interdependence is characterised by conflict, e.g. the Balkans in Europe; however, it can well constitute a spot characterised by cooperation and even ‘regionality’. In this case, discourses at the complex level can refer to it more positively, e.g. as a contribution or even a model for the overall security of the complex, e.g. Scandinavia. The case of sub-complexes vis-à-vis ‘centred’ security complexes is particularly relevant for the purpose of this empirical study on the Baltic Sea region. The structure of a ‘centred’ security complex is associated by the Copenhagen School with the notion of a ‘gradated empire’, which follows from Adam Watson’s reflections on international society (Watson 1992). International society, according to Watson, is not based solely upon states; rather, it includes other and not necessarily state-based political units in the formation and development of a sovereign system. Hence, ‘supernational’ systems are those characterised by the monolithic dominion of the states. ‘International’ systems are characterised by a greater degree of differentiation and by varying degrees of autonomy, which are best captured by the image of ‘concentric circles’, as in the figure below. In such a pattern, those actors which are located closer to the centre of the system fall under the ‘direct rule’ of the international system, which 72 A Critical Region Building Approach weakens the further actors are placed from the centre. Hence the denomination: ‘gradated empires’. This construction has been resumed by Wæver, and has been applied in particular to Europe (Wæver 1996, 223-226). For the purpose of this theoretical discussion, it can be argued that Watson’s argument is applicable to different aspects of the RSCT. Firstly, it gave birth to the notion of ‘centred complexes’, which, as opposed to the ‘standard’ and anarchic type of complex, share with Watson a certain (uni-) polarisation of power within the complex (see Buzan and Wæver 2003, 84-89). Secondly, Watson’s arguments fit nicely with the passage from the traditionalist and state-based classical RSCT to its critical and ‘widened’ version, because of his focus on international ‘society’ rather than merely upon states. More interestingly for the purpose of this discussion, however, Watson’s construction can suggest something vis-à-vis the relationship between ‘centred’ complexes, such as the EU or Russia (ibid.), and sub-complexes. Since the argument is that the ‘direct rule’ of the centre weakens proportionate to the distance between the actors and the centre, the formation of ‘autonomies’ within the complex is more likely to take shape at the periphery of this imperial structuration of the complex. In other words, the formation of sub-complexes can be regarded as one possible form of ‘autonomy’ within the complex, one in which security interdependence develops differently than at the centre of the complex, but nevertheless remains embedded within it. Figure 13: Sub-complexes within a ‘centred’ security complex Source: adapted from Wæver 1996, 225 and 233. Complex/macro-region Sub-complex /meso-re g ions 73 A Critical Region Building Approach From this follows that it is more likely for security interdependence at the meso-regional/sub-complex level to take shape at the fringes of a centred complex, because the ‘direct rule’ of the complex is weaker there; however, this claim requires further specification. The point on peripheralisation applies more suitably to the case that security interdependence at the meso-regional level is on the enmity side of the analytical spectrum. This is because a situation of confrontation at the core of the complex may be too destabilising for the very survival of the complex (imagine transposing the security interdependence characterising the Balkans and the Caucasus, in Benelux or the Moscow area). Conversely, a meso-regional/sub-complex interdependence characterised by amity may well be less peripherialised, provided it remains at the level of ‘autonomy’, i.e. firmly embedded within the complex, and does not form a sort of ‘independence’. The second argument deriving from Watson’s imperial formulation is that this peripheralisation of sub-complex formation can play a role in the inter-complex relations. In the event that two or more adjacent security complexes are similarly characterised by such ‘centred’ power structuration, then the peripheries of the ‘empires’ involved will be characterised by security interdependence that is less firmly tied to the centre and thus more difficult to define. Here the formation of subcomplexes may play an interesting role at the complex level. Figure 14: Sub-complexes in the inter-complex relations Complex 2 Complex 1 Sub complex C Sub complex A Sub complex B Source: adapted from Wæver 1996, 225 and 233. 74 A Critical Region Building Approach Once again, the consequences of sub-complex formation at this ‘double’18 periphery for the complexes/macro-regions will depend upon the type of security interdependence characterising the subcomplex. A sub-complex that is a conflict zone is potentially capable of exacerbating the relationship at the macro-regional/complex levels, especially in a case in which the two adjacent complexes/macroregions have contrasting interests in it (see ‘sub-complex A’ in the figure above). The disagreement between the EU and Russia in the 1990s over intervention in the Balkans conflict can again provide a good example. Alternatively, enmity at the sub-complex level can foster cooperation between the two complexes/macro-regions. Here the security interdependence in two complexes can converge in the subcomplex (see ‘sub-complex B’ in the figure above). Within this framework, relevant actors at the complex level will practice the security speech act so that cooperation and coordination of action between the two complexes is encouraged, and the two complexes intercede to favour amity and possibly the formation of a security regime at the sub-complex level. Thirdly, if the sub-complex at this double periphery is characterised by mature cooperation or even regionality, it can constitute a forerunner, i.e. a model for enhancing peaceful interdependence between the two complexes (see ‘sub-complex C’ in the figure above). In this framework, the sub-complex will practically function as borderbreaking element. It can even constitute a sort of hook for the intercomplex relations, possibly in view of a partial fusion of the security interdependence of the complexes involved, i.e. the formation of one “loose supercomplex” (Wæver and Buzan 2003, 431)19. 6: Summing Up In conclusion, then, the three arguments characterising the critical RBA can be summarised as follows: 1) The continuum of security interdependence can be correlated to the continuum of regional approaches and, therefore, the security analysis can provide an explanation for regional cooperation. This is because regional cooperation is considered one of the elements of the security analysis; however, this correlation is 75 A Critical Region Building Approach possible only in the event that the security analyst selects relevant actors for the security analysis not only for their leadership, but also on the basis of their awareness of being part of a region. 2) This correlation between regional security analysis and regionalism is not exclusive. This can be best understood by focusing on the most advanced stage of amity, which is security community. Security community corresponds to total desecuritisation. Hence, it refers to a status quo in which security is no longer an argument for the regional actors. This has been associated with the concept of regionality in regionalist terms. Regionality is a form of cooperation characterized by factors other than security, of a cultural, social, economic or otherwise nature, and can thus exemplify how security defines regionalism in a non-exclusive manner. Furthermore, one can also derive from this correlation between security community and regionality that the more the security speech act is securitized, the more the status quo wanders from security community/regionality, the more security comes to define the regional discourse. 3) The territorial construction of a security region is functional in terms of the security interdependence in the critical RSCT, and so is the construction of a region in the critical RBA. This implies that the region cannot be only socially constructed (as in the classical RBA), but must instead correspond to the existence of security interdependence. From this follows a more specific categorisation of the regional territory. It is argued that security complexes generally correspond to that which in regionalist terms are defined as macro-regions. Security sub-complexes correspond territorially to meso-regions. Particularly relevant for the purpose of this study on the Baltic region is the relationship between subcomplexes/meso-regions and ‘centred’ security complexes/ macro-regions (such as the EU and Russia). Two observations have been elaborated in this respect: firstly, the sub-complex is more likely to form at the periphery of the centred complex if its security interdependence is characterised by enmity. This is because the power of the complex at its outermost circles is weaker than at its core. This argument on peripheralisation does not necessarily apply to the case of a sub-complex characterized by amity. Secondly, the formation of a peripheral sub-complex 76 A Critical Region Building Approach can produce relevant implications at the inter-complex level in the event two ‘centred’ complexes are adjacent. In this case, the sub-complex may either aggravate the inter-complex relations, or may instead improve them, even becoming a sort of hook for deeper inter-complex relations. 77 A Critical Region Building Approach Notes: 1 These are the conditions under which, given the constructed nature of security, the ‘speech act’ works. These may be internal (‘the grammar of security’) or external (the ‘authority of the securitising actor’ or the existence of objects that are generally referred to as threatening, e.g. tanks) (Buzan et al. 1998, 32-33). 2 As Johan Eriksson put it, for critical security students, “to observe is to advocate” (Eriksson 1999, 327, emphasis added). Thus, the analyst, by observing, advocates, i.e. makes a claim, a choice; (s)he passes a judgement. The stress is not merely on the constructivist approach of CSS, as the analyst decides what security is constituted of; the focus is also on the strong political role played by the analyst. In the case of the Copenhagen School, instead, “to observe how to advocate” is at the centre of the problem (ibid.). This statement confirms that the securitisation approach is even more constructivist than the CSS insofar as the philosophical position is concerned, while the purpose of the School is more traditionalist. The methodological/ethical implications of this claim will be dealt with in Chapter 3. 3 “Economic security concerns access to the resources, finance and markets necessary to sustain acceptable levels of welfare and state power. Societal security concerns the sustainability, within acceptable conditions for evolution, of traditional patterns of language, culture and religious and national identity and custom. Environmental security concerns the maintenance of the local and the planetary biosphere as the essential support system on which all other human enterprises depend” (Buzan 1991, 19-20). 4 According to Buzan, and later the RSCT, ‘overlay’ is defined as a condition in which “great power interests transcend mere penetration, and come to dominate a region so heavily that the local dynamics of security interdependence virtually cease to operate. It usually results in the long-term stationing of great power armed forces in the region, and in the alignment of the local states according to the patterns of great power rivalry” (Buzan and Wæver 2003, 13). 5 However, the definition of a security region should not go ‘too far’ beyond the ‘territorial contiguity’ criterion. This is an important point of difference between Morgan and Lake’s approach to RSCT on the one hand, and Buzan and Wæver’s approach on the other. For the former, regions are correlated to the global to the point that the regional level is practically incorporated, also in analytical terms, in the global (see Lake and Morgan 1997). Both the classical and the critical RSCT put forward by Buzan and Wæver retain the inextricable correlation between the global and the region, but are more keen to stress the value of security regions as analytical subsystems, both at the conceptual and territorial levels (Buzan and Wæver 2003, 116). 6 An RSC may also not be present. One of the reasons is that a state may have too few capabilities to operate beyond its domestic concerns. There would not be a trace of interdependence in such a scenario. Secondly and more importantly, (as opposed to the previous scenario) an external state can intervene, thereby obstructing the possibility of forming a RSC, which is where the ‘overlay’ figure applies. 78 A Critical Region Building Approach 7 Patrick Morgan has elaborated on the same concept, that of the spectrum between confrontation and security community, distinguishing five possible steps in between. The less co-operative form of RSC is that in which security is obtained “via the use of power to restrain power,” which resembles the neo-realist paradigm where power distribution and the presence of a (hegemonic) leadership is essential to guarantee security. The second level is that of the “security via a concert,” where confrontation is compensated for by the collective responsibility of the great powers. The third step is “security through multilateral collective management,” where decision-making can no more be achieved with a concert of major powers, but via consensus of all the states of a RSC. The fourth level, already noted above, is that of the “security via a Pluralistic Security Community.” The fifth level is that of security achieved via integration (cf. Morgan 1997, 33-38). 8 Buzan and Wæver note that the English School would replace Locke’s model with that of Grotius (Buzan and Wæver 2003, 82). As anticipated in the preface, this question will be dealt with more thoroughly in Chapter 7 below. 9 One could argue that at the time of the publication of the works on the RBA, only Buzan’s classical version of the RSCT was present. A few years still had to pass until the constructivist ‘updating’ of the RSCT was developed in full length. This may explain why Neumann focuses on Buzan’s classical framework rather than on the ‘critical’ – and more constructivist – framework. The following criticism of the critical RSCT to the RBA, however, can be made from the point of view of both Buzan’s classical version and the constructivist ‘critical’ one (see footnote below). 10 While, admittedly, this problem is less clearly stated in Buzan’s formulation in People States and Fear (Buzan 1991, 186 ff), and other scholars have referred to this security framework to examine regionalism (see Møller 2000, 47-48; Söderbaum 2002, 5). 11 There is a sort of ‘Freudian lapse’ made by Emanuel Adler, which I regard as quite emblematic to exemplify this aspect. In his excellent article on constructivism in International Relations (Adler 1997), he refers to Neumann’s essay on the RBA (Neumann 1994). However, instead of “A Region Building Approach to Northern Europe,” he refers to the title of the article as “A Security Community-Building Approach to Northern Europe” (in Adler 1997, 359 emphasis added). 12 Some analysts oppose this strict association between security community and community building (cf. Stadsholt 2001, 12). 13 This linkage, Adler and Barnett continue, emerges rather explicitly in Tönnies’ original differentiation between association and community, i.e.: Geimenschaft and Gesellenschaft, “whereas the former refers to an organic community involving bonds generated by tradition and culture, the latter sees society as a more contractual arrangement that emerges from self-interested behaviour” (Adler and Barnett 1998, 60). See Chapter 7. 14 After making his distinction between regionalisation (referred to here as ‘old’ regionalism) and regionalism (here ‘new’ regionalism), Wæver argued that “Regionalists are advocates of regionalism, not of regionalisation […] and probably ‘regionalist’ has an even more radical ring than regionalism” (Wæver 1997, 300 79 A Methodological Trait D’Union The Actor/Analyst Debate At the heart of both the Copenhagen School framework and the classical RBA rests the social construction of reality, which is also the case for the critical RBA. However, the construction of the two frameworks is characterised by remarkable differences, which must be spelled out in order to unveil how I intend to proceed with the critical RBA. Towards the end of the First Argument of the critical RBA, it was argued that the crucial condition for correlating the analysis of regional security and regionalism is that actors in the identified security complex possess an awareness of belonging to a region. But this, I concluded, implies a number of reflections regarding the epistemology of social construction in both the classical RBA and the Copenhagen School. I would argue that the differences in the two approaches to social constructions boil down to how the relationship between actor and analyst is conceptualised in the two frameworks. By spelling out these differences, I will be able to observe the epistemological criteria guiding the social construction of the critical RBA. In Adler’s terms, the classical RBA would be categorised within the fourth group of constructivist approaches: the genealogical and deconstructivist variety. Figure 15: The RBA constructivist approach Point of observation ’old’ regionalisms ‘new’ regionalisms The RBA refers to the construction of the region by actors. Actors may perform this operation in the most diverse ways and according to the most diverse criteria. Actors themselves may be quite different and reside in very distant places, all becoming ‘part’ of the region on account of their own political will to construct it. In other words, the role of the actors in the construction of the region is seemingly without restrictions. Within the context of this debate, this implies that actors perform the political act of regional construction and hold responsibility for it (see also Neumann 1992, 32). 86 A Methodological Trait D’Union The role of the analyst in the classical RBA is to deconstruct the construction and to uncover its deep genealogical roots. The need for an RBA was based precisely on the assumption that all approaches to regionalism “never acknowledge the inevitable political dimension of their analytical endeavours,” and that “choosing the criteria by which to define a region cannot be a politically neutral act” (Neumann 1992, 5). The RBA stands in opposition to these trends, because it allows actors to construct and reconstruct their reality and limits analysts to deconstructing this “ongoing genesis” (Neumann 1992, 6) of actors’ discourses1. This explains why, in the figure above, I have portrayed the point of observation of the social construction in the RBA as ‘from below’2. The Copenhagen School frames this epistemological debate considerably differently, and should be included in Adler’s first group of constructivisms: the modernist type. Figure 16: The Copenhagen School Constructivist Approach Point of observation Confrontation Security Regime Security Here the analyst places him/herself outside and beyond the continuum of analysis in order to determine how relevant actors talk security into existence and construct the analysis accordingly. In the Copenhagen School framework, the analyst is not as “sovereign” (Neumann 1994, 57) as was the case in Buzan’s classical objectivist framework (Buzan 1991). Reality, according to the Copenhagen School, is also socially constructed. This was the basis for carrying out the correlation between the critical RSCT and the RBA as part of the First Argument in Chapter 2 above. However, analysts unavoidably ‘subjectivise’ the security environment around them by observing it. Objectivists will argue that observation does not genuinely imply an act of emancipation by the analyst, since the security reality exists and unequivocally displays itself. In the Copenhagen School framework, on the other hand, the analyst must recognise his/her responsibility. (S)he observes the nature 87 A Methodological Trait D’Union of the security interdependence that has been constructed and interprets it, i.e. the analyst decides “how to advocate” (Eriksson 1999). This argument underlines a certain measure of ‘teleological clarity’ characterising the constructivist epistemology of the Copenhagen School. This teleological clarity is motivated by the need to minimise relativism and the fragmentation of the actual analytical outcome. I should recall an argument by Buzan from the preceding chapter in order to explain this point: “while our philosophical position is in some sense more radically constructivist in holding security to be always a political construction, not something the analyst can tell about how it ‘really’ is, we are in our purposes closer to Traditional Security Studies” (Buzan 1997, 20). Hence, ‘teleological clarity’ denotes the need to delimit the terms of social construction in order to restrain the analysis from floating indefinitely across the continuum. In this manner, which is all but immune to criticism (see McSweeney 1996; 1998), the proponents of the Copenhagen School rebut the accusation of being excessively individualistic – as the radical constructivist approach of the framework could imply – by focusing on the importance of collectivities and of the work of ‘endorsement’ of the security move by leaders, which is at the centre of the speech act approach. All of this suggests that the Copenhagen School modernist approach to Constructivism is more ‘top-down’ as regards the point of observation (figure above)3. The epistemological position of the critical RBA will stand somewhere in between that of the Copenhagen School and of the RBA. It will be closer to the Copenhagen School in terms of its modernist clarity, since the aim of the critical RBA is to explain regional cooperation through the lenses of security. Hence, it holds that the analyst bears a strong responsibility in carrying out the analysis and must be aware of his/her ethical role, since (s)he will select the relevant actors, judge their leadership, the receptiveness of a collectivity, etc. On the other hand, viewing the critical RBA as one “element” (Buzan and Wæver 2003, 75) of the RSCT means that one must only take into account the case in which actors do actually perceive themselves as being part of the region. This is the aspect of the epistemology of the critical RBA that lies closer to Neumann’s genealogical approach, since the awareness of actors is crucial to constructing the regional reality. In other words, the critical RBA takes 88 A Methodological Trait D’Union the strong ethical and political responsibility of the analyst from the Copenhagen School. From the classical RBA, it takes the focus on the subjectivity of the actor4. Figure 17: The Constructivist approach of the critical RBA Security Analysis Point of observation Regionalisms ’Old’ Regionalisms ‘New’ Regionalisms Confrontation Security Regime Security Community Re g ion From this combination of the two different epistemologies it follows that actors must be aware of the fact that their discourses have implications for the construction of the region. On the other hand, the analyst must acknowledge that (s)he is ‘advocating’ by studying the regional reality in a certain manner. Hence, (s)he will be examining the leadership of the actor and the fact that this leadership is recognised by a collectivity. On this basis, (s)he will analyse security interdependence and then derive the outcome in terms of the construction of the region. In other words, as shown in the figure above, the type of social construction envisaged in the critical RBA remains ‘from above’, as in the RSCT, due to the noteworthy responsibility of the analyst. Yet the analyst must always take the subjectivity and awareness of regional actors into account, which explains the ‘oblique’ positioning of my point of observation in the figure. In view of these reflections regarding the different epistemologies, allow me to conclude this discussion with a few ontological clarifications. The security analyst will coincide with the regionalist analyst, once (s)he examines the security interdependence and then chooses to deepen the analysis on that one ‘element’ of the RSCT that the Copenhagen School considers being the construction of the region. Likewise, actors performing the two speech acts – region building and regional security – will coincide, once the relevant actors performing a 89 A Methodological Trait D’Union certain security move are aware that they are also constructing (or attempting to construct, or obstructing the construction of) a region. If regional security actors and regional actors do not coincide, the analysis of the regional security will not correspond to a ‘reasonable’, ‘fair’ and ‘commonsensical’ state of the regional construction (see endnote 1). This is where the modernist Constructivism of the Copenhagen School is stronger, since such incompatibility between the regional security and the regional construction is to be attributed to the manner in which the analyst has selected ‘politically relevant’ discourses, and to how (s)he has evaluated the leadership of actors. If, for instance, a security community scenario should emerge from the security analysis while the state of the construction of the region is – ‘on the basis of common sense’ – at the level of balance of power, this means that the analyst has not chosen actors who are ‘politically relevant’ and their discourses are therefore not representative for the regional ‘collectivity’. In other words, if discourses on regional security and region building are not performed by ‘dominant’ actors, then the regional security and region building scenario these actors portray are not shared by a collectivity, are not strong enough to be enforced within the regional milieu, and thus, do not paint the dominant vision of the regional security and of the regional construction. Conversely, once security discourses do indicate a type of regional construction that is widely agreed upon by a collectivity, then it means that the analyst has selected relevant actors. It then follows that once actors are selected ‘properly’ in the critical RBA, then the question of the contrasting nature of the two analytical operations (RSCT and RBA) is actually a tautology on the ontological level. Relevant actors merely construct one reality, and it is up to the analyst to decipher its security dimensions and its regionalist dimensions, as well as their correlation. The actors are those who define regional security and the region, and the analyst should keep clearly separated from the actor in both dynamics. The analyst observes and judges while actors operate on empirical grounds. The reader ought to acknowledge the responsibility that this entire exercise involves, and the analyst must also bear it in mind. In order to make this exercise more transparent to the reader, I shall now turn to the actual tools of this analysis. 90 A Methodological Trait D’Union 2: Methods and Sources The proponents of the Copenhagen School argue, with respect to the selection of methods and sources in the security complex analysis, that “security experts usually draw upon a varied repertoire of sources, newspapers, conversations, theories and other academic writings on the case, which all come together in their skilful judgment” (Buzan et al. 1998, 178). In view of the ‘moderately modernist’ constructivism embraced by the critical RBA, this analysis will also be characterised by a ‘varied repertoire’ that can be divided in two main methods (and their sources): discourses (and their analysis) and case-study research. Prior to analysing these two elements, however, one short note should be devoted to the fact that, in view of my framework and the constructivist epistemology related to it, methods and sources employed in this analysis will be of a qualitative nature. W. Lawrence Neuman argues, “Qualitative researchers try to look at social life from multiple points of view and explain how people construct identities” (Neuman 2003, 146). Choosing qualitative instead of quantitative research is justified by the great importance attributed by this framework to the analyst and his/her ability and responsibility to contextualise and interpret given evidence, i.e. that which Neuman refers to as a ‘bricolage’ operation (Neuman 2003, 147-148). In this respect, the selection of the ‘repertoire’ employed in the analysis has been made on the basis of their (subjectively evaluated) meaningfulness with respect to the research questions of this study. Discourses Discourse analysis constitutes the major method to be employed in this study. I have thus far discussed that which discourses ultimately mean for this analysis, both in terms of security interdependence and the regionalist outcome, i.e. the speech act, the receptiveness of collectivity etc. Within this context, instead, I ought to indicate the methodological criterion behind the choice of texts to be analysed. A first defining criterion for distinguishing between discourses is that of hegemonic and competing texts. This distinction, originating from the work of Chantal Mouffe and Ernesto Laclau (1985, 105-145), focuses on that which has also been termed “intra-discourse analysis” (Kazan 2003, 112-113). Intra-discourse analysis focuses on discourses simultaneously practiced at different levels. In the case of this study, 91 A Methodological Trait D’Union intra-discourses focus on that which is said (and by whom) in the Baltic Sea Region at any given time. Within the hegemony/competing texts dichotomy, preference will be granted to those discourses that are actually dominant in the regional discourse; or rather, that the analyst considers dominant, on the basis of the procedures of the critical RBA. Wæver’s reading of Laclau and Mouffe’s conceptualisation can also be of some interest as pertains to this dimension of discourse analysis (Wæver 1989a, 188-191; Kazan 2003, 113). By distinguishing between official and debated texts, Wæver’s focus remains on the intradiscoursive dimension, but texts are divided along a slightly different line, i.e. between actual policies and not. The two intra-discourses differentiations – official/debated texts and Laclau and Mouffe’s hegemonic/competing ones – are complementary and both can be embraced. There is a third distinction of importance for the actual design of the analysis, which goes beyond the intra-discourse. Intra-discourses, it was argued, ought to scrutinise texts regarding the BSR at a given point in time. This is the synchronic dimension, since one looks at different texts produced at different levels within the same unit. This categorisation, however, neglects the contextualising element of the whole exercise, i.e. the need to integrate discourse analysis in a wider spatial and temporal framework. Hence, the diachronic dimension refers to discourses produced ‘outside’ the BSR (e.g. Brussels) but relevant for it, and, perhaps more importantly, it refers to how discourses change over time. The diachronic dimension is to be distinguished but is inseparable from the synchronic one, and provides the overall framework of reference. I shall now place the sources that will be employed for the analysis of the Baltic Sea within these distinctions of discourses. I can distinguish between at least three types of relevant texts (ordered according to when and where in the analysis these will be most relevant). The first and rather unconventional type of primary source for discourse analysis is constituted by a few specific academic and research works, primarily published during the first half of the 1990s, which focus on the topic of Baltic Sea cooperation (mainly in Chapter 4)5. Presenting this source at the top of my list of discourses is perhaps 92 A Methodological Trait D’Union rather debatable. Normally, academic and research works would be considered secondary sources. Being the work of analysts, these texts – in principle – do not tell much about how discourses are practiced within this framework (unless relevant actors are quoted in the analysis). Hence, academic and research texts are normally used as background and inspiration, and this is the manner in which they will be employed in this analysis as well. Nevertheless, the distinction between actor and analyst made above is again brought up in order to understand the methodological peculiarity of the few specific research works I am relating to here. Indeed, the analysis will refer to a group of intellectuals working with Baltic Sea cooperation who thus occupy a rather unique position in the region, simultaneously serving both as analysts of and actors in this regional phenomenon. This is to say, not only have these individuals taken the responsibility – like any other analyst – to judge and interpret the behaviour of actors in the security and regionalist milieu, and to derive their reflections accordingly. The methodological peculiarity is that these texts are written by analysts turned actors. Their actorness consists of two features: first, the texts produced by these analysts embody a wider trend concerning the Baltic Sea region. In this study, these texts are regarded as representative of a vast, although rather unstructured, variety of actors characterising the Baltic Sea area security and cooperation in this phase (mostly, it will be argued, grassroots networking). The relevance and ‘leadership’ of the discourses of these intellectuals consists in giving a voice to this widespread actorness. The (analytical) operation of systematising a wider variety of actors becomes a political act, if these actors do not do so on their own. Second, these intellectuals systematised the diversity resulting from the wide actorness they ‘represented’ by making statements, which constitute actual political propositions and ‘visions’ on how the regional security and cooperation in the Baltic Sea should develop. The fairly wide circulation of these texts and the experimental nature of the Baltic Sea Region contributed to the transformation of these meta-academic works into actual manifestos of how and what the region is and ought to be. With respect to the categories of discourses employed here, these texts cannot be viewed as either hegemonic or official. The weight of policy statements and official positions regarding the political 93 A Methodological Trait D’Union establishment is consider ‘heavier’ than are the discourses of Baltic Sea intellectuals in the overall economy of the analysis, quite simply because of the very ‘actorness’ of politicians and senior officials, i.e. their ability to reach (and influence) public opinion, their leadership (which endorses the security and regional speech act) and, not least, policy-making capabilities. Hence, the texts produced by Baltic Sea intellectuals should be viewed as competitive rather than hegemonic and are certainly debated, if one considers the different philosophical and ideological orientations that can be contained within my framework of analysis. My point is that these intellectuals and their works go somewhat beyond the traditional way of considering academic literature, i.e. secondary source, instead becoming a significant primary source of this analysis. This takes me to the second and more traditional type of primary sources that will be employed in the analysis more prominently in Chapters 5, 7 and in the Conclusions of this dissertation. This category comprises of selected speeches and statements from major foreign policy decision makers (as well as policy documents from governments or regional organisations) and qualitative interviews with senior governments officials. Both types of texts should be considered dominant (or hegemonic in the terminology of Laclau and Mouffe), because they reflect a position of leadership. A politician or senior official from one of the Baltic Sea countries that publicly expresses a judgment with respect to the state of the regional cooperation or the security situation in the region is a source that provides a fair indication of the terms by which the political establishment in that specific country is discussing the topic. Moreover, the very leadership embodied by these individuals has a dramatic impact on the common understanding of the issue. When, for instance, Russian politicians lash out in public against citizenship policies in Estonia or Latvia, they project an understanding of security and possibly an opinion concerning the prospect of regional cooperation that individuals who are interested in the issue most likely take into account. Once again, it should be stated that the meaningfulness of these speeches, documents and interviews is left to the judgment of the analyst, from which the radically qualitative nature of this research follows. There are a number of obvious limits to the analysis of speeches and public statements 94 A Methodological Trait D’Union owing to the rather wide chronological extension of the analysis, but also because of some inescapable technical limitations, e.g. the linguistic heterogeneity of the region – nine different languages for nine littoral countries – and the degree of comparability of the different statements. Concerning the former, I have embraced the rather comfortable – and admittedly realistic – claim that English has become a sort of lingua franca in the Baltic Sea region (see Joenniemi and Wæver 1992, 6), especially when it comes to issues concerning regional cooperation. In relation to the latter aspect, the comparability – synchronic and diachronic – of different discourses, I have attempted to consider primary sources referring to the same major gatherings focusing on BSR cooperation among foreign policy makers. One such gathering is the Annual Ministerial Sessions of the Council of the Baltic Sea States, where foreign ministers of the member states meet annually and deliver statements regarding different aspects of the regional security and cooperation. The second such main source is the Baltic Sea Region Security and Cooperation Conference Series organised by the American Embassy in Stockholm and held annually since 1996, which have gathered major foreign policy makers of the littoral countries, including foreign ministers, prime ministers, presidents and other senior officials. In all, 73 speeches and statements from major foreign policy makers from the littoral countries and 14 official documents issued by regional organisations and governments of the littoral countries are under analysis6. With respect to the twelve interviews carried out with diplomats and civil servants, a few remarks should be made concerning the importance and use of “unstructured, nondirective, in depth” (Neuman 2003, 390) qualitative interviews. I deliberately list all of these adjectives to stress the fact that the aim was not to base these interviews upon a questionnaire, to compare answers and, possibly, elaborate on them in a graph, as one would do in surveys. The object of these ‘chats’ was predominantly to set the context, encourage elaborations, and confirm the public statements of policy makers, i.e. a form of background work. Admittedly, this understanding of the interview makes my role as analyst central to both the elaboration of the answers and to the course of the interview itself (Neuman 2003, 395). On the other hand, the interview form also provides considerable freedom to the individuals that have been interviewed in terms of 95 A Methodological Trait D’Union applicable to my framework of analysis. Its constructivist nature, I noted, may suffer from the risk of being too subject to the responsibility of the analyst. Applying Ruggie’s argument, the possibility to overcome this relativism ultimately depends upon the ability to differentiate between actors involved in the regional process and to identify the dominant dimension that can enable the analyst to define regional change. This explains why the different parts of this analysis focus more specifically on some actors more than others. With this respect, focusing on one dominant actor ‘at a time’ (i.e. Scandinavian intellectuals first, then littoral countries, then Europeanisation etc.) can assist both the discussion of transformation as well as the observation of the constellation and interdependence between the actor in focus and other players in the regional debate. The second Durkhemian concept analysed by Ruggie is the determinant of change. This refers to the ‘dynamic density’ of a system, which is the “quantity, velocity and diversity of transactions that go on within society” (Ruggie 1983, 281). This touches upon another central element in the construction of this analysis. It relates to the multiplicity of actorness in this region and to the plurality of interactions within it. The point here is that focusing the analysis on one or a few of these transactions at a time does not mean ignoring the others. Nor does it mean that these social transactions are isolated from others. Most importantly, by focusing on one dynamic, I do not establish chronologically isolated compartments, i.e. one and only one actor defines the regional discourse at a certain point in time. The reason for focusing the analysis more prominently on one actor ‘at a time’ is ultimately of a methodological and ethical nature. Methodologically, I am driven by the need to underline the dimension of the change (as in Ruggie’s first point) and, not least, to grant the narration a measure of fluidity. Ethically, the reason for focusing on one actor at a time stems from the fact that this analysis is my own interpretation of the evolution of the Baltic regional security and cooperation. Therefore, while attempting to portray the plurality within the region, the analysis is also – and more strongly – driven by the need to define a pattern, which represents the view of the analyst on the issue. Nonetheless, the determinant of change at the ontological level is to emerge as clearly as possible. Hence, whenever possible, the analysis 102 A Methodological Trait D’Union will refer to the plurality of actors, even though a dominant actor is in focus. This is somewhat easier (and clearer) when the future prospects for this region are put forward (Chapter 7). That is because while the analysis of the past and present can count on how reality has unfolded, the definition of the dimension and of the determinant of change will necessarily be hypothetical and thus fuzzier when predicting the future scenario (even if on the basis of actors discourses and a case study). Structure of the Analysis In view of these arguments, I can now lay out the structure of the empirical analysis, as it will be discussed in the following chapters. The analysis will be broken down into three chronological phases, which portray my interpretation of the evolution and future prospects for regional security and cooperation in the Baltic. Each phase corresponds to one ‘loop’ in my research cycle, i.e. (1) selection and analysis of discourses, and (2) security analysis and regionalist outcome. Each phase is constructed upon the dimension and the determinant of reality. Pursuant to the constructivist epistemology behind this study, it should not come as a surprise that the construction of this three-phase structure followed the empirical analysis itself. Provided that, as noted, it is inconceivable to generate a purely inductive analysis, it can be argued that I have come up with such a construction largely a posteriori, i.e. after the analysis itself. This is actually where the distinction between ‘methodology’ and ‘method’ returns. The division in three phases reflects a choice made by the analyst, and this choice has been considered on the basis of the social construction at the heart of the methodology of this study. Such anticipation, however, should not stand in contrast with the constructivist epistemology of the study. The reader will indeed be accompanied through the actual analytical construction following the criteria and strategy presented above. The chart below indicates the three phases. The ordinate in the chart refers to the security continuum and thus spans from ‘confrontation’ to ‘security regime’, to ‘security community’. The abscissa indicates the timeframe from 1989 to 2004 and beyond. In each phase of the analysis, I will interpret the regional development on the basis of how dominant actors have practiced the security discourse in the Baltic. 103 A Methodological Trait D’Union Hence, the interpretation of security discourses, following the critical RBA, will indicate the kind of regional cooperation generated. Figure 19: Three Phases of Baltic Sea Region Security and Cooperation First PhaseThe Rise Second Phase – The Framing Third Phase: After 2004 Security Community Security Regime Confrontation 1989 1992 2004 Source: Author’s own elaboration (later discovered to bear a resemblance to Wæver 1989, 310). The First Phase of the analysis, ‘the Rise’, begins in the very early years of the post Cold War period and ends, rather shortly thereafter, in the early 1990s. This phase will be discussed in Chapter 4. Ideally, I indicate the year 1992 as marking the end of the first phase, although this timeframe will be interpreted flexibly, i.e. as an indicative landmark to facilitate the narration rather than a lock isolating the different phases. The Second Phase, ‘the Framing’, is longer and spans from the early 1990s (1992) to the year 2004. It will be elaborated in two chapters – Chapters 5 and 6 – to analyse the security discourses of two dominant (and deeply interdependent) actors/dynamics: the countries surrounding the Baltic Sea and the European level. The Third Phase, ‘After 2004’, will be discussed in Chapter 7, which presents a scenario based on the current discourses of dominant actors on the future of the region. Hence, it does not require an exact time-span – unless a time-span is indicated in the discourses taken into account – since the goal here is not to ‘predict’ a short, medium or long term for the future of this region, but rather, to analyse the security discourses about the future of the region and to test them against the model of environmental security. 104 A Methodological Trait D’Union Notes: 1 On the other hand, it ought to be noted that the RBA – more or less explicitly – acknowledges that the deconstruction of facts cannot be totally neutral. In relation to the ‘Japan’ example mentioned in the chapter above, for instance, this can be seen by the fact that Neumann notes that “it does seem more ‘natural’, ‘reasonable’, ‘fair’ to include the Scots than the Japanese” (Neumann 1992, 32) in the BSR (see also Chapter 4 below). 2 Morten Kelstrup pointed out this rather suggestive way of examining this issue to me (as the one ‘from above’ that is explained right below). 3 In order to eradicate possible confusion, I should state that the ‘top-down’ and ‘bottom-up’ labels attached here to the RBA and to the Copenhagen School, respectively, do not have to do with the way both regional constructions and regional security complex form, i.e. the ontological discussion. The continuum of regionalisms remains comprehensive for all approaches, and likewise, the formation of RSCs can be both top-down and bottom-up (see Buzan et. al. 1998, 198-201). Here I am dealing only with the epistemological debate, i.e. the formation of knowledge in the RBA and in the Copenhagen School. 4 The fact that the critical RBA is ‘in between’ the types of Constructivism envisaged by the RSCT and by the classical RBA could make my framework correspond to one of the other options described by Adler. The ‘Grotian’ scenario associated with the future of the BSR is quite consistent with the second type, where insights from international law are preponderant (see Chapter 7 below). 5 It may be useful to list the works I am referring to: Wellmann 1992; Jervell et al. (eds.) 1992; Wæver 1992, 1993; Joenniemi (ed.) 1993; Neumann 1992, 1994; and to a lesser extent Wæver 1997; Joenniemi 1999, which appeared later and, to some extent, represents interpretations of the early discourse. Particular mention should be made regarding the report authored by Pertti Joenniemi and Ole Wæver entitled: Regionalization Around the Baltic Rim: Notions on Baltic Sea Politics (Joenniemi and Wæver 1992), which perhaps put forward the most comprehensive argument. 6 Other minor sources of speeches and statements are Baltinfo – the CBSS newsletter - and Radio Free Europe/Radio Liberty Reports, and The Economist. 105 A Methodological Trait D’Union 106 First Phase: The Rise PART II The Analysis 107 First Phase: The Rise Figure 20: The Baltic Sea Area Source: The Library of the University of Texas. This map is in the public domain. 108 First Phase: The Rise 4 The Rise This chapter explores the early region building discourse in the BSR and its transformation from the end of the Cold War until the early 1990s. It will primarily rely upon the regional vision forwarded by a small group of Scandinavian intellectuals, hereafter referred to as ‘region builders’. The other, more traditional, primary source for the analysis in this chapter is comprised of official statements of regional institutions and policy makers on which, I will argue, is based the second argument for assessing the rise of the regional discourse. Before engaging in the actual discussion, however, a few preliminary observations must be made with regards to space and times in the Baltic Sea Region. I will then tackle the early region building by discussing the ‘remaking’ of Norden, which is the dynamic that most region builders regard as the conceptual ancestor of the BSR. In the third section, I will discuss the core of ‘Baltism’, which is considered as the main idea behind the notion of region building advanced by the region builders. In the following two sections, I will shift the focus of the discussion and examine the institutionalisation of the region and how this transformed the region building rationale. Particular attention will be devoted to the Council of the Baltic Sea States, whose establishment indicates, ideally, the culmination of this First Phase. The final section builds on an appraisal of this discussion in the light of my analytical framework. Here the correlation between security and regionalism is systematised and explained. 1: Prologue: Space and Time Baltic Space(s) To paraphrase the observation of geographer Mercator on the Mediterranean, the Baltic Sea is “given different names depending upon the shores it washed up against” (in Prodi 2002a, 8)1. Similarly, then, it would hardly be likely for the term ‘Baltic Sea Region’ not to refer to different things. Hence, I consider it suitable and useful to 109 First Phase: The Rise introduce this discussion on the rise of the BSR by defining, or at least sketching, the territorial dimensions of the region2. Both ‘postmodern’ and ‘traditionalist’ notions of regional space are quite problematic when applied to the Baltic case, although, ironically, they have produced somewhat similar (and equally debatable) definitions of the Baltic Sea space. As noted, the postmodern inclusiveness of some radical critical approaches to regionalism can theoretically push the borders of the Baltic region to Japan, if one considers trade relations with the Baltic Eastern shore (Joenniemi & Wæver 1992, 30). On the other hand, some littoral states are not Baltic countries tout-court simply because they border the Sea. Hence, those traditionalist studies that tend to refer to Russia as a whole (Hovmand 2002, 48) will implicitly push the borders of the BSR as far as Vladivostok in East Asia, the eastern frontier of that country ( which is not too far from Japan!). Arguably, “‘regions’ divided up according to the different conceivable criteria [never] coincide perfectly” (Bourdieu 1991, 222); however, the point here is that if one were to adhere to either of these criteria, the outcome in terms of defining the Baltic territory appears equally misleading. More importantly, in line with the constructivist epistemology followed by the study, the spatial delimitation of this region should emerge from the analysis itself. Hence, the regional territory will be reduced3 or widened4, state-centric or not; and its borders can be more5 or less6 flexible depending upon the dominant security dynamic in the region. Yet, as part of the ‘modernist constructivism’ to which this study has chosen to adhere, I noted in the Third Argument of the analytical framework that the spatial construction can be functionally defined and then ‘bent’ according to the actual construction of the analysis. I can therefore provide a few geographical coordinates in order to offer an initial landmark, a sort of mental map the reader can refer to as a point of departure. This map of the Baltic Sea Region corresponds to: the German Länder (states) of Schleswig-Holstein, Hamburg, and Mecklenburg-Vorpommern; Northern Poland7; Lithuania; Latvia; Estonia; the Leningrad and Kaliningrad oblasti (regions) and the St. Petersburg Municipality (Russian Federation)8; Finland; Sweden and Denmark9. 110 First Phase: The Rise Roots The post-Cold War rise of regional cooperation in the Baltic Sea area was accompanied by frequent references to previous historical experiences. Thus, prior to engaging in the actual discourses, a few notes should be devoted to interpret the historical developments that have been part of the ‘cognitive evolution’ characterising the rise of the post-Cold War regionalism in the Baltic Sea. I will briefly recall four historical experiences. The first reference is to the Hansa period. It is of relatively little interest to this study to delve into the historical disputes on the actual features of the Hanseatic League (cf. Kirby 1990). What is of importance, however, is to sketch how and on what basis this experience was rediscovered in the 1990s. Intellectuals and policy makers alike pointed to the innovative constellation of power characterising this experience. The Hansa fostered close ties among more than one hundred cities surrounding the Baltic Sea for more than three centuries, stretching from the fourteenth to the sixteenth century. This aggregation fuelled increased regional economic transactions and served as the basis for social and cultural networking. These features were at the centre of the image of the ‘New’ Hansa (Joenniemi and Wæver 1992, 4), which emerged in the late 1980s. The grain trade was a major impulse for regional cooperation in the fourteenth century, just as the ‘new’ Hansa was envisaged as taking shape for reasons relating to prospects for regional economic growth. Just as the first ‘old’ Hansa promoted social and cultural interaction around the Baltic Sea without being tied to any of the great powers of the region, so also was the post-Cold War BSR supposed to nurture a sense of ‘we-feeling’ without necessarily being attached to states. In relation to my theoretical framework, the Hansa experience is also suitable to exemplify the linkage between security and regionalism. The Hansa provides a model in which regional cooperation was initially rather fuctionalised in relation to some basic security needs, e.g. economic security for trade in the Baltic Sea. This economic securitisation enabled a social and cultural ‘we-feeling’ to emerge among the cities participating in the network. This generated a rather ‘Germanised’ commonality, since the activism of the German merchants remained instrumental to ensuring its development; however, the point vis-à-vis the correlation between security and 111 First Phase: The Rise reluctant” (Wæver 1994, 4). Indeed, the post-structuralist Scandinavian region-builders were heading towards Europe. Europe is to become the centre and Norden would inexorably be left at the periphery. It was no longer viable for Norden to construct itself in opposition to Europe, as had traditionally been the case, and the need to grow nearer to Europe became apparent (ibid. 5). In Joenniemi’s words, this encounter with Europe transformed Norden into a somewhat more ‘modern’ and state-based entity (Joenniemi 1998, 11). This resulted in a sort of compromise in which: “(t)he argument can be made that Norden holds something that the EU is still aspiring for. In terms of time, Norden is ahead in being far less governed by modern, security-centred accounts and in having turned postmodern not by rephrasing security but by leaving the whole concern aside. Security has, in the case of the EU, been as inroad to postmodernity while the Nordic case indicates that the departure can also be one of dropping the theme of security.” (Joenniemi 1998, 18) This is where ‘Baltism’ enters the picture. According to Scandinavian region-builders, the Nordic path to Brussels passed through the Baltic Sea. Embracing the south-eastern shore of the Baltic rim was a means by which to mark Nordic willingness to adapt to new conditions and challenges. Nordics “learned that they did not become less European by turning East; on the contrary, they were seen as very useful to continental actors if they ‘handled’ Baltic challenges” (Wæver 1997, 312). Hence, the Baltic challenge was to become a new identifying element of ‘Nordism’. This is where the conceptual shift is stronger. In the light of the process of European integration, indeed, Nordism concerns something ‘external’, i.e. the Baltic Sea region, rather than merely internal, i.e. Norden itself (Wæver 1994). This transformation from ‘internal’ to ‘external’ has the character of an identity project and also determines significant security implications; their importance for the analysis at hand will be better assessed in the final section of this chapter. At least three important dynamics, however, ought to be spelled out here. The first dynamic is eminently external (and rather functionalistic). The promotion of Norden in Brussels through the Baltic Sea project was regarded as 118 First Phase: The Rise serving the rather straightforward purpose of protecting Northern Europe from the unknowns of the new post-Soviet entities situated in its immediate vicinity (cf. Wæver 1997, 315). The second challenge is of an internal nature, as epitomised by the title of Wæver’s article “Balts, Books and Brussels” (1994). The Nordic shift towards the Baltic region, as noted, implied that the most challenging security threats were to be handled in light of the Europe-wide dimension. What remained for Norden itself, then? Mostly ‘softer’ issues, such as culture, research or education (and therefore ‘books’), in which the Nordic identity marker could be preserved. The third and final implication, somewhat half-way through the internal and external challenge, points to the need to enhance the Baltic project in order to differentiate it from Norden itself (cf. also Wæver 1997, 323-324). The rationale was not to make a larger ‘Norden’ out of the Baltic Sea. Nordic experience and values were to be exported and made available to the Baltics, but Norden was neither to expand nor to ‘blend’ into them. Norden provided values and experience capable of influencing the Southern Baltic transition, but never really aimed at enlarging the Nordic model itself15. On the other hand, the Baltic project could be construed as reinvigorating Norden, its values and its core beliefs. In Wæver’s words, it could “serve to keep a Nordic profile alive” (Wæver 1997, 324). 3: ‘Baltism’ The critical assessment of the ‘ideology’ behind the Baltic Sea region building during this First Phase will be carried out by reframing the relevant discourses of the Scandinavian region-builders, as they arose in the aftermath of the Cold War. The actorness of region builders, I noted, is viewed as relevant, because it ‘gave a voice’ to a much wider and unstructured variety of actors and systematised the diversity of these actors by putting forward coherent political propositions. In order to provide a comprehensive account of the various elements contained in these BSR ‘manifestos’, I have reframed the discussion on ‘Baltism’ into four main pillars: the European dimension, Multiplicity, Identitybuilding and Security. 119 First Phase: The Rise The European Dimension This element follows directly from the discussion of post-Cold War Norden. The EEC could already be characterised as a rather experimental polity in the late 1980s. Scandinavian region builders grasped this diversity despite being uncertain as to which direction post-1989 European integration would actually take, e.g. ‘interstate’, ‘super-state’, ‘sub-state and around the state’ (Joenniemi and Wæver 1992, 23), or all of these options together. They opted to place emphasis upon the latter of these options in order to introduce the importance of regionalism in Europe. In Joenniemi and Wæver’s report (1992), this is explicitly expressed in different stages of their comprehensive argument. One can actually argue that the European dimension is the element pervading the entire narrative. Indeed, according to the region builders, social and political forces were necessary to re-balance the new Europe and: “regionalism enters the picture as a policy aiming at transcending and softening the choice between centrality and periphery” (Joenniemi and Wæver 1993, 4). Region-builders viewed Europe as the ‘proper’ region: “Europe is a region from the perspective of international relations, and all groupings inside Europe are therefore subregional” (Wæver 1993, 25). This discussion is reminiscent of the distinction between ‘regions’ and ‘subregions’ referred to above in the Third Argument of the analytical framework. The peculiarity here is that while Norden is viewed as one of these sub-regions, the BSR is instead considered to be a “protoregion” (ibid., 43)16, a sort of new entity that conceptually and spatially could be placed somewhere in between the other two. Joenniemi and Wæver also stress the fact that it is precisely the eminent diversity of Europe that will allow for a natural repartition of competences among different actors, including ‘proto-regions’: “We will see an increasing – not yet settled – division of labour between the EC-level, the state-level and the level of trans-regions (like the Baltic Sea). They take on different functions, they are about different things, and they are organised differently. Thus, it is misleading to imagine the Baltic Sea Region as a kind of mini-EC (or the EC as a giant Baltic Sea cooperation).” (Joenniemi and Wæver 1992, 1) 120 First Phase: The Rise On the other hand, this characterisation of the Baltic entity is also likely to pose significant challenges to the construction of Europe itself. “There are reasons to assume,” Joenniemi and Wæver argued, “that even transborder regionalization, such as the one occurring around the Baltic Rim, is stimulated by the ongoing EC integration, although the relationship is a rather complex one and is sometimes even conflictual” (Joenniemi and Wæver 1993, 4 and see Wæver 1997, 311). The uncertainty surrounding the future structuration of Europe renders this problem rather sensible, also today. Europe also represents one of the keys to understanding the economic dimensions of the prospective BSR. The argument was that the BSR would constitute itself as one of Europe’s ‘bananas’ of economic growth. The Sea will indeed resemble a “blue banana” of growth running parallel to the lines of the ‘yellow banana’ of growth in continental Europe (Joenniemi and Wæver 1992, 4; Neumann 1992, 25). Finally, there is a more traditionalist argument, which, paradoxically, is also suitable for paving the way for the second poststructuralist dynamic characterising Baltism, i.e. ‘Multiplicity’. It was argued that the dramatic geopolitical changes following the end of the Cold War – the demise of the Soviet Union and the independence of the Baltic States – sparked the regional discourses. Within this context, region builders argued that the reunification of Germany was a crucial development. The traditionalist argument here is that the reconstitution of this large continental actor came to pose new challenges to the balances of the Continent. Nevertheless, this geo-strategic element was only the ‘visible’ part of the story. In fact, German influence in Europe – referred to as the Germanisation of Europe – was to be mostly ‘invisible’ (Wæver 1993, 30-31). The new ‘German’ Europe was to be characterised by the participation of different German Länder in several parts of the Continent (not only in the Baltic Sea, but also in Southern, Central and possibly Eastern Europe) and by their connection to numerous European economic hubs. Hence, for the region builders, vast, eminently diverse and heterogeneous German (economic) influence was to characterise the continent; influence that would fit the Baltic Sea quite comfortably, since this, after all, is what the Hansa metaphor essentially involved. 121 First Phase: The Rise Multiplicity The German sub-regional influence in Europe contributed indeed to enhancing the recurrent image of the BSR as a new Hansa. This was, however, merely a partial means of examining the issue. The essence of the new Baltic Sea cooperation was not only in the functionalistic Hansa-like economic networking, but also and foremost in that which cross-border cooperation implies conceptually and symbolically. This represents the actual core of the neo-medieval metaphor of regional cooperation: “the new structures can be seen as neo-medieval [in] the sense that overlapping authority is returning” (Joenniemi and Wæver 1993, 5 and see also Ruggie 1993, 150). The novelty of the Baltic regional experience was enshrined in the post-state nature of this multiple cooperation, which is what ultimately incarnates the political subjectivity of the region. Within this context, the point of crucial importance stressed by region builders was that states and governmental institutions could participate in this regional project but were not supposed to steer – and thereby suffocate – it (see Joenniemi and Wæver 1992, 21 and 34; Wæver 1993, 26). Multiplicity in the Baltic, therefore: “is a configuration that connects a variety of different actors, and is composed by a setting of multiple overlapping networks. ‘Balticism’ operates without constituting any distinct hierarchy; rather it challenges the old ones. Region formation around the Baltic – in the form of trans-regionalisation – brings together a number of views that are sometimes very different or even in conflict with each other, and links various levels such as the subnational, transnational and supranational.” (Joenniemi and Wæver 1992, 35) The fact that the BSR was a “meeting ground for different agendas, priorities and modes of understanding” (Joenniemi and Wæver 1992, 7) was indeed rather self-evident. Cooperation literally blossomed among businessmen, NGO activists, artists and scientists representing all of the countries around the rim. It is during this First Phase that an Ars Baltica concerning cultural cooperation was established (since 1988). The Baltic University Programme and an Organisation of the Baltic Ports have existed since 1991. The Pro-Baltica Forum (an 122 First Phase: The Rise association for political and economic integration) was launched in 1992; the Baltic Media Centre came to life in 1993; and the TransBaltic Network was established in 199417. These examples are merely illustrative of the fact that the region builders were indeed conceptualising something that was actually proceeding around them. They viewed grassroots and trans-local multiplicity as the core of the region, because that is where the construction of the region actually began. This diversity also implies, however, that different views of the region were simultaneously put forward. The question, then, is how to build one Baltic region out of this multiplicity and plurality. For poststructuralist region builders, this did not actually represent an issue. Multiplicity does not only imply a plurality of ways by which the BSR can unfold, but also, and more interestingly, it generates several Baltic Sea Regions, which also serves to define this phenomenon: “A deconstruction of the debate on the Baltic Sea region suggests that it is not only about one region, delineated and defined in the same way by the various parties. There are a variety of voices and projects that compete with each other, and sometimes conflict with each other, but that also seem to reinforce each other and the general trend towards Baltization. Thus, divergence is not only a source of weakness, but also a source of strength.” (Joenniemi and Wæver 1993, 7) On the other hand, it would appear as though there are also certain ‘reasonable’ limits upon the workings of the most radical poststructuralists with respect to the number of Baltic regions that can be achieved: “the organisations and activities […] vary in terms of the number and type of participants, but they are all Baltic-specific and cover the region quite broadly” (Joenniemi and Wæver 1992, 15). This means, for instance, that the radical deterritorialisation that region builders put forward, i.e. the ‘Japan’ argument, can and should be delimited, i.e. “it does seem more ‘natural’, ‘reasonable’, ‘fair’ to include the Scots than the Japanese” (Neumann 1992, 32). Ultimately, this multiple translocalism is likely to generate one Baltic region, at least in the minds of its participants, because of the ‘voluntaristic’ nature of the regional endeavour: 123 First Phase: The Rise “These activities seem to have developed in a relatively spontaneous and fragmented way on the basis of the prevailing needs without a master plan, scheme or framework. It may be envisaged that this growth will continue in the near future across the board, although developments will presumably take place in a somewhat more coherent way, curtailing the fragmentation that so far has labelled the pattern.” (Joenniemi and Wæver 1992, 15-16) This quote suggests that multiplicity ought to contribute to a more ambitious project than the mere cross-border cooperation. The regional multiplicity makes the next step possible, which represents the actual ontological core of the theorisation of the region builders: regional identity. Identity Building As Wæver writes, “in the last instance identity is a question of who wants to see themselves as belonging” (Wæver 1993, 24). This suggests that the Baltic multiplicity could provide people around the rim with an additional (but not substitutive) element of we-feeling (Joenniemi and Wæver, 1993, 5): people cooperate and thereby come to share something in common. This does not quite point to what ‘Baltic identity’ is actually supposed to be about; however, the very post-structural inspiration behind these early discourses precisely implies that a clear-cut definition of regional identity is difficult to grasp. Baltic regional identity is about experimenting with new modes of cooperation; it is about doing things together and differently; it is about sharing experiences and cultivating person-to-person contacts; it is about ‘talking about’ the region and establishing social practices; it is a self-reinforcing, all-inclusive and ‘never complete’ dynamic (Wæver 1997, 302; see also Joenniemi and Wæver 1992, Neumann 1992 and 1994; Wæver 1993; Joenniemi and Wæver 1993). In such a context, the reference is to ‘identification’ rather than identity (Wæver 1993, 42). Region builders viewed this identity building (as well as the construction of a physical infrastructure) as the first of three steps in the project of constructing a Baltic region – where the formation of a regional economy and politics constituted the second step and security 124 First Phase: The Rise is the final one (cf. Joenniemi and Wæver 1992, 32; and Wæver 1997, 307). As this deliberately loose regional ‘plan’ unfolds, region builders also provide the ‘tools’ necessary to enforce it. They start by indicating how to build this regional identity in practical terms. Regional “self-referential cultural processes” (ibid.) are regarded as being at the centre of identity building. This is enshrined in the important linkage between culture and identity. Culture and ‘culture politics’ are viewed as the central vehicle for disseminating and deepening a common ‘we-feeling’, particularly within the context of a rather loose conceptualisation, as in the Baltic case. In Wæver’s words: “We are not used to thinking in terms of identity politics. Culture is the place where our communities (at an aggregate level) and our political systems are able to articulate visions, problematize and rearticulate self-conceptions, and address the question of who we are and where we belong.” (Wæver 1993, 40) This is the item on the post-structuralist regional agenda in which intellectuals possess an explicitly political role. Culture politics and intellectuals are indissolubly linked to the process of fostering a regional identity. Once again in the words of Wæver, “There are plenty of cultures and identities in the Baltic Region. Baltic culture and identity is that which we are now making – the product of intellectuals, conferences, and news reporting on conferences” (Wæver 1993, 40). This means that the very conceptual reflections of intellectuals carried an eminently political meaning, which is that of establishing “certain patterns,” i.e. the region, “as natural” (Wæver 1993, 24). Their first gatherings18 for instance, had a strong regionalising function. These created the ‘myth’ and the metaphors of the BSR, which can be perceived as the virtual constitution of this postmodern region. As Wæver explains: “Why should conferences and rumours suddenly be important? Because of the role they play in myth making – as ideology produces. The first step in this kind of cooperation has to be to create the image that The Baltic Region exists to make it more natural for business, political activists, tourists and students to follow this pattern.” (Wæver 1993, 39)19 125 First Phase: The Rise In sum, region builders ‘invented’ the region based on hypotheses and expectations20, and on few significant signals that could be observed on the ground, e.g. grassroots cooperation. Security Security is a potentially confusing concept in this particular context. First, one should note that the few observations in this section will not involve my own interpretation of the connection between security and regionalism in the First Phase, which will be discussed towards the end of this chapter. Here I merely complete this assessment of the arguments forwarded by the early region builders by referring to their security component. Second, one cannot avoid noticing the personal connection between the Scandinavian region builders and the ‘Copenhagen School’ of security studies, i.e. some of the Scandinavian region builders are the Copenhagen School. This implies that a conceptual overlap between the manner in which security is conceptualised by the Copenhagen School and the way region builders treated security in the BSR is unavoidable. Region builders have explicitly declared that security is one of the most significant aspects in the regionalisation process (Joenniemi and Wæver 1992, 18-19). They established the correlation between the post-Cold War security changes and region building, e.g. the role of Russia vis-à-vis the Baltics, Germanisation, etc., and they hinted at the security interdependence between Europeanisation and sub-regional cooperation (Wæver 1993, 26). Region builders also refer rather loosely to securitisation and desecuritisation processes (which predominantly occupied Ole Wæver, one of the Baltic Sea region builders) by advocating that the concept of security in the BSR is to be widened both spatially21 and conceptually22. Finally, as already noted, they refer to ‘security’ as the third and final step in the region building ‘plan’ (after ‘culture/identity building and infrastructure’ and ‘economic and political integration’). The few observations on security in the BSR sketched in the lines above are the primary and most significant elaborations on security that can be found in the texts of the region builders. The fact that most region builders worked considerably on security issues but dealt with them somewhat briefly in the BSR context suggests that region 126 First Phase: The Rise building was also but not only (and probably not principally) about security. 4: Institutionalisation In this section and the next, the critical review of the discourses of the region builders will continue. As opposed to the previous section, however, those arguments will be accompanied by – and confronted with – the rationale behind the establishment of regional institutions, most importantly the Council of the Baltic Sea States (CBSS). The purpose for doing so is to portray the differences between the two visions and to observe the transformation in the regional rationale during the First Phase, which is brought about by institutionalisation. It has been argued that the Baltic region envisaged by Scandinavian intellectuals was, more or less radically, post-statist. This meant that regional clustering was to be characterised by a sub/non/‘around the’ state sort of multiplicity, rather than the stiffened traditionalist ‘topdown’ formation in which states hold supremacy (Joenniemi and Wæver 1992, 35). Of interest here is that this post-structural inspiration is mitigated by the need to make this region building process not only innovative, but also credible and feasible. Hence, of all the prefixes that could be added to the term ‘state’, region builders did not perceive the regional experience as necessarily ‘anti’-statist: “Regionalism is often regarded as something anti-statist, but it could also be viewed as a safety-net which favours the states and the application of statist principles. States may join, too, and they are allotted the position of partners but have to refrain from claims of traditional supremacy.” (Joenniemi and Wæver 1992, 35)23 This meant that the ‘bottom-up’ core of the region was not to generate a confrontational power struggle with states. It simply meant that the ‘formalisation’ of the regional cooperation was not necessary. Formal groups would be created to facilitate the regional process rather than to embody it – if they were created at all: “It is not so that the region is the political institution” (Joenniemi and Wæver 1992, 32). Actual cooperation was to be found elsewhere and institutions were to be support structures: “in the Baltic Sea project, the political institutions are there to create the conditions for “real” cooperation” 127 First Phase: The Rise Figure 21: Regional cooperation in the Baltic Sea during the First Phase First Phase – The Rise Second Phase Third Phase Security Community Security Regime Confrontation 1989 1992 2004 The end of the Cold War revolutionised the patterns of power in the Baltic region, as in Europe at large. Contacts within the Baltic region prior to 1989 were practically absent at most levels, and the BSR symbolically incarnated a space that the overlay of the Iron Curtain divided in almost equal parts. Environmental security represented a minor, though significant, exception. Otherwise, confrontation was the characteristic security interdependence in this area, and regional cooperation was non-existing. The BSR happened to be right in the middle of the rapid shift in the power constellations following 1989. This shift brought about a pattern of widened cooperation in which a vast range of issues was tackled at the regional level by a wide variety of actors. I am primarily referring to non-governmental organisations, local administrations and single individuals who capitalised on opportunities to have a say in issues beyond local or national reach. The majority of these actors were new and unaccustomed to the management of trans-national challenges, and the typology of these regional phenomena did not have much of a precedent. This also rendered the post-1989 period in this region rather experimental. This was the security milieu as it appeared to the Scandinavian region builders when they tabled their vision for the BSR. This is a vision that had the unique quality of being able to systemise the much wider grassroots actorness into a single political message. This may explain why their vision seems to surpass the mere linkage between 134 First Phase: The Rise security and regionalism, and includes other and more central elements, e.g. economic cooperation, social cohesion and culture politics. Nevertheless, the focus on security will reveal another possible reading of this dynamic, which also explains the rationale behind the following step, i.e. institutionalisation. Region builders envisaged ‘regionality’ (Wæver 1997, Joenniemi 1999) as the ultimate form for this region to assume. This prospect pushes the features introduced by the Scandinavian region builders (identity, economic integration, the European dimension etc.) to a higher degree of interdependence in the regional framework, one which indeed closely resembles Norden (without necessarily requiring amalgamation). In this process, ‘security’ is indicated as the third and ultimate step of the (rather loose) plan of region building. The correlation here between security and regionality, as proposed in the Second Argument of the critical RBA, becomes apparent. Indeed, focusing on elements other than security, and simultaneously making ‘security’ the paramount objective of region building, suggests that the Baltic Sea was an embryonic ‘security community’; or at least potentially so (see also Joenniemi and Browning 2004). This evolution of the Baltic regional cooperation into a security community was not to occur in the foreseeable future. The Eastern transformation, historical cleavages and uncertainty regarding the future rendered predictions somewhat vague; however, the notion that the region ought to head towards regionality (and thus security community) was part of the region-builders agenda. Statements such as “Does a Baltic Sea Region exist? No, but it will” (Wæver 1992, 38), which time and again appear in the conclusions of the founding texts of the region builders (cf. also Wæver 1997, 330) appear to confirm this assessment. If one were to alter the word ‘region’ into ‘regionality’, or even into ‘security community’ in such a statement, the deepest meaning of the project put forward by the Scandinavian region builders would perhaps become even clearer: ‘Does a Baltic Sea Security Community exist? No, but it will’. This is to say that the argument of the region builders revolved around the powerful processes of desecuritisation following the end of the Cold War. These processes opened regional cooperation up to issues that were increasingly unrelated to traditional security, and which eventually were to sideline security altogether, thereby transforming the region into a security 135 First Phase: The Rise community, i.e. a Baltic Sea regionality (cf. also endnote 11 in Chapter 2). This reading of the early regionalist phenomena in light of security interdependence underlines the important question related to the distinction between actor and analyst introduced in the First Argument of the critical RBA, which was further specified in the methodological chapter. Choosing to analyse the discourses of the region builders has required consideration of their actorness as somewhat representative of the wider bottom-up debate. It has therefore implied that I, the analyst, have chosen to select the region builders as the leading speech actors. I have warned against this rather unique overlap in Chapter 3, where I noted that the region builders are analysts turned actors. The region builders did so themselves, e.g. when they underlined the political nature of their statements (see Neumann 1992 and 1994; Wæver 1997). However, that which the present analysis on the correlation between security and cooperation reveals is that region builders, in their capacity as analysts, did not – or simply did not wish to – explicitly distinguish between security analysis and region building (instead, this argument comes out explicitly in Joenniemi and Browning 2004). Security is present in their assessments; however, the focus on (prospective) regionality as the ultimate (political) goal of region building necessarily implies sidelining any security talk. The fact that the actual term ‘security community’ does not appear too frequently in those early writings, while the contents are rich with references to it, is a confirmation of the fact that region builders – as analysts – did not narrowly focus on security interdependence because, as actors, they practiced desecuritisation and aimed at its ultimate form, security community. This may also explain the territorial inclusiveness of the Baltic region in the formulation of the region builders. Within the context of the Third Argument of the critical RBA, it was argued that the space of regional security is delimited somewhat functionally, apart from the case of ‘security community’, which indeed sidelines security talk and thus also allows for the regional territory to be socially constructed in a more radical manner. Interpreting region builders discourses as aiming towards security community also serves to explain the radically constructivist core of the territorial construction, e.g. the ‘Japan’ argument. 136 First Phase: The Rise Correlating this kind of a security analysis with the regionalist outcome, the resemblance with a ‘new’ kind of regionalism appears rather self-evident. In the arguments above, one could spot ‘de’ and ‘reterritorialisation’, a ‘post’-statist regional space, participation and multiplicity ‘from below’. This may explain why the curved line depicting the post-1989 cooperation in the chart above rises almost vertically, towards regionality. Such dynamics, however, are also instrumental in explaining how institutionalisation entered into the picture. Region builders argued that states and institutions were intended to be an active part of their region building vision. As Joenniemi writes, “The driving forces often have a bottom-up character then to be complemented by various top-down measures” T(Joenniemi 1999, 4). As noted, one model closely resembling such a view of the institutional dynamic is Euroregions, where cross border cooperation is often regarded as being at the heart of ‘community building’. However, Euroregions are based on a different understanding of territoriality than the creation of mesoregions such as the Baltic, as they are not pan-regional in a Baltic Sea perspective and should rather be perceived as ‘one of’ the actors – and a very proactive one – involved in regional cooperation. Other forms of institutionalisation, ranging from local (UBC and BSSSC) to state-based institutions (CBSS), did not subscribe to the same vision of the region as that advocated by the region builders. Most importantly, they did not share the same vision in terms of regional security; in fact, they opposed it. To put it bluntly, while region builders preached progressive desecuritisation, regional institutions structured cooperation around the resecuritisation of an array of issues. Rather than viewing the BSR as a security community in the making, institutional actors had the preservation and promotion of common interests as the ultimate goal for regional cooperation. As Uffe Ellemann-Jensen stated, “It is important that we in our future cooperation focus on issues which are motivated by practical problems and which may best be dealt with within a regional framework” (Ellemann-Jensen 1992, 8, emphasis added). One may argue that regional institutions welcomed broader actor participation in the regional arena. Most importantly, they widened the agenda of cooperation to include ‘soft’ issues, such as the environment, economy, and crime prevention. However, the principal reason for this 137 First Phase: The Rise widening operation was in fact the need to contain security challenges that were utterly ‘hard’, e.g. threats arising from the regained independence of the Baltics (e.g. the presence of Russian troops on Baltic soil) or the uncertainties surrounding the reunification of Germany, etc. (e.g. Ellemann-Jensen’s ‘damage control’ argument). Hence, the dominant security discourses at the institutional level underline that institutional actors resecuritised the region around nonmilitary issues29 in order to contain more imminent and serious threats. As a result, regionalist scholars would argue that regional phenomena are viewed in a more functional manner as a ‘problem solving’ tool. This explains why the curved line in the chart turns to indicate a security regime status quo. For institutional actors, as for the region builders, the BSR represented a ‘new’ phenomenon; however, the premises, priorities and goals behind this label were intrinsically different: common interests for the institutional actors vs. identity building for the region builders; resecuritisation vs. desecuritisation; regime vs. regionality. Hence, the regional organisations, and most importantly the CBSS, did contribute to enhance the Baltic Sea as a ‘new’ region, but also slightly transformed it, if one uses the Scandinavian region builder as the ideological point of departure to gain an understanding of the bottomup component of regionalism. The institutional component of this analysis is also a crucial pass to proceed to the next phase. It suggests that, according to the present interpretation, dominant discourses regarding regional cooperation shifted progressively towards a more ‘formalised’ dynamic in which states, as well as ‘outside-in’ European developments, came to play a more important role. As Wæver writes, “the project has come under increasing pressure – both from the most powerful projects in the West and from re-mobilisations in East as well as West of state based security logic or (neo-) nationalism” (Wæver 1997, 330 emphasis added). In this respect, states have not quite been the type of actors for a regionality-in-the-making sort of project. Moreover, European integration is not (or at least not only) the loose and ‘poliarchic’ dynamic envisaged by region builders. This explains why, according to Wæver, the project came ‘under pressure’. Furthermore, this is also why, according to the present interpretation, the institutionalisation of 138 First Phase: The Rise the early 1990s is a suitable stepping-stone for discussing the transformation of the regional cooperation and the role of states and Europeanisation in it during the Second Phase. This transformation also points back to the problem of the ‘dimension of change’ that was sketched in the methodological discussion. The fact that states and Europeanisation are perceived as the dominant agents in the shaping of the regional discourse in the Second Phase does not mean that the non-state and local component, or even the vision that was put forward by the Scandinavian region builders, simply ‘disappear’. They do remain part of the discourse. In fact, they grew considerably during the 1990s in terms of number and type of activities. In order to follow the course of the regional cooperation, however, it is imperative for this study to identify ‘whose region’ one is dealing with. This must be carried out ‘modernistically’, i.e. by selecting dominant actors. As a result, this shift in focus towards states and Europeanisation indicates that those other ‘bottom-up’ actors progressively lost momentum in shaping the regional debate. Conversely, the institutional and governmental developments gained visibility in the project, though they also transformed its very rationale. 139 First Phase: The Rise Notes 1 E.g. the very term Baltic Sea translates as “Eastern Sea” in the Scandinavian languages, German and Finnish, and as “Western Sea” (Läänemeri) in Estonian. 2 In these analytical chapters, “Baltic region” will be used as synonym for “Baltic Sea Region”, and not, as is predominantly the case, used to refer to the three Baltic Republics of Estonia, Latvia and Lithuania. 3 When discussing the field of sub-regional cooperation within the framework of the Union of the Baltic Cities (UBC), the Baltic Sea Sub Regional Co-operation (BSSSC). 4 E.g. in the case of the institutional cooperation within the Council and the Baltic Sea States (CBSS). 5 When, for example, economic cooperation is the focus of the analysis, the very flexible metaphor of the ‘Blue Banana’ (Joenniemi & Wæver 1992, 4) applies. 6 As in the case the environmental security argument, see below in this chapter and in Chapter 7. 7 It could be argued that considering the varying degree of involvement towards the BSR within Poland (cf. Musial 2002), status as Baltic Sea political actors ought to be limited to the administrative units (voivodships) bordering the Sea: West Pomeranian Region (Zachodnio-Pomorskie), Pomeranian Region (Pomorskie), Warmia and Mazury Region (Warminsko-Mazurskie) (which are the three polish regions member of the BSSSC cf. below in this chapter). However, considerations of an administrative nature make this further distinction more complicated. Poland is not a federal state and therefore decentralisation of the decision-making is less evident and more complex to unfold than for instance in federal states such as Germany or Russia. 8 Some authors also include the Novgorod and Pskov oblasti (cf. Oldberg 2000) and the Republic of Karelia as Russian Baltic regions. 9 The same argument advocated for Poland may apply to Denmark, Lithuania, Finland, and Sweden (cf. endnote 6) and therefore, although at a domestic level, the spatial distribution of the actors with respect to the BSR is very diversified, we consider these entire countries to be ‘Baltic’. 10 In terms of drawing a distinction between Norden and Scandinavia, Pertti Joenniemi notes: “Norden is a somewhat broader term than Scandinavia (Denmark, Norway and Sweden) as it includes also Iceland and Finland. By and large these two terms may be used as synonyms, although ‘Scandinavia’ is more original, and ‘Norden’ has primarily entered the vocabulary after the world wars” (Joenniemi 1997, 228). 11 These other four entities, each with its own unique status with respect to the ‘mainland’, are the Faroe Islands and Greenland (Denmark), the Aalands Islands (Finland), and the Sami nation (Norway) (for an account of these special cases cf. Lindström, 1997). 12 EEC refers to the pre-Maastricht period and European Union to post-Maastricht (which came into force in November 1993). For the sake of clarity, because the themes discussed in these chapters are somewhat blurred chronologically, I will use the ‘EU’ abbreviation in the text, unless the distinction is of importance for the narration, as in this case. 140 First Phase: The Rise 13 I.e. the two ‘Winter Wars’ against the Soviet Union during the Second World War. 14 In this respect, comparison is recurrent in the literature on Norden, between the successful independence wars ‘from above’ in Italy and Germany, led by Piedmont and Prussia, and the one failed in Norden – ‘not’ led by Sweden, here seen as the core of Norden (Mouritzen 2001, Østergård 1997, Joenniemi 1997). 15 This dynamic can be observed in the somewhat natural ‘division of labor’ among the Nordic countries, each one of them deepening cooperation with a single Baltic ‘buddy’ (Denmark is traditionally associated with Lithuania; Sweden with Latvia; Finland with Estonia; and Norway to NW Russian regions, cf. van Deveer and Dalbelko 1999). Despite being rather markedly state-centric, this process is quite emblematic to exemplify Nordic altruism, and yet its deliberate focus on maintaining the Nordic and Baltic projects separated. Chapter 7 will discuss this relationship further, particularly in relation to the case of environmental cooperation. 16 This label instead fits with the arguments about the differentiation between macro and meso regions. One could actually add that in the early 1990s, the BSR was far from being a sub-region of the EU-Europe, to the same extent as, for instance, Scandinavia was. It was rather ‘in between’ EU and ‘Russian Europe’ (see Chapter 6 below). 17 For a review of these and other non-governmental institutions, cf. Suominen et al. 2000. 18 As was the case at the conference “Regional Identity Under the Soviet Rule: the Case of the Baltic States” in Kiel in 1987; the conference in Kotka, Finland in the Summer of 1990 (Joenniemi and Wæver 1992); the workshop: “A Region in the Making?” in Tallinn in June 1991 (cf. Neumann 1994, 53), and the workshop: “The Baltic Sea Region: Conflict or Cooperation?” in Kiel in September 1991 (cf. Wellmann 1992). 19 To this point he adds, “Why this incredible overemphasis of the role of intellectuals? […] Can the intellectuals really do this much? No, the intellectuals are not the root cause, but again to the extent that a regionalisation around the Baltic Sea is at all possible, it is going to follow this track. Step one is the myth making of the intellectuals” (Wæver 1993, 48). 20 Quite revealingly, Pertti Joenniemi and Ole Wæver introduced their report prepared for the Presidium of the Nordic Council for the 2nd Parliamentary Conference on Cooperation in the Baltic Sea with the section: “The Baltic Sea Region in 2008” (Joenniemi & Wæver 1992, 2). 21 “The security problem (security understood in societal rather than territorial terms) turns into the price of not participating, into issues of centrality and distance” (Joenniemi and Wæver 1993, 4). 22 “Security, in the classical sense, is less significant than it used to be” (Joenniemi and Wæver 1993, 2; see also Wæver 1997, 309). 23 Later, in 1993, Ole Wæver summarises the concept more explicitly: “Baltic Sea cooperation in the last instance has to take a non state form” (Wæver 1993, 38) for three reasons: “1) Regarding state-to-state cooperation, others will remain more important (Nordic Council and especially the EC) 2) The German factor […] in the 141 First Phase: The Rise Baltic will not be the Federal Republic of Germany (FRG) but the Länder, together with firms, organisations, towns, and individuals; 3) a problem of unequal units: how are we to have a close and balanced cooperation with big states like Russia, Poland, and the FRG on the one hand and small ones like Estonia and Denmark on the other?” (Wæver 1993, 39). 24 Markus Perkmann, when distinguishing typologies of Euroregions, identifies a proper “Scandinavian grouping”, because of the number of sub-regional arrangements that were created since the 1960s (such as the Öresundkommittén since 1964; the Finnish-Swedish Skärgården since 1978; or the Danish-Swedish so-called ‘Four Corners Co-operation’ (since 1980) (Perkmann 2002). 25 The typical example here is the Öresund experience (cf. Maskell 1999; Lyck & Berg 1997). 26 On the other hand, one may detect an echo of the theorisations of the region builders when considering the membership criteria of this organisation: “Different approaches to define what a Baltic Sea Region city is are possible. First, in sensu stricte, the Baltic Sea Region city is located only in the sub-regions adjacent to the Baltic Sea. In sensu largo, the Baltic Sea Region city is located in the sub-regions and states which are not located directly on the sea but who are traditionally, economically and culturally connected with the region” (Union of the Baltic Cities, 2003). 27 Making the story even more confusing is the circumstance that the Baltic Assembly and the Baltic Council of Ministers were preceded by Council of the Baltic States in 1990, which was symbolically intended to reconnect the cooperation among the three Baltics to the interwar period. It has nothing to do with the Council of the Baltic Sea States discussed in the next section. 28 This poses significant questions regarding the post-enlargements future of the Organisation, as will be discussed in the Third Phase. 29 Or ‘civil security’, following the denomination given by the CBSS to this kind of issues (cf. Council of the Baltic Sea States, 1996). 142 The Framing: States 5 The Framing: States The objective of this chapter is to analyse how the countries in the BSR have dealt with regional security issues and the nature of the ensuing regional cooperation. Speeches and statements issued by leading foreign policy makers are indicative of the defining trend in the security policies of a country with respect to the BSR and are therefore the most significant sources for analysis. These texts will span diachronically in order to observe whether change has occurred during this Second Phase, the extent to which change has occurred, and the manner in which it has occurred. I also wish to emphasise that the purpose of this aspect of the analysis is not to provide an account of the foreign policy doctrines of each Baltic Sea state. The purpose is rather to outline a pattern capable of indicating the character of the regionalism promoted by the security policies of the Baltic Sea states during the Second Phase. Hence, the analysis of these discourses regarding the region will provide a rather impressionistic depiction enabling the placement of state actors within my analytical framework and therein also the regional dynamic. A few explanatory remarks should therefore be made to account for the manner in which I intend to carry this assessment out. At the more conceptual level, the old behaviouralist foreign policy analysis, which could serve as a starting point for discussion of state security policies, is an IR sub-field in which rationalist and state-centric approaches have traditionally been dominant (White 1999, 38-39). However, constructing the analysis upon a continuum such as the one characterising the analytical framework in the project at hand, I will also include more recent and post-positivistic elements of foreign policy analysis that focus more prominently on identity-based arguments and their impact on the formulation of national security. Hence, in order to preserve the comprehensive character of the analytical continuum – but also to retain the ‘lightness’ of this impressionistic analysis of security policies – I will build the discussion upon three basic pillars. These three pillars (freely inspired by Wæver 1998a) represent the ‘internal’ core security values of the 143 After 2004 knowledge at the region-wide level, enhances networking and create contact. The Baltic Sea is the meeting-place that makes this possible and thus enhances a ‘thin’ degree of identification. In Grotian terms, the BSR provides a less-than-universal space for enhancing (quasi-) universal values. Such argument raises the point of what exactly constitutes these values, and whether such ‘quasi-universal’ values may end up fostering ‘otherness’. In the Baltic region, admittedly, there are indeed ‘Westernising’ elements in the conduct of education and research, which to some extent have continued the othering of Russia. Nevertheless, the Grotian point is that there are also values that go beyond cultural markers and are indeed (quasi-)universal, or European, in the inclusive as opposed to the EU-centric sense of the word. Hence, the lesson suggested by a ‘soft’ and originally functional field such as the environment is that culture and education may not necessarily embody a goal such as that of Baltic Sea identity building. On the other hand, they may be instrumental for enhancing inclusiveness and overcoming ‘otherings’ by fostering a wholeEuropean dialogue20 (see Henningsen 2002; Parts 2004). One final note regarding the wider lesson that can be drawn from the ‘bottom-up’ dimension of environmental security is the place occupied by intellectuals, who, it was noted, played a central role in the rise of the regional discourse in the early 1990s. The role of intellectuals and experts, i.e. the epistemic community, is central in a liberalist-like framework to “decrease uncertainty about the effectiveness of policy initiatives” (Buzan et al. 1998, 73). Within this context, therefore, epistemic communities are particularly powerful in terms of influencing strategies and decision-making. The case of environmental security in the BSR, in this respect, makes not exception (see Hjorth 1994). However, epistemic communities have a particularly peculiar role in a Grotian-like post-structural liberalism. In a framework in which culture and education do not generate identity but a ‘thin’ identification, epistemic communities not only play a role in influencing the political decision-making, but also in enhancing this light awareness – e.g. through networking or training activities. In this sense, intellectuals and experts have a role of their own. This role is similar to that of Scandinavian region-builders, but not entirely the same, since in this post-structural liberalist framework, intellectuals are not exactly region building actors themselves. They enhance regional 246 After 2004 awareness by implementing certain research-policy strategies, which they may have contributed to endorse. At the same time, they contribute to shape policy-making based on the needs emerging from the grassroots dimension they are in contact with. Hence, intellectuals are rather instrumental to connect the institutional actors and the grassroots and to enhance their complementarity; however, they remain somewhat outside of the construction of the region as such. The second element deriving from the environmental case involves the ‘top-down’ institutional dimension and may refer to the more functionalistic aspect of the Grotian tradition. The case suggests that coordination is not only vertical (‘top-down’ and ‘bottom-up’ actors, but also, and to some extent more importantly, horizontal. One of the greatest question marks posed by the institutional inclusiveness in the BSR is the existence of a large number of institutions doing similar things and thus possibly overlapping, e.g. the Nordic Council of Ministers and the Nordic Council, the Baltic Assembly and the Baltic Council, the CBSS. As Michael Emerson has written, “Evaluation of these […] mechanisms […] is a real challenge. What do they all do? How do they relate? Where do they overlap or complement each other?” (Emerson 2001, 41). The issue is complicated further by the existence of external initiatives and various projects focusing on different sectors (the most comprehensive ones being the NDI and the NEI). The result is that both financial and human resources are fragmented and the potentials of such wide participation end up becoming an organisational burden. Officials often lament the excessive amount of obligations required by such a high number of institutions21. To quote Chris Patten (2001, 3): “I have to confess that, given the unavoidable limits on human and financial resources we all face […], we have been encouraging the various regional bodies to consider further co-ordination and streamlining of their respective activities and to look at the possibility of developing a new “division of labour” between them. I am sure that co-operation in the Baltic region, and the implementation of our Northern Dimension policies could benefit from such rationalisation: more results, less structures and fewer meetings.”22 247 After 2004 The case of the environment suggests that a rationalisation of institutions, and thus horizontal coordination among the organisations, may be beneficial to the effectiveness of the cooperation in which each organisation is involved. This is what the NDEP has essentially been about. This partnership suggests that a consortium of institutions – which in different ways have long been active in the region – can pool resources and focus on specific objectives. The result, although the endeavour is too young to be evaluated comprehensively, has been the active inclusion of Russia. The wider lesson of this element provided by the environmental case primarily relates to the most representative regional organisation in the BSR, the CBSS. It can be argued that during its decade of existence, the CBSS has represented a significant landmark, not only in terms of the cooperation within the region, but also for the message of peace that this region has aimed to send to the whole continent. Since 1992 the Council has shown that it was possible to overcome the legacies of the Cold War and to make Russia and the Baltic States communicate a mere couple of months after the formal dismantling of the Soviet Union. Nonetheless, the post-enlargement scenario poses a different set of challenges. The full-fledged participation of Russia (as well as Norway and Iceland, two other non-EU members) certainly continues to represent an element maintaining the significance of this Council. As does the experience gained from over ten years of increasingly structured cooperation. However, the EU membership of four ‘New European’ countries and the enhancement of the EU presence in the region (i.e. the NDI and, possibly the Wider EU strategy vis-à-vis its neighbours, see the Conclusions of this study) will contribute to the accomplishment of some of the current primary CBSS goals. This requires a streamlining of this institution. Analysts and policy-makers alike have been raising enquiry regarding the question of the future of the CBSS (cf. Gänzle and Hubel 2001; Ojanen 2002; Tassinari 2002b; Stig Møller 2002; Ojuland 2002 among others). The Council, in view of its increased institutionalisation during the 1990s, is now more an actual organisation than a dialogue forum. It has engaged in a variety of different issues and has progressively grown in structure and complexity. The overlap of these activities with the other institutional 248 After 2004 bodies is self-evident. This is not necessarily perceived as a negative development, however, and it has actually been referred to as a ‘positive overlap’ (Gänzle and Hubel 2001, 18). According to such a view, the CBSS may support and complete the work of the other larger organisations in detail (mainly the EU, but also OSCE, and the Organisation for Economic Co-operation and Development (OECD)23 and would thereby become a sort of implementing agency of the softsecurity initiatives of other institutions (cf. Ojanen 2002a). However, recent developments indicate that this alleged ‘division of labour’ in terms of implementation remains somewhat vague24. The CBSS, despite acting through its various working bodies at a rather local level (see Ojanen 2002a), has thus far focused on “large-scale network” (quoted in Gänzle and Hubel 2001, 20) and it is not designed to act as an actual ‘implementing agency’. In view of the post-structural liberalist model proposed in my case study, it could be argued that institutional multiplicity is to be enhanced when it is meaningful, i.e. when there is a niche for each actor that can be coordinated with the others. Accordingly, a post-2004 streamlining of the CBSS could indicate a sort of ‘return to the origins’. This implies that the Council is unlikely to be an EU implementing agency or, to quote Igor Ivanov, “the primary coordinator of intergovernmental cooperation in the Baltic Sea region” (Ivanov 2001, 5). It may rather remain a forum for dialogue and consultation between EU members and non-members where, in Joschka Fischer’s words: “the bulk of practical, day-to-day relations between EU and Russia will take place” (Fischer 2001, 1, see also Interview Borisovas 2003, 2). The other option is a more radical restructuring of all of the regional institutions25. Some have voiced the possibility of at least reducing the number of institutions that are not pan-regional, i.e. those that group Nordics and Baltics, following the transformation of ‘Nordic 5 + Baltic 3’ into the “Nordic Baltic 8 cooperation” (Interview Borisovas 2003, 2). Moreover, the hypothesis that the CBSS itself could be discontinued has also time and again been raised (cf. the report of a recent seminar at the Danish Institute of International Studies by Tassinari 2004a), supported by the fact that the mandate of the CBSS Commissioner for Democratic Development was terminated in December 2003. Finally, there is the question of formalising 249 After 2004 coordination among the plethora of institutional actors in Northern Europe (cf. interview Arz Von Straussenburg 2004). On this note, one rather drastic idea could be that existing institutions (e.g. Nordic Council of Ministers, the Baltic Assembly, the CBSS etc.) agree upon the establishment of some kind of ‘Northern European Council’ following the example indicated by the NDEP. This new subject would serve as an umbrella-institution for the work of the various organisations, and thus coordinate the overlapping efforts at the financial, managerial, and operational levels. Moreover, it would become the only institutional referent vis-à-vis the policies implemented by the EU in the region. In sum, the liberalist trend characterising pre-2004 regional cooperation in the Baltic Sea envisaged in the close of Chapter 6 (relatively state-centric, inclusive and focusing on non-military security) is also consistent with the model of environmental security. However, the model specifies that claim. According to my framework, enhancing regional cooperation through the securitisation of nonmilitary issues does not lead to a non-security-based ‘regionality’. Conversely, the inclusiveness of the BSR environmental security overcomes ‘otherings’, i.e. Russia, East/West disparity, and may generate a thin degree of Gemeinschaft. This is at the heart of my poststructural liberalist interpretation based on the Grotian tradition. In terms of practical outcomes with respect to the regional cooperation, environmental security indicates that the post-structural liberalist model supports all-inclusive coordination of both the civil society operating ‘from below’ (particularly as far as culture and education are concerned) and institutional actors operating ‘from above’ (in relation to a horizontal streamlining of the activities). Moreover, as to the relations between EU and Russian complexes are concerned, this Grotian model purports that the Baltic sub-complex is characterised as ‘in between’ the sub-complex types B, i.e. functional cooperation within the two complexes, and C, i.e. border-breaking regionality, that were indicated in the Third Argument of the critical RBA (figure 14 in Chapter 2). This interpretation will be visualised in the three-phase scheme in the Conclusions of this study. In order to do that, however, I now conclude this chapter by discussing whether, to what extent, and in which way the model of cooperation suggested by the case of 250 After 2004 environmental security is applicable to other sectors on the basis of the current relevant discourses on the future of regional security. 4: Applicability The classical ‘liberalist’ assessment characterising the most recent pre2004 pattern of regional cooperation delimits the range of issues in which this ‘Grotian’ interpretation of the post-2004 scenario can be suitably tested. Indeed, the post-structural liberalist model continues to postulate a correlation between security and cooperation (and thus sidelines the non-security ‘regionality’ scenario). On the other hand, this interpretation also embraces the argument deriving from the pre2004 liberalist model, which excludes significant securitising moves at the Baltic regional level in the political and military fields, which instead tend to coalesce in the wider European dimension. Hence, the applicability of the environmental case will focus on the other nonmilitary security sectors, i.e. the societal and the economic. Moreover, this reference to the European complex suggests a first, rather self-evident argument with respect to the applicability of the post-structural liberalist model. The environmental case, it was argued, is also unique because the territorial and conceptual construction of the regional security fits with the Baltic regional framework, i.e. the BSR is an environmental security complex. The observation here is that, as opposed to other aspects in which the environmental model may be followed (e.g. inclusiveness, coordination etc.), it remains an exception in this one element. Environmental security is functional to the features of the physical environment, and the two coincide in the BSR. For most other issues, however, the two do not coincide. To quote Igor Ivanov, “solution to many problems of the Baltic Sea region is unthinkable outside the context of mainstream European processes and thus requires a comprehensive approach” (Ivanov 2001, 3). This argument recalls one of the concluding remarks of the pre-2004 scenario in which the European complex was distinguished from the Baltic Sea sub-complex. That distinction will be specified in the discussion at hand. Since the territorial ‘coincidence’ between regional security interdependence and the regional framework of the environmental case cannot apply to the other non-military security, I will rather discuss whether, to what extent, and in which way the other 251 After 2004 conceptual features of my ‘Grotian’ model apply to other non-military sub-complexes in the post-enlargement scenario. Civil Security The notion of ‘civil security’ deserves explanation. The argument was raised when presenting the RSCT and its different analytical sectors that the concept and applications of societal security refer to ‘identity’. This implies that the buzzword of societal security is ‘nation’, which is “not the state as either a government or a territorial entity but the shared identity which constitutes a common social ‘we’ and accordingly the security of a society can be threatened by whatever puts its we identity into jeopardy” (Williams 1998, 435). In fact, societal security is not an actual security sector unto itself, but rather, “a trendy response to current concerns about nationalism” (Buzan and Wæver 1997, 241). This opens up to questions about what it is that qualifies an identity and the nature of threats one is dealing with. The radical constructivist focus of the Copenhagen School framework sparked an animated discussion during the mid-1990s concerning the definitions of ‘identity security’ in the RSCT. In such a context, interpretations of identity may vary considerably, ranging from the more individualistic ones, e.g. family, ethnic group, to the more objectivist ones, e.g. the nation-state (cf. McSweneey 1996; Buzan and Wæver 1997; Williams 1998). Regardless, the adherence of the notion of ‘societal’ with identity implies that “the referent objects in the societal sector are whatever larger groups carry the loyalties and devotion of subjects in a form and to a degree that can create a socially powerful argument that this “we” is threatened” (Buzan et al. 1998, 123). The ‘logic of threats’ in societal security will then depend on the different possible constructions of identity. The European case provides a fine illustration of this variety of options (upon which the ‘Copenhagen School’ made its case for defining societal security, in Wæver et al. 1993). Multilevel governance generates multilevel identities and thereby multilevel societal securitisations, e.g. migrations, minorities, etc., in the European case. The point here is that ‘identity security’ in the Baltic Sea area has coalesced at various levels: the ethnic level, e.g. in Latvia and Estonia vis-à-vis Russia, the nation-state or, in a somewhat more complex manner, the European level. My assessment has excluded the 252 After 2004 applicability of regional identity discourses to the BSR proper, making it implausible to talk about a ‘societal security’ in the region. However, it is not uncommon for some of the threats that are sometimes ‘talked into existence’ as societal security, e.g. migrations, minority issues, as well as crime prevention and health issues, to be dealt with at the Baltic Sea-regional level26. This is to say that while the absence of identity prevents these regional security moves from being constructed as ‘societal’, the ‘logic of threats’ is quite consistent with it. This explains the label civil security, which avoids confusion with the terminology employed in the analytical framework. It is also somewhat explanatory in terms of the nature of the threats I am dealing with. Moreover, the term is also consistent with the notion attributed to these kinds of threats by the CBSS since 1996 (see CBSS website)27. Civil security issues have been at the forefront of the current dominant discourses on regional cooperation (Tuomioja 2003; Söder 2003; Cimoszewicz 2003; Stig Møller 2003a; Lindh 2002; Kwasniewski 2002a, 10; Heimsoeth 2000). Paavo Lipponen, for one, pointed at the Baltic regional cooperation in the field of crime prevention as a model for the entire EU: “The objectives set for cooperation in justice and home affairs can be achieved only through comprehensive action. The European Council therefore approved political guidelines for strengthening the Union’s external action to achieve these objectives. The results achieved by cooperation between the countries of the Baltic Sea region provide an example of practical regional cooperation in fighting organized crime.” (Lipponen 1999a, 4) Moreover, the manner in which current priorities are set on the regional agenda responds to a criterion that is not dissimilar to the one presented in the example of the environment. As for the environmental case, a review of the civil security problems and priorities in the BSR is beyond the scope of the present discussion, which is not on policies but on securitisation practices28. Nevertheless, one cannot fail to note that much of the current debate has been developing along the same lines indicated by the environmental model, – i.e. East/West disparities and the need to engage Russia. On the one hand, civil security threats such as crime, health problems, and corruption stem most prominently 253 After 2004 from the southeastern Baltic rim (see Moroff 2002b). The regional securitisation of these issues has also highlighted a dual pattern, as noted by a working party of the Council of the European Union on these matters: “cooperation so far appears to be mainly bilateral” (quoted in Moroff 2002b, 225). On the other hand, the current dominant debate concerning the post-enlargement scenario envisions Russian inclusion as the most exacting challenge, also as far as civil security is concerned, with particular focus on illegal migration, infectious diseases and organised crime (see Pursianen 2002; Sergounin 2003a, 68-73), as the case limit of Kaliningrad confirms29. These brief notes suggest that the post-2004 pattern of regional cooperation in the field of civil security may be consistent with the post-structural liberalist model indicated by the environmental case. There are arguments against and in favour of such a claim. On the one hand, the wider lesson of the environmental field stresses a ‘Grotian’ kind of regional inclusiveness and multiplicity. As far as civil security in the BSR is concerned, this multiplicity is rather questionable. The reason why this is the case can be reconnected to the distinction drawn above between ‘societal’ and ‘civil’ security. Initiatives ‘from below’ are more likely to focus on those referent objects where the ‘identity security’ is perceived as relevant, e.g. the individual, ethnic, national and, possibly, European levels, and are somewhat weak at the regional level (Gänzle and Kungla 2002, 60)30. On the other hand, the post-2004 scenario for civil security in the BSR is consistent with the environmental model as to the need for horizontal coordination of institutional actors. A plethora of organisations has been dealing with the various issues, which are included here under the oversimplifying heading of civil security31. These initiatives are “widespread but fragmented” (see Pursianen 2002, 135-142). Quite significantly, however, the pattern of coordination inaugurated by the environmental field has recently taken root, and the model launched by the NDEP has been swiftly followed by its civil security homologue. A “Northern Dimension Partnership in Public Health and Social Wellbeing” has been established in October 200332, and has been hailed by policy makers as one of the most prominent initiatives in the field thus far33. 254 After 2004 A Regional Economy? This sub-section presents a sketch of how and to what extent current economic security talk referring to the post-2004 scenario affects regional economic cooperation. Economic security is a rather contested concept, as was argued in Chapter 2. The very reference to threats and risks at the centre of the securitising move is controversial in the economic context, since the notion of ‘taking risks’ is at the heart of market economy (Buzan et al. 1998, 95). This is particularly relevant within the liberalist setting, e.g. that which closed my pre-2004 assessment, since liberalist approaches place economics at the centre of the social and political structure and, consequently, “triggers survival issues in all of the other sectors” (Buzan et al. 1998, 116, see also Mansfield and Milner 1997). Without making the conceptual debate too weighty (for this see Buzan et al. 1998 95-117), this brief note suggests that the controversy regarding the concept of economic security allows a certain degree of flexibility and, to some extent, may necessitate a “manufacturing” (Joenniemi and Browning 2004, 17) of the way the concept of security is formed and employed. Economic issues are generally very high on the regional agenda. They are constantly prioritised in the dominant recent discourses as one of the overall goals of the post-2004 development of the regional cooperation (Ellemann-Jensen 2003a; 2002a and 2002b; Stig Møller 2003; 2003a; 2002; Ojuland 2002; Valionis 2002). Securitisation is not really practiced in this case, and the economic becomes politicised. On the other hand, it is not uncommon to find expressions in the very same discourses such as “it is time to move!” (Ellemann-Jensen 2002a, 7); “time is short” (Danish Ministry of Trade and Industry 2001, 2); or “we cannot rest on our laurels” (Ellemann-Jensen 2002b, 8). These expressions indicate a rather explicit sense of urgency. That which the dominant actors are communicating in this setting is that the management of economic issues should be taken somewhat beyond the routine politicisation of the economic cooperation, i.e. taken to a level where action must be taken swiftly. This is where the ‘manufacturing’ of economic security applies. This urgency is not motivated by threat or danger. Despite flourishing economic cooperation in the BSR and steadily increasing intra-regional interaction, the underlying message is that there are unexpressed resources that the region must exploit in order to enhance its competitiveness in the context of the enlarged 255 After 2004 7 “We attach special importance to the co-operation of the northern countries in environmental protection. […] It would be well to extend joint measures for protecting the marine environment of the Baltic, now being carried out by a commission of seven maritime states” (Gorbachev 1987). I thank Carl Einar Stålvant for pointing this out to me. 8 Just to mention a few illustrative cases: Coalition Clean Baltic was founded in 1990 and groups 27 member NGOs spread throughout the nine Baltic Sea countries; the Baltic University Programme focuses its educational activities on environment and sustainability; UNESCO-sponsored Baltic Sea Project focuses on grassroots education and awareness raising campaigns on the environmental problems of the Baltic. At the level of sub-regional cooperation, Euroregion Baltic, BSSSC and the UBC all have environmental programmes. 9 The following two observations draw heavily from Tassinari and Williams 2003. 10 In April 1992, a new Convention for the protection of the Sea was adopted within the framework of HELCOM. The Convention generated a Baltic Sea Joint Comprehensive Action Programme that remains the most comprehensive strategy to combat environmental threats in the Baltic Sea (Helsinki Commission 1993). The CBSS also proved responsive to the environmental threats and in June 1998 adopted an “Agenda 21 for the Baltic Sea Region” to support sustainable development in the region (see Baltic 21, 1998). Another useful framework for intergovernmental cooperation is “Vision & Strategies around the Baltic 2010,” established in 1992 to deal with spatial development planning. At the EU level, the NDI channelled funds from Tacis and PHARE programmes (about EUR 55 million between 1994 and 2000) for environmental projects in the BSR. Finally, the NEI, UNDP and some International Financial Institutions (e.g. the World Bank) provided funding and expertise through projects specifically targeting the BSR (see Bober 2002, 3-7; Tassinari and Williams 2003, 317-322). 11 Given the highly interdisciplinary nature of the environmental field, a quantitative type of analysis would be more suitable to demonstrate this claim. This, for instance, has been done by applying the so-called Environmental Sustainability Index (Ciesin, 2002), which is a widely recognised tool for evaluating environmental performances on the basis of five indicators: ‘Reducing Human Vulnerability’, ‘Reducing Stresses’, ‘Environmental Systems’, ‘Global Stewardship’, ‘Social and Institutional Capacity’. Applying this categorisation to the nine BSR countries (see Tassinari and Williams 2003, 321-322) suggests that in categories such as ‘Social and Institutional Capacity’ and ‘Global Stewardship’, the Nordics and Germany perform far better than do the Baltics, Poland and Russia. On the other hand, in categories such as ‘Reducing Stresses’, the three Baltic States and Russia score higher standards than the other five countries (ibid.). 12 About two thirds of the Russian nuclear powered vessels are placed in North-West Russia. Moreover, a number of nuclear power plants, such as Sosnovy Bor, close to St. Petersburg, are of the so-called RBMK type, the same that caused the 1986 Chernobyl disaster. These plants provide about 40% of the total energy supply in North-West Russia and are not going to close down (cf. Pursianen 2002, 108-109), 262 After 2004 despite two of the same kind of plants in Ignalina, Lithuania being scheduled for dismantling in 2005 and 2009. 13 The pollution of the Neva River is often indicated among the most serious threats. This, together with the insufficient treatment of the municipal wastewater, makes St. Petersburg ‘the single most important point source’ of pollution in the Baltic Sea area (NDEP website). 14One could mention the 1995 arrest with the accusation of treason of Alexander Nikitin – the author of a report on the aforementioned highly nuclear-powered Russian Northern Fleet (see VanDeveer and Dabelko 1999, 241); or Moscow’s reluctance to ratify the Kyoto Protocol on Climate Change. 15 The rationale behind this initiative was introduced at the Northern Dimension Ministerial conference in Luxembourg in April 2001 to form a working group on the environment including the European Commission and the financial institutions operating in the region: the European Bank for Reconstruction and Development (EBRD), European Investment Bank (EIB), the Nordic Investment Bank (NIB) and the World Bank. Other contributors include all of the Nordic countries (including Norway) (see NDEP website). 16 The EU Council in Gothenburg invited the partner institutions to call a ‘pledging conference’ to create the NDEP Support Fund. After having identified twelve initial projects and the working rules to establish the support fund, the July 2002 pledging conference received about 110 million EUR (NDEP website). 17 This element emerged from several discussions involving the author in Brussels. 18 This argument is somewhat the opposite of the one put forward by the early Scandinavian region builders. Ole Wæver, for instance, constructed a two-phase model of the regional economic cooperation (to which I will return towards the end of this chapter). In this model, a ‘Phase 1’ according to which “Self-referential cultural process creates regional identity” precedes a ‘Phase 2’, where “Self-referential economic process creates regional economy” (Wæver 1997, 307). 19 See also Hubel 1999, 251-254 and Stadsholt 2001 for applications of the Grotian tradition to the Baltic Sea case. 20 The strong focus of the new NDI Action Plan on ‘knowledge society’ in the region may confirm this pattern (see also Musial 2002a). The argument is that “Northern Dimension partners will therefore work to [promote] a dialogue that enables the region to develop as a dynamic and open cultural area” (Council of the European Union 2003, 8). Moreover, in relation to the linkage between security and regional knowledge, “scientific research is vital to the development of the region with the key objective being to ensure that research activity is effectively co-ordinated and recognizes the impact of ongoing economic, environmental, political and social change in the region” (Council of the European Union 2003, 8-9). 21 This argument emerged from discussions between the author and officials in Brussels. 22 Also Russia expressed a similar concern: “Also urgent is the need to improve interaction between the CBSS and other regional cooperation structures” (Ivanov 2001, 4, see also Ivanov 2002, 2). Moreover, this need for more coordinated 263 After 2004 participation of the regional institutions within the NDI is also explicitly stated in the Second NDI Action Plan (Council of the European Union 2003, 3). 23 This argument emerged from discussions between the author and officials in Brussels. 24 For instance, Hubel and Gänzle argued that the CBSS Commissioner for Democratic Development could progressively assume the OSCE duties in Estonia and Latvia (Gänzle and Hubel 2001, 19). Yet, as noted, the Commissioner’s mandate has recently been discontinued. 25 Paavo Lipponen implied a similar argument in a recent speech in Oslo (see Aftenposten 2004). 26 Nor is it uncommon for these threats to be referred to – indiscriminately and thus erroneously – as societal (see for instance Gänzle and Kungla 2003, 58). 27 The CBSS itself uses the terms civil security and ‘civic’ security without distinguishing between them (see also Stålvant 1999, 65). 28 For a comprehensive review of the issues, see Moroff 2002b; Pursianen 2002; Gänzle and Hubel 2001. 29 The fact that the need for Russian engagement is very high on the agenda of the non-military security debate among practitioners and researchers alike may be confirmed by the fact that the London-based International Institute for Strategic Studies, which is traditionally considered to be a reliable source of the mainstream debate, has recently launched a new journal, Russian Regional Perspectives. The first issue was for the most part dedicated to soft security cooperation with the northwestern Russian regions. 30 Outgoing CBSS Commissioner for Democratic Development Helle Degn has instead praised recent progress in this respect (Degn 2004). 31 The CBSS Task Force on Organised Crime and the Task Force on Communicable Diseases; the Baltic Sea Network on Occupational Health and Safety; the OECD Baltic Regional Programme; and the World Health Organisation are just a few of the organisations engaged in civil security issues (see CBSS Website and Gänzle and Hubel 2001). 32 The Partnership includes all of the Baltic Sea countries together with Norway, Iceland, Canada, Spain, France, the World Health Organisation, the CBSS, and the EU Commission (see Baltinfo 61, 2003). 33 This opinion was expressed in discussions involving the author in Brussels (see also Lipponen 2002, Doyle Ong 2003). 34 For the specifics of the data see Danish Ministry of Trade and Industry 2001; Dansk Industri 2003; Fredriksson 2001. 35 See also the article in the Baltic Times: “Lithuania Fears Becoming Nordic Investment Colony” (in The Baltic Times March 29 2001). 36 “Challenge Number One is infrastructure,” which “must be addressed now” and “the European Commission and the EU members must appreciate the priority of our proposals,” argued President Adamkus with respect to existing problems for the regional economy (Adamkus 2002, 1-2 emphases added). 264 After 2004 37 Among the several bodies dealing with regional economic cooperation: e.g. the Baltic Development Forum, the Baltic Chambers of Commerce Association and the Business Advisory Council, just to mention a few. 38 An exception to this claim would be those emergencies, e.g. nuclear disasters, which would be securitised at a quasi-global level. 265 Conclusions Conclusions Many European initiatives […] attest to the recognition of the strong potential of our region and its significant role in ensuring the stability and prosperity of the entire continent (Vladimir Putin, 2001) The Baltic area […] encompasses virtually the entire spectrum of Europe’s present and future (Günter Verheugen, 2002) The main objective of this study was to assess the development of the regional cooperation in the Baltic Sea area through the analysis of regional security interdependence from the end of the Cold War to the present, and to envisage a likely post-2004 scenario. As presented in the Introduction, three research questions were attached to this formulation of the problem. The answers to the first two research questions have been provided in the previous chapters and will be summarised in the following section. As far as the third research question is concerned, i.e. the wider relevance of the Baltic regional case for EU foreign policy and EU-Russia relations in particular, important elements were discussed in the previous chapters as well. These will be systematised and expanded in the second section of this chapter, which concludes this analysis. 1: Conclusions How can the analysis of regional security be correlated to regional cooperation in theoretical terms? Exploring the correlation between security analysis and regional cooperation (regionalism) has constituted the core of the analytical framework of this study. This has been conducted by putting forward a ‘critical’ Region Building Approach (RBA). This framework is modelled upon its ‘classical’ version, as elaborated by Iver Neumann (Neumann 1992; 1994), and the analytical framework for regional security elaborated by the so-called Copenhagen School. The latter is also referred to in this study as ‘critical’ Regional Security Complex Theory (from which the label ‘critical RBA’ derives). 266 Conclusions I reviewed and systematised regionalist approaches in International Relations theory in the first chapter, using the notions of ‘old’ and ‘new’ regions (Hettne 1994, 2) as my inspirational point of departure. These notions have proven to serve as a suitable starting point for organising approaches in view of the analytical framework of the study. I therefore constructed a bi-dimensional matrix containing both ‘new’ and ‘old’ regionalisms: the one dimension of the matrix follows the regional actorness, the other marks the levels of analysis. The former dimension is characterised by a spectrum spanning from ‘bottom-up’ approaches to ‘top-down’ approaches. The second dimension analyses the systemic dynamics in which the regional phenomena are to be contained and is characterised by a spectrum spanning from ‘insideout’ approaches to ‘outside-in’ approaches. Such formulation explicitly recalls one argument contained in the classical RBA, i.e. the ‘continuum’ of approaches to regionalism, but specifies it, adding one dynamic: the regional actorness. By adding this other dimension, I intend to mark my choice of categorising approaches to regionalism according to their various theoretical characterisations rather than only according to levels of analysis. I associated ‘old’ regionalisms with neo-realist and liberal approaches. The former approach focuses on systemic changes. It views regions as an outcome of the anarchic international arena, and regionalism is associated with central pillars of neo-realist thinking, e.g. alliance building and hegemony. The state represents the only, rational and selfish actor in the regional phenomena. I then sketched two representative trends within the liberal tradition: neo-functionalism and liberal institutionalism. Neo-functionalism focuses on the internal dynamics of regionalism and, in view of its focus on states and elites, can be qualified as a top-down approach. According to neo-functionalists, the collective action of states is at the centre of the regional phenomenon. Their action produces consequences, i.e. spillovers, which affect integration. Liberal institutionalists retain the traditionalist focus on the state as the only rational actor in the international area and maintain that the formation of regions is to be explained on the basis of the need of the states to maximise their interests. The coordination of international action is then viewed as a means by which to cut transaction costs, and institutionalised (regional) cooperation is viewed as a likely outcome. 267 Conclusions I also selected two interpretations in relation to new regionalism: globalisation and critical theories. The focus of globalisation is systemic and the reflections have been upon the changes brought about by globalisation regarding territoriality and power. Post-Westphalian deterritorialisation implies a reframing of the relationship between power and space, and thereby, reterritorialisation. Regionalism then emerges as a possible form of reterritorialisation, which is compatible and not at odds with globalisation. Critical theories have then been introduced. These also focus on this ‘post’- dimension of power and territory: regions are conceptualised in a manner that goes beyond traditionalist spatial canons and focuses on cohesion, we-feeling and identity. As opposed to the traditional rationalist claim, these have been depicted (perhaps simplistically) as ‘inside-out’, and ‘bottom-up’. A rather typical way to frame critical theories to regionalism is to discuss the concept and characterisations of Karl Deutsch’s security community and its later elaborations (Adler and Barnett 1998), which then return in the context of the actual framework of this study (i.e. the Second Argument of the critical RBA). The bi-dimensional matrix is brought back to a one-dimensional continuum à la Neumann in the end of Chapter 1. This is less explanatory but more manageable in relation to the framework. My continuum spans from ‘old’ to ‘new’ regionalisms, instead of from outside-in to inside-out approaches as in the classical RBA. The ‘oldnew’ continuum allows me to conceptually retain the peculiarity of both the characterisation of actors and the levels of analysis, since it derives from my bi-dimensional matrix. In view of this basic framework, I proceed to introduce the major trends in security studies in Chapter 2. First, the shift from the ‘narrow’ focus of realist security studies (i.e. on military and political sectors) to the ‘wider’ liberalist agenda (which includes other ‘softer’ issues) is explained. Both approaches, despite diverging in the characterisation of the security agenda, are state-centric. The newer schools of security studies are then introduced. The focus is initially on some radically constructivist approaches, e.g. Critical Security Studies (CSS). A few of the shortcomings of this approach are discussed – particularly the inability of CSS to delimit the scope and focus of security – in order to 268 Conclusions introduce the approach to security studies considered most suitable for the chosen framework of analysis, the so-called ‘Copenhagen School’. Like CSS, the Copenhagen School is also characterised by a radically constructivist epistemology and is to be included among the wideners in security studies. The focus, however, is first on the reconceptualisation of security. In this respect, the Copenhagen School defines security in terms of ‘speech act’, which determines the concept of security based on the very utterance of the word. The focus on discourses and textual analysis then follows as the crucial methodological and ontological tool for deciphering the speech act. The utterance is not generally sufficient to qualify an issue as a security issue. In order to be so, the issue must be endorsed by a leader – who practices the speech act and holds the ethical responsibility for it – and must be perceived by a collectivity. The Copenhagen School has applied these conceptual formulations – predominantly elaborated by Ole Wæver (Wæver 1995) – most successfully to the regional level of analysis, in that which is known as the Regional Security Complex Theory (RSCT), which instead follows from Barry Buzan’s original formulation (Buzan 1983, 105-115 and Buzan 1991, 186-229). Buzan’s classical RSCT retained a rather traditionalist focus on states. Nevertheless, it widened the security agenda and introduced the sectoral approach, dividing security into military, political, environmental, economic and societal sectors. The main interest of this study for the RSCT is therefore related to its critical version, which is the one resulting in the actual Copenhagen School framework (Buzan et al. 1998; Buzan and Wæver 2003). Here the characterisation of sectors remains, but not necessarily the traditionalist focus on states, which, in view of the speech act, is substituted by a wider and socially constructed characterisation of actors and levels of analysis. The latter include regions. Security regions, according to the RSCT, must be characterised by geographical proximity and by an unavoidable pattern of interdependence among the actors within the region, which spans from amity to enmity. Elaborating upon this range, a more complete continuum can be developed: traditionalist confrontation can be viewed as the ultimate enmity, and ‘security community’ as the ultimate amity. All possible forms security interdependence are contained within this spectrum, depending on the different (de)securitisation practices. 269 Conclusions The analytical framework of the study is put forward on the basis of these reviews of approaches to regionalism and security. I begin by completing the presentation of Iver Neumann’s classical RBA. At the centre of the classical RBA lies the eminently constructivist nature of the region building. According to this approach, an analyst is supposed to unravel the genealogy of the region – i.e. ‘whose region’ the regional construction refers to – in order to understand how regions are characterised and, before that, ‘imagined’ by political actors. Before introducing this main feature (and in order to do so), however, Neumann argues that this new approach to regions is required in order to overcome the main limitation of all regionalist approaches, i.e. the assumption made by the ‘sovereign’ analyst as regards the regional reality. This “oversight” (Neumann 1994, 57) is epitomised by the “most useful” (Neumann 1992, 12) of all regionalist approaches, i.e. Buzan’s RSCT. Buzan’s framework is particularly useful because his amity/enmity pattern combines (in Neumann 1992, 13) inside-out and outside-in approaches and therefore resembles the continuum of approaches introduced above; however, the RSCT also takes the regional reality for granted. This reference to the RSCT is central for defining the critical RBA. While my framework retains the constructivist core of the classical RBA, it further elaborates on three arguments. First, the critical RBA shares with Neumann the usefulness of Buzan’s framework in combining different approaches. The constructivist epistemology of the critical RSCT introduced by the ‘Copenhagen School’ helps to overcome the major ‘oversight’ of Buzan’s approach (and of all other approaches to regionalism), which is that of taking regions as given. This correlation, however, raises a problem contained in the classical RBA. By associating the regionalist continuum to the security continuum, the classical RBA neglects the fact that the two processes refer to two different operations: security analysis and region-building by actors. This entails that the security continuum cannot simply be combined with the continuum of approaches to regionalism. It is argued that this correlation is only possible when it regards the one “element” (Buzan and Wæver 2003, 75) of the security analysis in which the analyst focuses on those security actors who are conscious of being part of a regional framework. In that case, the analytical 270 Conclusions construction of regional security can explain what kind of regionalism is being constructed (see figure 10 in Chapter 2). The Second Argument explains why this correlation between security and regionalism is not exclusive. This argument is accounted for by referring to the ‘security community’ end of the continuum. Security community, it is argued, is characterised by a status quo in which actors have rendered war to be an unimaginable event. This defines the most mature level of desecuritisation, a scenario in which security is in fact no longer present. Hence, because of this a-security status, regional cooperation will go beyond security and focus on other elements of a societal, cultural, economic or otherwise nature. The regionalist outcome of such a pattern is defined as ‘regionality’ (Wæver 1997; Joennniemi 1999). This focus on ‘community’ rather than ‘security’ suggests that the more security actors desecuritise, the less security will be predominant to define the regionalist outcome. The Third Argument of the critical RBA completes this correlation between security analysis and regionalism by introducing one specification about the constructivist nature of the framework. This is a rather empirical element, which is included in view of the case of the BSR, but it does serve to raise a more general point. According to the Copenhagen School, the construction of security regions is functionally defined (Buzan and Wæver 2003, 76), depending on where the security interdependence occurs. It then follows that the analytical construction of security regions “may or may not coincide” (ibid.) with the construction of regions by actors. Since the analyst in the critical RBA only deals with security actors who are aware of belonging to the region, a situation is then envisaged in which the delimitation of the region (and of the regional actors) must coincide with this functional delimitation of security regions. I applied this claim to the territorial construction of the region. The distinction between ‘macro’ and ‘meso’ regions applies here. In RSCT terms, indeed, security complexes (e.g. the EU) are viewed as ‘regions’, while smaller formations (e.g. the BSR) are viewed as transor sub-regions. The critical RBA introduces the distinction between ‘macro’ and ‘meso’ in order to not rank the subjectivity of a regional construction a priori on the basis of the ‘size’ of its security interdependence (as would be the case with the region/sub-region distinction). Moreover, the unit of sub-complexes introduced by the RSCT is retained within the framework of the critical 271 Conclusions cooperative arrangements in the BSR, which go back to the 1970s. However, the significance of this cooperation goes beyond these functional aspects. The case exemplifies the “nexus of civilization and environment” (Buzan et al. 1998, 76), which widens the referent object of this securitising move to individuals and to humanity as such. This interpretation of environmental security may explain the somewhat unusual correlation of environmental security in the BSR with the political changes occurring in the late 1980s on the eastern side of the Rim, particularly in the Baltic States. Moreover, the uniqueness of this case lies in the multiplicity of actors and the rather defined ‘division of labour’ among them. This aspect is highlighted by the current priorities of the cooperation – broadly defined in terms of East/West disparity and Russian engagement – and by the related policy responses, which foster the coordination of human and financial resources, e.g. the NDEP. In order to assess the wider applicability of this case, the environmental model was placed along the security continuum of the analytical framework and interpreted. The environmental question is representative of a trend in which inclusiveness does not lead to ‘regionality’ but at the same time goes somewhat beyond mere functional interdependence. These features characterise my case as being somewhat ‘in between’ security community and the liberalist security regime, which is most suitably described by the so-called Grotian tradition in IR. The Grotian tradition is state-centric but places strong stress on individuals in keeping with Hugo Grotius’ arguments. It postulates that states and individuals alike respond to a solidaristic view of the world (which Grotius regarded as being indicated by the natural law), which is not applicable universally, but at a more delimited level, e.g. regions. These arguments generated the more recent formation of a neo-Grotian tradition put forward by representatives of the so-called English School of IR and by a number of regime theorists. This neo-Grotian tradition differs from the Grotian one in terms of its more marked state-centrism and because of its more positivistic and less solidaristic perspective on human interaction. The Grotian tradition as a whole, which I have labelled poststructural liberalism, is ‘in between’ the Lockean (i.e. security regime) and Kantian (i.e. security community) traditions. Hugo Grotius’ original arguments bear closer resemblance to the Kantian tradition, 278 Conclusions while those of the neo-Grotians are closer to the Lockean tradition. The environmental case contains both the Grotian (e.g. the Nordic environmentalism) and the neo-Grotian arguments (which entails a more marked functionalisation of the environmental cooperation, e.g. on the part of Russia), and is therefore suitable to embody the Grotian tradition as a whole. This comprehensiveness is well represented by the more practical lessons that can be drawn from this case in the light of this poststructural liberalist interpretation. It was argued that the environmental case combines something that Baltic regionalism has always had, i.e. inclusiveness, with something it has never had, i.e. coordination. This coordination is characterised by a natural ‘division of labour’, which is missing in other sectors, between two dimensions: the grassroots ‘bottom-up’ developments, focusing mainly on education and information, and the institutional ‘top-down’ level, which focuses on capacity building and financing. Both dimensions suggest something about the wider applicability of this case. The ‘bottom-up’ dimension suggests that the grassroots level can retain a valuable role in this post-structural liberalist model of regional cooperation. In this context, civil society does not pursue identity building, but rather enhances inclusiveness. This is observed in the role played by cultural cooperation. This is no longer culture-politics aiming at the creation of a Baltic identification, as in the framework of the Scandinavian region builders. The we-feeling it fosters may be ‘Baltic’ in the sense that the BSR is the meeting-place where knowledge is circulated, but cultural cooperation is not about fostering ‘Baltism’. Rather, regional culture represents a ‘matrix’ in which – in Grotian terms – ‘quasi-universal’, in our case ‘whole-European’, values are disseminated and deepened. This enhances a fuller inclusiveness. Moreover, intellectuals are not actual regional actors in such a framework. The intellectuals serve as ‘connecting figures’ between policy makers – whose choices they may influence and optimise, i.e. the traditional role of the epistemic communities – and the grassroots level, where their presence is instrumental to implement these policies, e.g. through teaching and training. The model provided by the environmental case also promotes the horizontal coordination of the ‘top-down’ institutional dimension, which may refer to the more functionalistic dimension of the Grotian 279 Conclusions tradition. This is particularly useful for envisioning the future of various institutional actors operating within the region. Analysts and policy makers have often highlighted the need to streamline this multiple institutional presence. According to the environmental model, regional institutional streamlining is feasible and fruitful, as the NDEP experience suggests. This is particularly relevant with respect to the CBSS, which relevant discourses underline as being one of the main institutional actors that may/should need rationalisation. In order to avoid overlapping and enhance coordination, the Council in the post2004 scenario is envisaged in this study as being less an implementing organisation and more a forum for dialogue between EU members and non-members, as was envisioned at the time of its establishment. Following the environmental model (i.e. the NDEP), moreover, the future establishment of a new coordinating umbrella-institution for the work of the various organisations operating the region is also envisaged in order to increase effectiveness of decision-making and resource management and to provide one single institutional referent vis-à-vis the policies implemented by the EU in the region. This post-structural liberalist interpretation of the environment rendered it possible to conclude the post-2004 scenario by sketching the potential applicability of this model to other security sectors. The ‘civil security’ sector broadly corresponds to the ‘societal security’ introduced within the RSCT as far as the construction of the logic of threats is concerned: hence, it refers to a range of issues such as migration and minority problems as well as crime prevention and communicable diseases. Civil security, however, differs from societal security in terms of ‘identity’. According to the Copenhagen School, identity represents the main element of this securitising move, while according to my Grotian argument, I noted, ‘Baltic identity’ is not applicable: hence, civil security. As far as relevant discourses focusing on civil security within the BSR are concerned, the broad categorisation of the regional threats that was modelled on the basis of the environmental case – East/West disparity and Russian engagement – may stand. Moreover, the broader lesson of this model is less applicable as far as bottom-up multiplicity is concerned (which remains somewhat deficient in the civil security field), while there are significant steps in the direction of a horizontal coordination of the institutional dimension, as demonstrated by the recently established 280 Conclusions ‘Northern Dimension Partnership in Public Health and Social Wellbeing’. The economic sector in the BSR has also been undergoing a form of securitisation, although a certain ‘manufacturing’ of security is noted in the economic case given that the idea of ‘taking risks’ is somewhat intrinsic in market economies. However, the notion of economic security was associated with the competitiveness of the region, i.e. to avoid marginalisation and to enhance the potentials of a Baltic regional economy in the enlarged Europe. In the economic case, as in the other sectors, the logic of threats in dominant discourses can also be suitably read in the light of the broad categories of East/West disparity and Russian engagement. As far as the broad lessons of the post-structural liberalist model are concerned, it can be argued that the notion of (‘bottom-up’) multiplicity is ‘by definition’ applicable in the free market context, and it is for instance visible in the efforts made at the regional level to include small and medium enterprises in the Baltic Sea market. The concept of ‘coordination’ is instead less suitable to be associated with the context of free-market economies, though it may be related to the frequent recent calls to establish a ‘Baltic Sea Economic Space’. In sum, in order to frame how the analysis of the relevant security discourses defines the post-2004 scenario put forward in this study, the security sectors must be reunited. The analysis of security discourses indicated that the political and military aspects of the regional cooperation are likely to coalesce in the wider European framework and will be primarily dealt with in terms of bilateral EU-Russia relations (cf. the case of Kaliningrad and the question of the Russianspeaking minority in the Baltics). It also emerged that the regional discourse has not been practiced in terms of security community, because securitisation remains a central element in triggering cooperation, and identity-building is not part of the dominant discourses. Given that ‘hard’ sectors coalesce in the European dynamics, this securitisation will continue to take place at the nonmilitary level. The case of environmental security is both a model and an exception. It is a model in the sense that the pattern of inclusiveness and coordination, and the focus on East/West disparity and Russian engagement, are also largely consistent with the civil security and economic sectors. This makes the post-structural liberalist model 281 Conclusions indicated by the Grotian tradition – in all its nuances as visualised by the (*) in the chart above – suitable to contain the different options of the post-2004 non-military cooperation. The environmental case, however, is an exception because in this sector the BSR is constructed as an actual security complex. In the other non-military sectors, instead, the bulk of securitisation dynamics does not ‘line up’ in the BSR and take place at the level of the EU complex or in terms of EU-Russia quasi-imperial confrontation. However, securitising practices in the economic and ‘civil’ fields do not entirely coalesce in the European complexes and a certain securitisation focusing on the BSR is detectable. This provides occasion for the return of the notion of sub-complex. The present analysis of security discourses indicates that Baltic regional cooperation after 2004 is likely to develop on the basis of a ‘mild’ nonmilitary regional interdependence. This will render the BSR a subcomplex, included in the EU’s complex, placed at its periphery, and thereby providing a “soft-soft” (Archer 2001) option for engaging the other European quasi-Empire, Russia. I shall conclude this study by following up on this final point, thus summarising that which this interpretation of fifteen years of Baltic Sea regionalism may suggest vis-à-vis the interdependence of the two wider complexes in the light of current security discourses. 2: Whether and Whither to Widen: towards a ‘Wider Europe’ Widening Dimensionalism? I shall first sketch three cases of dimensionalism on the EU periphery, which frequently appear in literature and discourses on transnational regional cooperation in Europe. Since the 1990s, it became quite en vogue among peace researchers to discuss the affinities between the BSR and the regional cooperation in the Mediterranean (see Holm and Joenniemi 2001; Maestro 2001). This was favoured by a number of self-evident similarities between these two phenomena: both are ‘closed seas’. Both of them involve EU-member states (in the Mediterranean case: France, Italy, Spain and Greece), non-members (North Africa and the Middle Eastern countries) and applicant countries/‘would be’ members (Cyprus and Malta, and, in the longer run, Turkey and the Balkans). In both cases, moreover, 282 Conclusions the EU has been significantly engaged in fostering coordination and synergies. However, beyond these basic similarities, any comparison between the BSR and the Mediterranean Sea security and cooperation is largely regarded as unfeasible. First, the Mediterranean Sea has not quite followed the same chronology and modality as the Baltic with regards to the development of regional phenomena. While post-Cold War BSR was a ‘tabula rasa’ where experimental cooperative initiatives took place, cooperation in the Mediterranean has been perceived as a “tabula plena” (Holm and Joenniemi 2001, 3), where power relationships and historical cleavages had been firmly established long ago. Secondly, the BSR, rather than being a model, has in many respects been inspired by the Mediterranean precedent. For instance, the NDI was inspired by the so-called ‘Barcelona process’ (see Lipponen 1997, 29), which is the EU partnership framework in the Mediterranean since November 1995, although the two differ sensibly in character, contents and ambitions (see Holm and Joenniemi 2001, 14-19)1. Thirdly, the Mediterranean dimension has been rather deficient with respect to inclusiveness and multilevel participation. This is partly because of its more marked state-centric character (ibid, 31), and partly because this Sea is a more heterogeneous microcosm that can itself be divided into a number of sub-dimensions (the Adriatic area, the Aegean Sea etc.), each of which are characterised by different challenges. Discussions concerning the applicability of the Baltic case to another mare clausum, the Black Sea, have developed along similar lines (see Hubel and Gänzle 2001, 21-22; Vahl 2001, 9-10; Emerson 2001, 41-42; Haukkala 2002, 29-30). The Black Sea is also surrounded by prospective/‘would be’ EU members (Turkey, Romania, and Bulgaria), non-members (Ukraine, Russia and Georgia), and, enlarging the focus somewhat, one member (Greece). As in the BSR, the limited Black Sea cooperation has focused mainly on the coordination of nonmilitary activities, e.g. by establishing the Black Sea Economic Cooperation Pact, a regional body in some respects similar to the CBSS2 (see Haukkala 2002, 29-30). Also in this case, however, the differences seem to outweigh the similarities. The EU presence is much more timid in the Black Sea than is the case in the BSR (see Vahl 2001, 9). Most importantly, the range of challenges in the Black Sea area cannot be limited to non-military 283 Conclusions issues. Especially in the Caucasus, they include ‘hard’ issues, e.g. territorial disputes (between Armenia and Azerbaijan, and within Georgia), political stability (as confirmed by Georgia’s November 2003 ‘rose revolution’), and the cumbersome and somewhat ambiguous Russian presence (The Economist November 29 and 8 2003; Emerson 2001, 38-40). The dimension that most recently has drawn the attention of policy makers and observers has been the so-called Eastern Dimension (ED), focusing on EU’s ‘new’ neighbours: Ukraine, Moldova, Belarus (i.e. the Western Newly Independent States, WNIS), plus Russia, though the latter is hardly a ‘new’ neighbour (see Cimoszewicz 2003; Polish Ministry of Foreign Affairs 2003; Haukkala 2003 and 2002; Joenniemi and Browning 2003). The ED has been gaining a certain amount of attention, probably because the main sponsor of the initiative has been Poland, the most ‘weighty’ among the countries accessing the EU in the 2004 enlargement round. This attention is also owing to the fact that the focus of the proposal is in fact a rather intricate cocktail of primary interests and exacting challenges of EU foreign policy, such as economic and political stabilisation of the three new neighbours, energy cooperation, and also, we shall see, EU-Russia relations. The prospect of an ED has been voiced by Poland since 1998 (see Cimoszewicz 2003) and has been substantiated by a ‘non-paper’ issued by the Polish Ministry of Foreign Affairs and circulated since January 2003 (Polish Ministry of Foreign Affairs 2003). The initiative, by definition, recalls the NDI. The Polish proposal, on this point, underlines that the ED “is not meant to compete with the Northern Dimension, but to be complementary to it. It should build on the experience derived from the Northern Dimension as well as from the Barcelona Process” (Polish Ministry of Foreign Affairs 2003, 1)3. As for the other ‘dimensions’, however, the logic of threats dealt with within the context of the ED differs substantially from the postCold War BSR. Ukraine, Moldova and Belarus are countries still afflicted by political instability, latent conflicts (e.g. the Transdniestria region in Moldova), authoritarianism (in Belarus), poverty (Moldova and Ukraine) and, last but not least, heavy dependency on Russia (see The Economist October 25 2003; Haukkala 2002, 12-29; Joenniemi and Browning 2003, 475). Moreover, the bottom-up participation is rather weak thus far (see Haukkala 2002, 30), as is the coordination of 284 Conclusions different actors, both of which are aspects that Warsaw values as essential to bring the initiative success4. Another major difference between the ED and the BSR is that the ED focuses on countries that do not have immediate prospects for EU membership. According to the Polish proposal, these states ought not to be denied access. At present, however, that option is off the table. Quite interestingly, moreover, Poland itself is not ‘part of’ the Eastern Dimension (see Joenniemi and Browning 2003, 471). The ED targets four “Eastern European states” (Polish Ministry of Foreign Affairs 2003, 1) in which Poland is not included. This differs sensibly from, e.g. Finland (and theoretically Poland itself), which has been included in the NDI all along. Poland, instead, has “experience and knowledge of the region concerned” (Polish Ministry of Foreign Affairs 2003, 1) and can exploit it to tackle the rising challenges. The message, as Joenniemi and Browning rightly argue, is not about inclusiveness and participation, but of the very otherness and securitisation that the NDI (partly) and the BSR experience have aimed to overcome. Ukraine, Moldova, and Belarus are threats to be accounted for, problems to be solved, as opposed to challenges to be incorporated. As Joenniemi and Browning have written, “the ED seems to pertain to the export of ready-made policies and is rather bluntly geared to achieving domestic reforms” (Joenniemi and Browning 2003, 472). Finally, the Polish ‘non paper’ refers to Russia only en passant, almost as if it were ‘one of’ the countries within the ED, without really accounting for Moscow’s strategic weight nor its influence towards the other three countries (see for this argument Joenniemi and Browning 2003, 475 and Polish Ministry of Foreign Affairs 2003, 2; Cimoszewicz 2003). This simplified sketch suggests that EU relations towards its neighbours define a rather fragmented patchwork. These four dimensions are all somewhat tied together, as illustrated in the figure below. Rather significantly, the figure actually suggests that Russia is a ‘common denominator’ in three of these dimensions, which represents a significant aspect on which I will elaborate below. In relation to ‘comparability’ and ‘exportability’ of our Baltic model, however, it could be argued that the three examples that have been sketched seem to merely focus on the coordination of institutional activities (or even lack thereof), rather than on ‘dimensionalism’ proper. While these initiatives are affine to the BSR for geo-physical morphology (the 285 Conclusions Mediterranean and the Black Seas) or for geo-political vicinity (the ED), the BSR stands out as intrinsically different. Figure 29: A Patchwork of Dimensions Russia Belarus Ukraine Moldova Caucasus (Turkey) Maghreb Middle East Enlarged EU BSR ED ’Barcelona Process’ Black Sea A defining difference between the Mediterranean, Black Sea, and Eastern Dimensions and the BSR is the way they have been affected by Europeanisation. These three dimensions have not been affected by the East/West confrontation and subsequently by the EU enlargement process as massively as has been the case around the Baltic Sea. The Mediterranean and the Black Seas, in fact, are barely touched by the 2004 enlargement. A second argument, not unrelated to the first, is that the BSR has virtually eliminated ‘hard’ issues from the regional agenda, while in the other dimensions these issues remain the overall priorities in different ways. Finally, at the heart of my definition of the post-structural liberalist sub-complex was the notion of inclusiveness. Inclusiveness has been the trademark of Baltic Sea cooperation since its very rise, and it is in this regard that the unique role played by Norden has been most visible. In these other ‘dimensions’, this appears still lacking. It could actually be argued to some extent that the three ‘dimensions’ introduced here are comparable to one another in terms of 286 Conclusions logic of threats and type of cooperation, and that the BSR is the one case standing out as the exception rather than the rule. Hence, the overall impression is that the difference between EU’s neighbourhoods appears too great to rearticulate the Baltic experience, and that therefore, the Baltic Grotian model is largely not exportable. A somewhat different pattern is indicated by most recent developments, however. The Wider Europe-New Neighbourhood Concept The EU Commission has recently launched a proposal to overcome the irreconcilable diversity characterising the present (or emerging) dimensions. This proposal, widely known as the ‘Wider Europe’ concept, aims to determine a common and comprehensive New Neighbourhood Instrument including all of the new EU neighbours: the Southern Mediterranean, the WNIS and Russia (Commission of the European Communities 2003 a and b)5. In view of the arguments in the preceding section, the idea of a single initiative aiming to create, in the words of EU Commission President Romano Prodi, “a ‘ring of friends’ surrounding the Union” (Prodi 2002, 3) initially seems vague and simplistic. However, the impression is that this apparent ‘simplicity’ is precisely what the ambitious challenge before Wider Europe is all about. The March 2003 Communication of the EU Commission, which lays out the foundations of the Initiative, recognises that “the situations of Russia, the countries of the WNIS and the Southern Mediterranean are very different judged by most standards” (Commission of the European Communities 2003a, 4). Yet the idea of creating a single instrument is motivated by the fact that “the EU should aim to ensure a more coherent approach, offering the same opportunities across the wider neighbourhood” (Commission of the European Communities 2003a, 16). And further, that “although the political, economic and social context varies from border to border, the key cooperation objectives to be addressed are broadly valid for all” (Commission of the European Communities 2003b, 18). This phrasing of the concept points at a rather innovative rationale. The stability of the enlarged EU is first and foremost at the heart of Wider Europe. Quite illustratively, for instance, Prodi’s notion of a ‘ring of friends’ changes quite bluntly in the words of Chris Patten, who argued that: “Europe [read EU] is ringed – from Kaliningrad in 287 Conclusions Russian security complex (Buzan and Wæver 2003, 529-533) and, at present, they cannot really be counted as an actual ‘sub-complex’, as was argued for the Baltic region, because of the diversity characterising their security interdependences (ibid. 533). However, the idea of an Eastern Dimension – including the WNIS, Russia, Poland and, possibly, Latvia and Lithuania11 - within Wider Europe may prove of great strategic importance. The differences between the Baltic/Northern and Eastern dimensions are significant, since, it was argued, the two security discourses are largely incompatible12. The aim here, however, should not necessarily be to rearticulate the whole content of the BSR experience. A prospective ED may look at the BSR as a model to foster stability in the WNIS and to enhance EU-Russia partnership. The common denominator between the BSR and the ED is that they both reside at the “near abroad” of the two quasi-empires (Emerson 2001, 4). This point relates to the theoretical observations in the last of part of Chapter 2, where it was argued that sub-complexes/mesoregions placed at the ‘double periphery’ of centred complexes may serve as ‘hooks’ for the inter-complex relations. Both dimensions have the potential to link up the EU and Russian quasi-empires, and may play the function of ‘border-breaking subjects’ for the benefit of the two complexes. In this context, notwithstanding the irreconcilable incompatibilities between these dimensions, the two meso-regions may mirror one another. The BSR has been striving – and has now succeeded – to become part of the EU, and the ‘soft’ regional cooperation over the past fifteen years has been instrumental for strengthening the transition of Poland and the Baltics towards it, as well as for the fostering of Russian participation in it. Likewise, accepting the WNIS as included in the Russian complex (see figure above) would imply that regional cooperation in the ED based on soft issues could prove a valuable basis for coordinating activities and enhancing EU inclusiveness in the area. This might consequently foster political and economic transformation in the WNIS, and EU-Russia relations may be provided with a further channel of dialogue (for a similar argument see Lynch 2003, 84-86 and Haukkala 2002, 34-37). Such a scenario would admittedly downplay the ‘grander’ purpose of the ED and possibly of the Wider Europe concept as such. Indeed, accepting the WNIS as ‘embedded’ in the Russian complex would 294 Conclusions implicitly mean ‘legitimising’ the undemocratic and Russia-dependent regimes in these countries (see on this point The Economist February 21 2004, 28). On the other hand, at least at the beginning, this would help to overcome Russian suspicions as to the EU-centric mark of the Wider Europe initiative. This proposal appears to be the most logical path for the rearticulation of the Baltic experience. Moreover, the idea of viewing the two dimensions as hooks for the two quasi-empires may also provide a realistic answer for understanding where the component of the Wider Europe initiative as the “new pillar of the wider strategic partnership” between EU and Russia could actually lie. To quote Emerson, “a new model is emerging, based on two paradigms, which could be promising if followed through with substantial policies. First, the two empires talk of strategic partnership. Secondly, there is also developing a pattern of borderland regionalism” (Emerson 2001, 34). The BSR experience suggests that when hard issues are at stake, EU-Russia cooperation in the region has regressed. Hence, in order for the partnership to avoid becoming bogged-down in a bilateral stalemate, this joint EU-Russia ‘middle ground’ is to first focus on nonmilitary questions. This is where Moscow’s alleged European values are to be tested. This is also, however, where the presumed EU willingness to include Russia in its decision-making should be demonstrated and where, therefore, the EU must take “the first step” (Lynch 2003, 85). One could point at, among other matters, energy and economic cooperation as two primary and deeply interconnected issues in the current debate where there seems to be: “a stable equilibrium of mutual dependence” (Emerson 2001, 2) and where the model proposed in the project at hand may apply. These two issues are relevant for both parties, since the EU is increasingly dependent on Russian energy supplies, and is Russia’s first economic partner (Lynch 2003, 63-65 and Commission of the European Communities 2003a, 19-26). An attempt at making the EU-Russia debate on these two sectors more ‘inclusive’ can be spotted in the launch of an Energy Dialogue since October 2000 and in the establishment of a High Level Group on the Common European Economic Space since May 2001 (see Lynch 2003 90-91; Vahl 2002). Progress in this debate has been regrettably slow, however, on account of the strategic importance of the energy sector for Russia, 295 Conclusions and because the prospects of Russian membership in WTO are still very problematic, and constitute the sine qua non for any economic talk with the EU (Emerson 2004, 33-34). As Russian Deputy Prime Minister Victor Khristenko recently noted, “Russia has already begun modifying its economic conditions with European standards […]. But oil and gas exports are too important for Russia; hence it looks to reach special accords with the EU on this issue” (Khristenko 2004). The experience of the BSR indicates that in order to overcome this impasse, EU and Russia could begin by fostering the regionalisation of these issues in the ED, which epitomises many of the current problems affecting EU-Russia relations. The heavily Russia-dependent economies of these three countries are severely dysfunctional. On the other hand, they play a crucial role for Russian energy exports by constituting an important transit route for Russian natural resources (see EIA Country Analysis Briefs, Ukraine 2003; Haukkala and Moshes 2004, 23). Following the Baltic Sea model, the regionalisation of these ‘soft’ sectors13 could promote sustainable growth and political stability in the WNIS by enhancing ‘top-down’ coordination and ‘bottom-up’ inclusiveness. The two complexes would not put the bulk of their bilateral relations at stake by encouraging this process, which, in return, could prove a significant test case for the wider pattern of EURussia relations. Finally, the post-structural liberalist model elaborated by this study underlines the importance of a proactive epistemic community to achieve these goals. The promotion of educational programmes, cultural exchanges, research, and civil society links would be of crucial importance for enhancing cooperation in the ED. This would soften Russian suspicions regarding the EU-centrism of initiatives emanating from Brussels and strengthen the whole-European inspiration of the bilateral dialogue14. These prospects for EU-Russia strategic partnership would arguably reiterate a rather well-known criticism that, as US scholar Robert Kagan stated, “focus[es] on issues ‘challenges’ – where European strengths come into play, but not on those ‘threats’ where European weakness makes solution elusive” (Kagan 2003, 32-33). Strength in tackling ‘challenges’ and weakness in tackling ‘threats’ are indeed two aspects also characterising the security and cooperation within the 296 Conclusions BSR, which, it was argued in this study, has been in many respects the epitome of the wider European dynamic. However, the story told by this region over the past fifteen years is that taking up ‘challenges’ can constitute a means of moving beyond ‘threats’ and, ultimately, providing that which the late Anna Lindh defined as ‘a model of peace’ (Lindh 2001, 1). This is the sense of the statements by Vladimir Putin and Günter Verheugen that opened this chapter. This is also where, most of all, the Baltic Sea Region has proven to be Mare Europaeum, a truly European Sea. 297 Conclusions Notes 1 The Barcelona Process, also known as the Euro-Mediterranean Partnership, involves twelve Partners: Morocco, Algeria, Tunisia (Maghreb); Egypt, Israel, Jordan, the Palestinian Authority, Lebanon, Syria (Mashrek); Turkey, Cyprus and Malta. The three main objectives of the process are to Establish “a common Euro-Mediterranean area of peace and stability; to create “an area of shared prosperity through the progressive establishment of a free trade area between the EU and its Partners and among the Mediterranean Partners themselves,” to “develop human resources,” and to “promote understanding between cultures and rapprochement of the peoples in the EuroMediterranean region as well as to develop free and flourishing civil society” (see Barcelona Declaration 1995, 3, 4 and 7). 2 The Black Sea Economic Cooperation Pact includes the littoral countries, plus Greece, Albania, Armenia, Azerbaijan and Moldova (see Vahl 2001, 9). In a sense, it could be argued that the Black Sea cooperation could resemble the BSR also as far as the ‘Wightian’ composition of the states is concerned, being surrounded by the three small South Caucasian post-Soviet states (like the three Baltics), two ‘great’ powers such as Turkey and Russia, and some medium powers (e.g. Ukraine, Romania, Bulgaria). 3 Just like the NDI, the Polish proposal also advocates the drafting of an action plan (see Cimoszewicz 2003a, 4 and Polish Ministry of Foreign Affairs 2003). Along the same lines, the US-sponsored e-PINE has also hinted at the possibility of ‘exporting success’ (see e-PINE US State Department Website). Also Ronald Asmus recently referred explicitly to the ‘eastern agenda’ (Ukraine and Belarus) as “the most pressing task” (Asmus 2003, 23). 4 The Polish Foreign Minister Cimoszewicz argued that “the Eastern Dimension formula would facilitate greater involvement of international institutions and private capital in assistance projects” (Cimoszewicz 2003, 2), which is a strategy recalling the ‘Partnerships’ launched within the NDI. Moreover, the importance of NGOs and bottom up initiatives is also stressed in the same speech. As the Minister put it, “investment in human capital always pays off” (Cimoszewicz 2003, 3). 5 The Commission states that the Initiative “does not […] apply to the Union’s relations with the remaining candidate countries – Turkey Romania and Bulgaria – or the Western Balkans.” Nor does it include the Southern Caucasus (and thus the prospective ‘Black Sea dimension’), because “the Communications focuses […] on those countries that are or will be in 2004, on the immediate borders of the Union and do not have the perspective of membership currently (Moldova is the exception). Once Turkey has joined the EU, the geographical scope may need to be re-visited” (Wider Europe – Neighbourhood Website). A November 2003 report by the European Parliament, however, advocated the inclusion of the Black Sea in the Wider Europe concept (in Emerson 2004, 24). At the time this dissertation was concluded “it seems likely that in May 2004 the countries of the Southern Caucasus (Armenia, Azerbaijan and Georgia) will also be added to the list of the EU neighbours” (Haukkala and Moshes 2004, 10). 298 Conclusions 6 The two overarching objectives of Wider Europe are “to reduce poverty and create an area of shared prosperity and values based on deeper economic integration” and “to anchor the EU’s offer of concrete benefits and preferential relations within a differentiated framework which responds to progress made by partner countries in political and economic reform” (Commission of the European Communities 2003a, 9). “Greater EU Political Involvement in Conflict Prevention and Crisis Management” only receives brief mention (Commission of the European Communities 2003a, 12). 7 In the more detailed Communication of July 2003, the Commission also drafted a list of 24 ‘Neighbourhoods Programmes’ (Commission of the European Communities 2003b, 14). 8 The point was reaffirmed by Conclusions of the European Council of November 18 2002 where it was argued that: “This initiative should be seen in conjunction with the EU’s strong commitment to deepening co-operation with the Russian Federation, which is a key partner” (quoted in EU Commission Wider Europe – Neighbourhood website; see also Commission of the European Communities 2003a, 5) 9 The other being the lack of financing, which Wider Europe should be able to overcome, at least from 2007 onwards. 10 On the other hand, Brussels has recently grown increasingly wary about the possibility of enhancing Russian inclusiveness in view of the most recent domestic developments vis-à-vis political freedom, the rule of law etc. (see The Economist February 21 2004 and Commission of the European Communities 2004). 11 The possibility that the experience of the Baltic Sea and the NDI can be usefully exported to Ukraine, Belarus and Moldova has been stressed by policy makers from Latvia and Lithuania (Paksas 2003; Adamkus 2002; Berzins 2002). 12 Following the categorisation of sub-complexes in the critical RBA (see figure 14 in Chapter 2), it could be argued that the WNIS could prospectively turn into a subcomplex type B, i.e. where functional cooperation is facilitated by both Russia and the EU. The Grotian model characterising the Baltic Sea, instead, already goes beyond this type B. 13 The ‘dimensionalist’ potential of these two sectors in the Eastern Dimension is stressed within both the Wider Europe (Commission of the European Communities 2003a, 7 and 10; Prodi 2002, 5) and the Polish ED contexts (Cimoszewicz 2003, 2 and Polish Ministry of Foreign Affairs 2003, 3). 14 It should also be noted that this brief mention of the economic and energy sectors is to be considered an example and a starting point as to where the ‘dimensionalist side’ of the EU-Russia strategic partnership may develop in the future. It is not ruled out that other sectors are to be considered, nor that the strategic partnership as such may include other issues. In May 2003, for example, the EU-Russia Summit in St. Petersburg reiterated the idea of the four common spaces (“common economic space, common space for freedom, security and justice, common space of cooperation in the field of external security, as well as a common space in the field of education, including culture” (Commission of the European Communities 2004, 1)), which ought to constitute the basis of the future partnership. 299 Conclusions 300 Bibliography Bibliography Speeches and Official Documents Adamkus, Valdas: Post-Enlargement Challenges for the Baltic Sea Region, speech delivered by Valdas Adamkus, President of Lithuania, at the Sixth Conference on Baltic Sea Security and Cooperation in Stockholm [retrievable from www.usemb.se/bsconf/2002/speech_adamkus.html last accessed November 14, 2003], April 24, 2002. Adamkus, Valdas: Speech delivered by Valdas Adamkus, President of Lithuania, at the Fourth Conference on Baltic Sea Security and Cooperation in Stockholm [retrievable from www.usemb.se/bsconf/1999/text/02adamkus.html last accessed November 14, 2003], November 4, 1999. Agreement on Partnership and Cooperation - establishing a partnership between the European Communities and their Member States, of the one part and the Russian Federation, of the other Part, 1997. Asmus, Ronald D.: American Views on Security and Cooperation in the Baltic Sea Region, Address by Ronald D. Asmus, Deputy Assistant Secretary of State, at the Second Conference on Baltic Sea Region Security and Cooperation in Stockholm [retrievable from www.usemb.se/bsconf/asmus.html last accessed November 27, 2003], November 6, 1997. Asmus, Ronald D.: Address by Ronald D. Asmus, Senior Analyst at RAND, at the First Conference on Baltic Sea Region Security and Cooperation in Stockholm [retrievable from www.usemb.se/BalticSec/ASMUS.htm last accessed November 27, 2003], November 19, 1996. Baltic 21: An Agenda 21 for the Baltic Sea Region – Baltic 21, Baltic 21 Series No 1/98 Stockholm, 1998. Barcelona Declaration: adopted at the Euro-Mediterranean conference [retrievable from http://europa.eu.int/comm/external_relations/euromed/bd.htm last accessed December 30, 2003], January 27-28, 1995. 301 Bibliography Bartoszewski, Wladyslaw: Address by Wladyslaw Bartoszewski, Minister of Foreign Affairs of the Republic of Poland, at the 10th Ministerial Session of the Council of the Baltic Sea States in Hamburg [retrievable from: http://www.cbss.st/documents/cbsspresidencies/9german/10thministrialse ssion/statements/dbaFile282.html, last accessed November 24, 2003], June 7, 2001. Berzins, Indulis: Statement by Indulis Berzins, Foreign Minister of the Republic of Latvia, at the 11th Ministerial Session of the Council of the Baltic Sea States in Svetlogorsk, Kaliningrad Oblast [retrievable from http://www.cbss.st/documents/cbsspresidencies/10russian/11cbssministeri alsession/dbaFile786.html last accessed on November 22, 2003], March 5-6, 2002. Berzins, Indulis: Statement by Indulis Berzins, Minister of Foreign Affairs of Latvia, at the 10th Ministerial Session of the Council of the Baltic Sea States in Hamburg [retrievable from: http://www.cbss.st/documents/cbsspresidencies/9german/10thministrialse ssion/statements/dbaFile279.html, last accessed November 24, 2003], June 7, 2001. Berzins, Indulis: Speech delivered by Latvian Foreign Minister Indulis Berzins, at the Fourth Conference on Baltic Sea Security and Cooperation in Stockholm [retrievable from www.usemb.se/bsconf/1999/text/03indulisberzins.html last accessed November 14, 2003], November 4, 1999. Bildt, Carl: “The Baltic Litmus Test: Revealing Russia’s True Colors” in Foreign Affairs (73:5), 1994. Bloomberg Jakko: Russia and the Baltics: a Test for the New Security Structure Address by Ambassador Jakko Bloomberg at the Second Conference on Baltic Sea Region and Security in Stockholm [retrievable from http://www.usemb.se/bsconf/blomberg.html last accessed November 11, 2003], November 6, 1997. Brundtland Commission: Our Common Future. The World Commission on Environment and Development. Oxford, OUP, 1987. Bury, Hans Martin: Statement by Hans Martin Bury Minister of State at the Federal Foreign Office of Germany, at the 12th Ministerial Session of the Council of the Baltic Sea States in Pori, Finland [retrievable from 302 Bibliography http://www.cbss.st/documents/cbsspresidencies/11finnish/pori/speeches/d baFile1952.html last accessed November 22, 2003], June 11, 2003. Cimoszewicz, Wlodzimierz: The Eastern Dimension of the European Union. The Polish View, Statement by Wlodzimierz Cimoszewicz, Polish Minister of Foreign Affairs at the Conference “The EU Enlargement and Neighbourhood Policy” in Warsaw, February 20, 2003. Cimoszewicz, Wlodzimierz: Statement by Wlodzimierz Cimoszewicz, Foreign Minister of the Republic of Poland, at the 11th Ministerial Session of the Council of the Baltic Sea States in Svetlogorsk, Kaliningrad Oblast [retrievable from: http://www.cbss.st/documents/cbsspresidencies/10russian/11cbssministeri alsession/dbaFile789.html last accessed November 24 2003], March 5-6, 2002. Commission of the European Communities: Communication from the Commission to the Council and the European Parliament on relations with Russia, Brussels, February 9, 2004. Commission of the European Communities: Communication from the Commission Paving the way for a New Neighbourhood Instrument, Brussels, July 1, 2003b. Commission of the European Communities: Communication from the Commission to the Council and the Parliament Wider Europe – Neighbourhood: a new Framework for Relations with our Southern and Eastern Neighbours, Brussels, March 11, 2003a. Commission of the European Communities: Commission Working Document: the Second Northern Dimension Action Plan, Brussels, 2003. Commission of the European Communities: Guidelines for a new Action Plan, Presidency Conclusions on the Future of the Northern Dimension Policies of the European Union, Brussels, October 2002. Commission of the European Communities: Communication from the Commission to the Council: The EU and Kaliningrad, Brussels, January 17, 2001. Commission of the European Communities: Commission Communication to the Member States laying down Guidelines for a Community Initiative 303 Bibliography States in Hamburg [retrievable from http://www.cbss.st/documents/cbsspresidencies/9german/10thministrialse ssion/dbaFile284.html?1=1&print=true last accessed November 11, 2003], June 7, 2001. Lipponen, Paavo: Speech by Prime Minister Paavo Lipponen at the Forum on the Northern Dimension in Social Protection and Health [retrievable from http://www.valtioneuvosto.fi/vn/liston/base.lsp?r=20257&k=en&old=754 last accessed December 22, 2003], September 16, 2002. Lipponen, Paavo: Speech by Prime Minister Paavo Lipponen at the plenary session of the European Parliament [retrievable from http://presidency.finland.fi/netcomm/news/showarticle1622.html last accessed December 22, 2003], October 27, 1999a. Lipponen, Paavo: Opening Speech by Finland’s Prime Minister Paavo Lipponen at the Foreign Ministers Conference on the Northern Dimension in Helsinki [retrievable from http://presidency.finland.fi/netcomm/news/showarticle1889.html last accessed November 30, 2003], November 12, 1999. Lipponen, Paavo: “The European Union Needs a Policy for the Northern Dimension” Speech by Finland’s Prime Minister Paavo Lipponen in Rovaniemi, Finland, in Heininen, Lassi and R. Langais: Europe’s Northern Dimension: the BEAR meets the South, September 1997. Luik, Juri: Baltic Security: Why NATO? Speech by Juri Luik, Estonian Ambassador in Brussels at the First Conference on Baltic Sea Region Security and Cooperation [retrievable from www.usemb.se/BalticSec/LUIK.htm last accessed November 14, 2003], November 19, 1996. Lykketoft, Mogens: Statement by Mogens Lykketoft Minister of Foreign Affairs of Denmark, at the 10th Ministerial Session of the Council of the Baltic Sea States in Hamburg [retrievable from: http://www.cbss.st/documents/cbsspresidencies/9german/10thministrialse ssion/statements/dbaFile275.html, last accessed November 24, 2003], June 7, 2001. North Atlantic Treaty Organisation: NATO Review, Spring 2003. North Atlantic Treaty Organisation: NATO Update June 11, 2002. 310 [Document text truncated for crawler view.]