securitization diagram

Securitisation: Taking stock of a research programme in Security Studies Ole Wæver February 2003; draft1 Appraising theories and research programmes – or continuing the paradigm wars? ―[H]e misunderstands paradigms. He believes that paradigms easily generate a family of theories (…). Paradigms are apparently like sausage machines: Turn the crank and theories come out. Yet nobody in any field is able to generate theories easily or even to say how to go about creating them.‖2 Kenneth Waltz here makes an important comment on how to evaluate a theory. A theory3 (or a research programme4) should be assessed on its own terms: what can it do and where does it fail? Because paradigms are not theory machines, very little can be achieved by debating vague and general paradigms, or at least: such debates do not constitute serious evaluation of any specific theory. 1 I appreciate helpful comments on previous drafts presented to research seminars at COPRI (Copenhagen Peace Research Institute) and the department of political science at the University of Copenhagen and to the Workshop on ―The Traditional and the New Security Agenda: Inferences for the Third World‖, Universidad Torcuato Di Tella Buenos Aires, September 11-12, 2000: Lene Hansen, Barry Buzan, Thomas Diez, Morten Kelstrup, Birgitta Frello, Karen Lund Petersen, Ib Damgaard Peteresen, Helle Johansen, Anders Wivel, Mikkel Vedby Rasmussen, Henrik Larsen, Helle Malmvig, Carsten Bagge Laustsen, Rut Diamint, Ann M. Florini, Gavin Cawthra, Mónica Herz, Iván Witker Barra, Abdur Rob Khan, Virgilio R. Beltran, Anna Leander, Stefano Guzzini, Tarja Cronberg, Pertti Joenniemi, and Chris Browning. This paper started as a presentation at a 1999 BISA roundtable organised by Theo Farrell, where the Copenhagen School was the most mainstream approach to be superseded by first cultural-constructivist security studies, then critical security studies and finally radical security studies. 2 Kenneth N. Waltz, ―Evaluating Theories‖, American Political Science Review, vol. 91:4, December 1997, pp. 913-918 (quoting from p. 913). ‗He‘ in the quote is John Vasquez, because the article is part of a ―Forum‖ on John A. Vasquez, ―The Realist Paradigm and Degenerative versus Progressive Research Programs: An Appraisal of Neotraditional Research on Waltz‘s Balancing Proposition‖, ibid., pp. 899-912. See also Kenneth N. Waltz, Theory of International Politics, 1979, pp. 13-16 and Waltz, ―The Validation of International-Political Theory‖, paper for the 1994 annual meeting of APSA, New York, Sept. 1-4 1994. On this theme, see also Miriam Fendius Elman and Colin Elman, “How Not to be Lakatos Intolerant: Appraising Progress in International Relations Theory,” in International Studies Quarterly, vol. 46:2, June 2002, pp. 231-262. 3 I can continue to follow Waltz: ―I define theory as a picture, mentally formed, of a bounded realm or domain of activity. A theory depicts the organization of a realm and the connections among its parts.‖ ―A theory is not a mere collection of variables. If a ‗gap‘ is found in a theory, it cannot be plugged by adding a ‗variable‘ to it. To add to a theory something that one believes has been omitted requires showing how it can take its place as one element of a coherent and effective theory.‖ (―Evaluating‖, pp. 913 and and 916) Naturally, I am aware that Waltz furthermore assumes that the primary products of theories are explanations and that they should be ‗tested‘ by inferring hypotheses and checking these against experimental or observational data. My concept of ‗theory‘ allows for other forms of understanding, and therefore I follow Waltz primarily in his more basic conception of how theories are made – ―creatively‖ (Theory, op.cit., p. 9), which implies that the core of a theory is not something one ‗finds‘ in reality, but some idea for how to conceive of a field. 4 Again as Waltz, I am not going to make a strong distinction between theory and research program. The core of a (good) research program is necessarily a theory, and the introduction of the term ‗research program‘ usually serves two purposes. One is to give connotations to a more dynamic, time-stretched and collective enterprise which is certainly appropriate in this context. The other is to set up more or less strict criteria for a Lakatosian or post-Lakatosian evaluation of the ‗progressiveness‘ of a research program. I will not on this occasion go far in this latter direction, first because the Lakatosian research program in the sociology of science has run into serious difficulties itself, and second because securitisation theory is still so young that attention naturally focus less on patterns in its evolution (progressive? degenerative?) and more on the character of the theory itself. On problems with Lakatos, see eg. Donald N. McCloskey, Knowledge and Persuasion in Economics, Cambridge: Cambridge University Press 1994; Paul Diesing, How Does Social Science Work? Reflections on Practice, Pittsburgh: University of Pittsburgh Press 1991. 1 When appraising a theory, focus should not be on its assumptions but how it works as theory; and especially we should not base our judgements on guilt by association. The latter is surprisingly common. As noted by Waltz, most of the critiques of his neo-realist theory attack statements by other realists or find ‗inconsistencies‘ in his theory by attributing to him views presumably held by all realists: ―One cannot judge the fertility of a research program by evaluating work done outside of it.‖5 Similarly, when the theory of securitisation has been attacked by mainstream authors, the target has most often been views falsely attributed to the theory because it shares other elements with constructivists, post-structuralists, neo-realists or ‗security wideners‘6. This tendency to place all discussions at the level of general schools or paradigms is a corollary of the often-noticed preference of IR for ‗grand debates‘. This preference has its advantages and disadvantages and in any case it is not gotten rid of simply by deciding so, because it is part of the intellectual and social structure of the discipline and thus partly a question of power.7 While therefore these general debates between very loosely defined schools or paradigms are here to stay, it is my hope that we are also able to conduct debates focused on specific theories and the present paper is an attempt to further one such debate. It is as such not a plea for adopting a wide or a narrow concept of security or for being more or less constructivist – it presents a theory, which has one very specific core idea and has been developed around this idea. An argument is made about what this theory can be used for and where it is useless, and some of the main criticisms are (briefly) assessed. Schools and debates in security studies – East and West of the Atlantic Seen in relation to the general discipline of international relations, the sub-field of security studies exhibits an unusual degree of divergence between European and American theoretical developments. In most other fields, scholars on both sides conduct – or at least are aware of – the same debates, even if these might be balanced very differently, witness the ‗grand debate‘ in IR about rationalism and constructivism or the debate in European studies between liberal inter-governmentalism and multi-level governance. However, within security studies most scholars on one side of the Atlantic would depict (and teach) the state of the discipline in terms of debates and studies that are not mentioned in a similar overview on the other side and vice-versa. Obviously, this does not mechanically and fully follow geographical criteria, but the trend is nevertheless clear. In Europe there is a vibrant debate over a number of competing schools in security studies: critical security studies, the Copenhagen school, radical post-modernists, feminists, Bourdieu-inspired approaches, and more traditional, realist positions. Most of these are not known at all to the majority of American scholars. (In this case, Canada is more European than American and partly itself with its own literature on ‗human security‘8, and the main 5 6 Waltz, ‖Evaluating Theories‖, op.cit., p. 915. As the section below on ‘strengths and weaknesses‘ shows, extremely many relevant, insightful and often effective criticisms have been made, but they have mostly originated (to stay in the unfortunate metaphorics of the paradigm wars) on the opposite, the radical, side of the theory (i.e. from post-structuralists, radical constructivists and feminists). 7 Ole Wæver, ‗The Sociology of a Not so International Discipline: American and European Developments in International Relations‘, in International Organization, vol. 52:4, 1998, pp. 687-727. 8 C. Thomas and P. Wilkin (eds.) Globalizaton, Human Security, and the African Experience, London: Lynne Rienner 1999; Astri Suhrke, ‖Human Security and the Interests of States‖, Security Dialogue, vol. 30:3, September 1999, pp. 265-276; Simon Dalby, Geopolitical Change and Contemporary Security Studies: Contextualizing the Human Security Agenda, The University of British Columbia: Institute of International Relations, Working Paper No. 30, April 2000; Kanti Bajpai, ‖Human Security: Concept and Measurement‖, 2 contrast is not Europe vs. North America but Europe vs. the U.S.) It is often argued by observers of European debates, that the field of security studies surprisingly has become one of the most exciting fields and the place where a lot of theoretical innovation takes place, of relevance to the whole field of IR.9 If you turn to the leading academic, American journals in security studies (or look at ph.d. theses written in the US), these debates register only very marginally.10 The leading debate is instead likely to be seen as the intra-realist debate between offensive and defensive realism11 (and other distinctions within realism) with numerous interventions refining the theoretical arguments and doing empirical case studies usually through in-depth historical studies. Also there have been major debates over particular hypotheses like the democratic peace12 and increasingly a debate that looks more like the European debates: the metatheoretical debate between constructivists and rationalists.13 However, the latter is also the Kroc Institute Occasional Paper #19:OP:1, August 2000 (64pp); Edward Newman, ‖Human Security and Constructivism‖, International Studies Perspectives, 2001:2, pp. 239-252; William Bain, ‖The Tyranny of Benevolence? National Security, Human Security, and the Practice of Statecraft‖ in Global Security, vol. 15:3, 2001, pp. 277-294; Roland Paris, ‖Human Security: Paradigm Shift or Hot Air?‖, International Security, vol. 26:2, Fall 2001, pp. 87-102; Nicholas Thomas and William T. Tow, ‖The Utility of Human Security: Sovereignty and Humanitarian Intervention‖, Security Dialogue, vol. 33(2), June 2002, p. 177-192; Alex J. Bellamy and Matt McDonald, ‖‘The Utility of Human Security‘: Which Humans? What Security? A Reply to Thomas & Tow‖, Security Dialogue, vol. 33(3), September 2002, pp. 373-377; Thomas and Tow, ‖Gaining Security by Trashing the State? A Reply to Bellamy & McDonald‖, Security Dialogue, vol. 33(3), September 2002, pp. 379-382. See also Canada‘s ‖Human Security Web Site‖: www.humansecurity.gc.ca, ‖Human Security Network‖: www.humansecuritynetwork.org, the Commission on Human Security: http://www.humansecuritychs.org/ and Harvard‘s ‖Program on Human Security‖: www.cbrss.harvard.edu/hs/ 9 Michael C. Williams, ‘Words, Images, Enemies: Securitization and World Politics‘, forthcoming in International Studies Quarterly, vol. 47(?). For similar claims about the vitality and importance of these debates, see Johan Eriksson, ―Introduction‖ in Eriksson ed. Threat Politics: New perspectives on security, risk and crisis management, Aldershot: Ashgate 2001, pp. 1-18, especially p. 18 (note 1); Jef Huysmans, "Revisiting Copenhagen: Or, On the Creative Development of a Security Studies Agenda in Europe", European Journal of International Relations, 4:4 (1998) 479-506; Christopher Corpora [add ref.]; Steve Smith, ‖The Increasing Insecurity of Security Studies: Conceptualizing Security in the Last Twenty Years‖ in Stuart Croft & Terry Terriff (eds.) Critical Reflections on Security and Change, London: Frank Cass 2000, pp. 72-101. 10 The divergence was maybe already signalled during the 1980s and early 1990s by the very different reception of Barry Buzan‘s People States and Fear (1983, 1991). It never made a big impact in the US, while it became a cenral reference, a standard textbook and a modern classic not only in the UK, but generally in Europe (and Canada?). 11 E.g. Sean Lynn-Jones & Steven Miller ‗Preface‘, in Brown, Michael, Sean Lynn Jones, & Steven Miller (eds). 1995. The Perils of Anarchy: Neo-realism and International Security. Cambridge: MIT Press 1995. Pp ix-xii; Jack Snyder, Myths of Empire: Domestic Politics and International Ambition, Ithaca og London: Cornell University Press 1991; Zakaria [review-essay]; Fareed Zakaria, From Wealth to Power - The Unusual Origins of America's World Role, Princeton, New Jersey: Princeton University Press 1998; John J. Mearsheimer, The Tragedy of Great Power Politics, New York: W. W. Norton & Company 2001; Stephen G. Brooks, ―Dueling Realisms‖, International Organization, Vol. 51, No. 3, 1997, pp. 445-477; Randall Schweller, ‗Neorealism‘s Status Quo Bias: What Security Dilemma‘, Security Studies, Spring, 5(3), 1996, pp. 90-121; Randall Schweller, Deadly Imbalances: Tripolarity and Hitler’s Strategy of World Conquest, New York: Columbia University Press 1998; Charles Glaser, ‗The Security Dilemma Revisited‘; World Politics. October, 50 (1), 1997, pp. 171201; Jeffrey Taliaferro, "Security Seeking under Anarchy: Defensive Realism Revisited," International Security, vol. 25, Winter 2000/01, pp. 128-61; Gideon Rose, ―Neoclassical Realism and Theories of Foreign Policy‖, World Politics, Vol. 51, 1998, pp. 144-172; Sten Rynning og Stefano Guzzini (2001): Realism and Foreign Policy Analysis, Working Papers 42/2001, København: Copenhagen Peace Research Institute: http://www.copri.dk/publications/WP/WP%202001/42-2001.pdf; Stephen M. Walt, ―The Enduring Relevance of the Realist Tradition‖ in Ira Katznelson and Helen Milner (eds.) Political Science: The State of the Discipline III, New York: W. W. Norton 2002. 12 Bruce Russett book + ref to key debates in IS and elsewhere. 13 Peter J. Katzenstein, ‖Introduction: Alternative Perspectives on National Security‖ in Peter J. Katzenstein 3 one that most clearly shows the differences. The first difference is that on the US side, reflections on the concept of security play no integrated role in research. Such considerations are at most involved in delineating the field and thereby locating the questions about which then to gain empirical, causal, historical and theoretical knowledge.14 If an article is defined as security analysis, this will typically mean that at most the reflection on this status consists in defining security studies as being about e.g. ―the study of the threat, use and control of military force‖15 and therefore a theoretical-empirical study of causes of war is relevant or the strategic use of sanctions might with a little more difficulty be justified too; the concept of security is not present in the analysis as such. In the European debates, questions about the concept of security became the launching pad for a general attention to the self-reflective nature of the discipline, i.e. that the discipline not only studied security, but it also had its own concepts of security and thereby its own security practices. ‗Doing security‘ therefore implied to reflect on the practice of speaking in the name of this concept.16 This pointed towards a general attention to the close connections between (sub-)discipline, theory, concept and the studied object (all called something like or with ‘security‘). A partly related, second difference is that in Europe, a particular debate emerged that was organised at first within and because of the particular questions related to security, but (ed.) The Culture of National Security: Norms and Identity in World Politics, New York: Columbia University Press 1996, pp. 1-32; Ronald L. Jepperson, Alexander Wendt, and Peter J. Katzenstein, ‖Norms, Identity and Culture in National Security‖ in Katzenstein (ed.) The Culture, op.cit., pp. 33-75; Peter J. Katzenstein, ‖Conclusion: National Security in a Changing World‖ in Katzenstein (ed.) The Culture, op.cit., pp. 498-537; Michael Desch, "Culture Clash: Assessing the Importance of Ideas in Security Studies," International Security, vol. 23, Summer 1998, pp. 141-70 + debate in International Security, vol. 24:1, Summer 1999, pp. 156-180; Ted Hopf, "The Promise of Constructivism in IR Theory." International Security, V.23, Summer 1998, pp. 171200; Dale Copeland, "The Constructivist Challenge to Structural Realism: A Review Essay," International Security, vol. 25, Fall 2000, pp. 187-212; Ido Oren, ―Is Culture Independent of National Security? How America‘s National Security Concerns Shaped ‗Political Culture‘ Research‖, European Journal of International Relations, vol. 6:4, 2000, pp. 543-573. 14 This is as clear in The Culture of National Security, where Katzenstein (―Introduction‖, op.cit, pp. 7-11) argues the strategic rationale of employing a narrow concept of security in combination with the new approaches, because – again – if you can beat the traditionalists in their home field, it is evidently easy later to transfer this gain to the new fields that are already home turf for the new approaches. Implicitly, the question of concepts of security is here reduced to a question of ‗issues‘, whereas the meta-criticism raised by Critical Security Studies, the Copenhagen School and others does not register. This is the approach to the concept that frames the book (it returns on pp. 523-528). Almost as an aside, Jepperson et al (―Norms, Identity‖, op.cit) reflect (pp. 72-75) on the possibility that different theories reflect external developments and suddenly the discussion employs a logic very close to securitisation/desecuritisation when it explains why various issues have been defined as on or off the security agenda as the result of political processes. Similary, Steven E. Miller, ("International Security at Twenty-five: From One World to Another", International Security, Vol. 26, No. 1 (201), pp.--) show in a grand overview of 25 years of International Security, how at best the widening debate register as a question of a re-drawing of the boundary of the issue area and meta-theoretical pluralism is proven by articles on the causal impact of norms, i.e. as a question of what variables to include as independent variable. 15 Stephen M. Walt ‗The Renaissance of Security Studies‘, International Studies Quarterly, 35, (1991), p. 212. 16 Thus, the difference is not primarily about the different fate of ‖the broad concept of security‖ although that is interesting too, and a question that can only partly be explained as a reflection of differences in US and European policy, cf. Robert Kagan, ‖Power and Weakness‖, Policy Review, June & July 2002, Number 113, pp.; Ole Wæver, ‗Widening the Concept of Security -- And Widening the Atlantic?‘ in Bo Huldt, Sven Rudberg & Elisabeth Davidson (eds.) The Transatlantic Link: Strategic Yearbook 2002, Stockholm: Swedish National Defence College 2001, pp. 31-48. This is not the whole story, because much of the literature in Europe is increasingly critical of the broad concept, and the most important debate is therefore whether conceptual reflections on security are seen as a fertile focus for debate or a margingal question. 4 which increasingly influences more general IR debates. In the US, influences clearly went the opposite way: the theoretical positions within security studies derive from general IR debates. This is probably most easily seen in the case of the debate on constructivism. The main salvo from the constructivists, the Katzenstein volume17, was launched explicitly as a move in a general IR debate where constructivists found that it was time to prove their worth on the home ground of realism: security. Most of the contributors were not primarily working in fields or institutions traditionally seen as ‘security‘. In contrast to earlier periods such as ‘the golden age‘ of security studies in the 1950s and 1960s, it is today not particular challenges, needs or debates within security studies that motivate theory development.18 In the context of American IR, this change is valued. It is generally assumed to be a sign of maturity to get away from particular theories and debates in sub-fields and instead develop general theories that are in turn applied to different fields like European integration, international security or trade disputes.19 The new European theories developed in relation to public discussions about security and attempts to develop specific theorising for this purpose. Thus, these theoretical developments were the product of complicated, personal processes of political and theoretical choices and settling or coming to terms with one‘s role in-between academia, expertise and citizen/public intellectual.20 It is often stated that IR research in the US is more closely connected to policy than in Europe21, but this is only partly true: Relevant research is more systematically promoted through various channels in the US, and quite large sub-systems (primarily think-tanks) are very directly linked to policy. Also academic journals like International Security have a more implicit policy orientation expressed by frequent discussions in terms of what ‗we‘ (the US) should do, where equivalents are much more rarely found in any European academic journal. However, the disconnect between large parts of academic IR and the policy circles is also very significant in the US, and European research typically has a broader concept of politics, not only as policy advice (This, I return to in the final paragraphs in a discussion of different understandings of ‗relevance‘.) Here, it probably also plays a role, that the field of practiced policy research in the security field is more closely ‗disciplined‘ and specialised in the US. Therefore there are few constructivists (very few radical constructivists), whereas it is possible to be both a poststructuralist and a policy researcher (and government advisor22) in some European countries (partly because of a different game over the status of different disciplines, partly simply because of less specialisation/division of labour in smaller countries). The bottom-line on this point is that the European theories developed as an integral part of struggling with security issues, the American ones much more detached as part of 17 Peter J. Katzenstein (ed.) The Culture of National Security: Norms and Identity in World Politics, New York: Columbia University Press 1996. 18 The golden age is presented by David Baldwin as the 2nd of 4 phases of security studies, see: David Baldwin, ‖Security Studies and the End of the Cold War‖, World Politics, vol 48, Oct 1995, pp. 117-141. 19 In relation to European integration studies, this argument is made forcefully by Andrew Moravcsik, The Choice for Europe: Social Purpose and State Power from Messina to Maastricht, , NY: Cornell University Press1998, pp. ##. 20 Ken Booth, ‖Security and Self: Reflections of a Fallen Realist‖, i Krause and Williams eds. Critical Security Studies. Minneapolis: University of Minnesota Press 1997, pp. 83-120; Jef Huysmans,'Defining social constructivism in security studies. The normative dilemma of writing security', Alternatives, vol.27, Supplement (2002) pp.41–62. 21 Notably, Stanley Hoffmann, ―International Relations, An American Social Science‖ Dædalus, vol. 106, 1977, pp. 41-60. 22 Aha – that explains recent European policy! ;-) 5 academic debates between various explanatory theories. This in turn is in the American optic the most policy relevant, because the role of the analyst is to provide the relevant knowledge of cause-effect relations that enable the optimal policy decision. Politics and knowledge are not seen as that separate in the European context. Thirdly, the seemingly similar discussions around constructivism in European and American security studies turn out very differently also because of different reference point regarding meta-theory. Realism remains much more central in U.S. security studies than it is in both general American IR/IPE (IO as well as ISQ type)23 and than it is in European security studies. Ironically in relation to the above point about avoiding ‗local‗ theories and participating in general theory debates, security studies has its distinct style. In security studies (as represented primarily by the journals International Security and Security Studies), the dominant form of research is more historical, less oriented towards formal rational choice than in IO-style IR.24 This is not an instance of ‗traditionalism‘ a la second debate (Hedley Bull) where judgement is seen as integral to research25, because a journal like International Security is at least as insistent as IO on strict causal and ‘positivist‘ social science. This shows in the actual publishing record but maybe more revealingly in places like the instruction sheet for contributors. It asks a potential author to sum up (at least for herself) her argument in an arrow diagram because if this is not possible, the argument is probably not clear. This implies that. an argument necessarily takes the form of propositions about cause-effect relations among a few factors (and not e.g. writing structured history, deconstructive interventions or normative IR). The spectrum of theoretical positions probably do not differ much between the US and Europe (and many individuals move back and forth), but the median point does. Therefore, the seemingly similar debates, that could be seen as being about ‗constructivism meets security studies‘ on both sides, turn out very differently. In US security studies, it is about a distinct type of mainstream constructivism that orients itself towards the canons of science among rationalists, where much (but far from all) constructivism in Europe blends in with more radical positions. Accordingy, the debate mostly in International Security over the role of constructivism in security studies26, turns out to be about ―assessing the importance of ideas in security studies‖, i.e. ideas and identity conceived as variables and judged in strict 23 As noted by several authors, (neo)realism has long been present in general IR (not-specifically-securityoriented) discussions mostly as a ghost: Throughout the 1980s and much of the 1990s, a very large part of theoretical arguments were justified as critiques of neo-realism, but to actually find a self-defined neo-realist at an ISA meeting or in the pages of IO, was not that easy. Obviously, this meant that neo-realism was influential in shaping debates but in a much more in-direct way than within security studies, where it remained something like the orthodoxy. 24Michael Brown, Steven E. Miller and Seam M. Lynn-Jones (eds.) Rational Choice and Security Studies: Stephen Walt and His Critics, Cambridge: MIT Press, 2000. Cf. also the statistics in Wæver, ‘The Sociology of a Not so International Discipline‘, especially figure 3 (p. 702) and Table A1 (p. 727). Unfortunately, these statistics include only IO and ISQ, not IS and SS, but I think it is clear that the distribution between different meta-theories in the ‘non-security‘ journals differ markedly from what one would find in the security journals. 25 Hedley Bull, 'International Theory: The Case for the Classical Approach', in World Politics, 3 (1966), 361377. See also the recent attempt to elaborate this methodological stance: Robert Jackson, The Global Covenant: Human Conduct in a World of States, Oxford, Clarendon Press 2000(?). 26 Michael Desch, "Culture Clash: Assessing the Importance of Ideas in Security Studies," International Security, vol. 23, Summer 1998, pp. 141-70 + debate in International Security, vol. 24:1, Summer 1999, pp. 156-180; Ted Hopf, "The Promise of Constructivism in IR Theory." International Security, V.23, Summer 1998, pp. 171-200; Dale Copeland, "The Constructivist Challenge to Structural Realism: A Review Essay," International Security, vol. 25, Fall 2000, pp. 187-212; Theo Farrell, 'Constructivist Security Studies: Portrait of a Research Program', International Studies Review, 4, 1, 2002, pp. 49-72. 6 causal terms. Also the participants explicitly engage in laborious efforts to define a ‗conventional‘ constructivism as distinct from a ‗critical‘ constructivism. The purpose of the present article is to present and take preliminary stock of one of the European research programmes, the so-called Copenhagen School, or rather the part of it that is defined by securitisation theory. It is the hope that this can serve both as an introduction to those un-familiar with the theory and (maybe by reading fast in the first half) as a summing up and ‘where to go from here‘ exercise for the insiders. More generally, this offers a look into the kind of debates that take place in parts of European security studies. It is not uncommon27 in (North?) European debates to see security studies as having three main ‗schools‘: traditionalism/realism, Critical Security Studies28 and the Copenhagen School. The Copenhagen School The so-called Copenhagen School in security studies is built around three main ideas: 1) securitisation, 2) sectors and 3) regional security complexes.29 In this brief presentation, I will focus on the first, because securitisation is what defines most distinctly the school in a 27 Johan Eriksson, ‗Observers or Advocates? On the Political Role of Security Analysts‘, i Cooperation and Conflict, 1999 nr. 3; Richard Wyn Jones, Security, Strategy, and Critical Theory, Boulder CO: Lynne Rienner 1999; Steve Smith, op.cit.; [add more examples] 28 Ken Booth, "Security and Emancipation", Review of International Studies, 17:4, (1991), pp.313-327; Keith Krause and Michael C. Williams (eds.) Critical Security Studies. Minneapolis: University of Minnesota Press 1997; Keith Krause and Michael C. Williams, ―Broadening the Agenda of Security Studies: Politics and Methods‖, Mershon International Studies Review, vol. 40, supplement 2 (1996), pp. 229-254; Keith Krause "Critical Theory and Security Studies: The Research Programme of 'Critical Security Studies'", Cooperation and Conflict, 33:3 (1998), pp.298-333; Richard Wyn Jones, Security, Strategy, and Critical Theory, Boulder CO: Lynne Rienner 1999; Bradley Klein, "Politics by Design. Remapping Security Landscapes", European Journal of International Relations, 4:3 (1998), pp.327-346; Bradley S. Klein, Strategic Studies and World Order: The Global Politics of Deterrence. Cambridge: Cambridge University Press, 1994; Lene Hansen, "A Case for Seduction? Evaluating the Poststructuralist Conceptualization of Security", Cooperation and Conflict, 32:4 (1997), pp.369-97. 29 The name ‗Copenhagen School‘ was coined by Bill McSweeney in a critical review essay which turned into an exchange: Bill McSweeney "Identity and security: Buzan and the Copenhagen school", in Review of International Studies 22:1 (1996), pp.81-94; Barry Buzan and Ole Wæver "Slippery? contradictory? sociologically unstable? The Copenhagen school replies", in Review of International Studies 23:2 (1997), pp.143-52; Bill McSweeney "Durkheim and the Copenhagen school: a response to Buzan and Wæver", in Review of International Studies 24:1 (1998), pp.137-140; Mike Williams "Comment on the 'Copenhagen Controversy'", in Review of International Studies 24:3 (1998), pp.435-441; Bill McSweeney, Security, Identity and Interests: A Sociology of International Relations, Cambridge: Cambridge University Press 1999. In a sociological, descriptive sense, the Copenhagen School is usually taken to refer to the work done since 1985 by the ―European Security‖ research group at the Copenhagen Peace Research Institute, notably its series of collective books: Egbert Jahn, Pierre Lemaitre, and Ole Wæver Concepts of Security: Problems of Research on Non-Military Aspects, Copenhagen Papers no. 1, Copenhagen: Center for Peace and Conflict Research 1987; Ole Wæver, Pierre Lemaitre and Elzbieta Tromer (eds.) European Polyphony: Perspectives beyond East-West Confrontation, London: Macmillan 1989; Barry Buzan, Morten Kelstrup, Pierre Lemaitre, Elzbieta Tromer, and Ole Wæver The European Security Order Recast: Scenarios for the Post-Cold War Era, London: Pinter Publisher 1990; Ole Wæver, Barry Buzan, Morten Kelstrup, and Pierre Lemaitre, Identity, Migration and the New Security Agenda in Europe, London: Pinter Publichers 1993; Barry Buzan, Jaap de Wilde, and Ole Wæver, Security: A New Framework for Analysis, Boulder CO: Lynne Rienner 1998. The most thorough review of the school in this respect is Jef Huysmans "Revisiting Copenhagen: Or, On the Creative Development of a Security Studies Agenda in Europe", in European Journal of International Relations 4:4 (1998), pp.479-506. Who others in Copenhagen or elsewhere to count as ‗members‘ is beyond the reach — and preoccupations — of the current essay. Especially in Scandinavia but also beyond, quite a lot of ‗applications‘ have been done, and in this respect there are many at least part time members. This analytical sense of the Copenhagen School as defined by its theory and approach is the object of this article. 7 meta-theoretical sense and determines its further development, its weaknesses and its political implications. However, it is worth remembering the other key ideas not least because the tensions and interactions between these three explain much of the dynamics in the development of the theory.30 ‗Sectors‘ refer to the distinction between political, economic, environmental, military and societal security. The concept of ‗security complexes‘ points to the importance of the regional level in security analysis and suggests an analytical scheme for structuring analysis of how security concerns tie together in a regional formation.31 In this article, the approach will be outlined briefly and its strengths and weaknesses discussed, both in relation to operational and political implications, and by taking up the quite numerous theoretical and meta-theoretical challenges that have been posed to the approach. Wideners vs. traditionalists — a third way Part of the background of the emergence of the Copenhagen School is the debate in politics and security studies in the 1970s and especially 1980s between ‗wideners‘ and ‗traditionalists‘, ie. the debate over a wide versus a narrow concept of security.32 Should the concept of security be taken beyond its (alleged) traditional focus on military issues and on the state? During the 1980s, this debate became increasingly polarised: The wideners used the narrowness of traditionalists as their main argument for their own position, and notably the traditional position became more narrow than it had actually traditionally been due to the argument that the ‗widened‘ concept of security meant ‗everything is security‘ and therefore it carried a risk of emptying the concept of meaning and ripping the field of its main intellectual tool. The Copenhagen School emerged as one suggestion for a viable middle position. In contrast to the traditionalist view, it is possible to extend the net widely and look for security in all sectors and with all possible referent objects. And against the view of the wideners, it is necessary to be able to discriminate and separate security issues from non-security. Actually, it is only by having a clear sense of what is security, that it is possible to open up without being swept away. In this – seemingly not too hopeful (―oh no, once more!‖) – attempt to return yet again to the discussion of ‗re-defining security‘, it paradoxically became clear that much of the literature on ‗what is security?‘ is really about ‗what is security also?‘; typically: ―it should be ‗not only‘ about military issues, but the environment too‖. Various forms of specifications have been added: common security, environmental security, individual security, but the 30 Sectors and regional security complexes stem from Barry Buzan altough the main reference now is to collective Copenhagen School books (Security, A New Framework from 1998 and Regions and Powers in 2003, respectively). Securitisation comes from Ole Wæver but also here the main reference is now a collective book (Security: A New Framework). 31 The concept of regional security complex was introduced by Barry Buzan in People, States and Fear: The National Security Problem in International Relations, Harvester Wheatsheaf 1983. The concept is at the centre of the most recent book from the project group in Copenhagen: Buzan & Wæver, Regions and Powers: The Structure of International Security, Cambridge University Press forthcoming. 32 Jessica Mathews,'Redefining Security', Foreign Affairs, 68:2 (1989); Johan Galtung, 'Twenty-Five Years of Peace Research', Journal of Peace Research, 22 (1985); Jan Øberg, At Sikre Udvikling og Udvikle Sikkerhed, Copenhagen: Vindrose 1983; Richard Ullman, "Redefining Security", International Security 8, 1 (1983):129153. Anti-wideners include Stephen Walt "The Renaissance of Security Studies", International Studies Quarterly, vol. 35 (1991) pp. 211-239 and Daniel Deudney, "The Case Against Linking Environmental Degradation and National Security," Millennium, 1990, Vol. 19: 461-476. 8 ‗security‘ in these combinations would remain the old one as well as un-reflected as long as there was no focus on security as such.33 But what is security then? The meaning of ‗security‘ is not understood by setting up some ideal definition of how the concept ought to be used (and then possibly criticise practitioners for not being logicians). Rather, one should try to capture the real functions of the term, the powers of the concept, as employed in political practice. Language users implicitly follow rules for what is seen as meaningful statements. This approach does not entail conducting opinion polls and asking people what they think security means, but analysing actual linguistic practices to see what regulates discourse. What do practitioners do in talking security? Security is about survival. In security discourse, an issue is presented as posing an existential threat to a designated referent object (traditionally, but not necessarily the state).34 The designation of the threat as existential justifies the use of extraordinary measures to handle it. The invocation of security has been the key to legitimising the use of force, and more generally opening the way for the state to mobilise or to take special power — eg. using conscription, secrecy, and other means only legitimate when dealing with ‗security matters‘. Therefore, it is implicitly or explicitly part of the securitisation of an issue that a point of no return is postulated – we have to deal with this in time, and because of this urgency, we can not leave it to the normal procedures. Also the threat has a general up-setting potential; it overflows other areas and therefore it should not be weighed and balanced as part of the normal political process. What good is it that we have invested a lot in education today if we fail to take care of defence and therefore we are invaded and lose our liberty tomorrow? Therefore, the security argument goes, sufficient defence is a necessity and thereby not of the same status as other investments. In the traditional perspective, this might be recognised but then seen as a side-effect, a secondary feature of security policy. However, such prioritising and dramatising effects are 33 The few attempts by traditional conceptual analysts — notably David Baldwin ("The Concept of Security", in Review of International Studies, 23:1, pp. 5-26, 1997) — at analysing security are at the same time valuable and instructive in their failure. By approaching ‗security‘ logically, these analyses assume that ‗security‘ has a general meaning independent of its context (i.e. the same in e.g. everyday usage, business and international affairs), and the task is then to analyse logically how states should deal with the aim of ‗security‘ in relation to other aims. This, however, misses the inner relations between ‗state‘ and ‗security‘ and the particular meanings that the concept of security has acquired in the international context, especially since the 1940s. There is not an abstract concept of ‗security‘, which can be applied to the state and the individual alike — when used in international affairs, the concept of security is marked by concepts like state and sovereignty, just as these in turn contain traces of ‗security‘. On the problems of these assumptions in traditional conceptual analysis, see Ole Wæver, "European Security Identities", in Journal of Common Market Studies, 34:1, pp. 103-32, 1996; Jef Huysmans, "Security! What Do You mean? From Concept to Thick Signifier", in European Journal of International Relations, vol. 4:2, pp. 226-255, 1998; and Wæver, ―Security: A Conceptual History for International Relations‖, paper presented at the annual meeting of BISA, December 2002. If approached less through time-less conceptual analysis and more through conceptual history, one can capture the particular meaning that ‗security‘ gained in international relations in the mid-20th Century. The conceptual analysts, on the contrary, by ignoring this – historically contingent – development make arguments about abstract security that ironically often apply to security in all contexts but that of international relations. 34 This section draws on the currently most extensive, operational presentation of the theory: chapter 2 in Barry Buzan, Jaap de Wilde, and Ole Wæver, Security: A New Framework for Analysis, Boulder CO: Lynne Rienner, 1998. The securitisation interpretation of security was first presented systematically in Ole Wæver, "Securitisation and Desecuritisation", in On Security, Ronnie D. Lipschutz (ed.), New York: Columbia University Press 1995, pp. 46-86. 9 systematically involved and it seems that this is central to the political struggles over security – the reasons for resisting suggestions for new security issues or for pushing items on to the security agenda. Therefore, I suggest that this exactly is what the concept does, and therefore also why it is used.35 It is the meaning of security.36 ―Security‖ is the result of a move that takes politics beyond the established rules of the game and frames the issue as above normal politics. Securitisation can thus be seen as a more extreme version of politicisation. Any issue can be located on the spectrum from nonpoliticised (meaning that the state does not deal with it and it is not in any other way made an issue of public debate and decision) through politicised (meaning the issue is part of public policy, requiring government decision and resource allocations or some other form of communal governance) to securitised (meaning the issue is presented as an existential threat). Take the example of gender in the Western world. Until the late 1960s it was mostly seen as belonging to the private sphere and not a proper issue for public policy, ie. non-politicised. The women‘s movement successfully politicised gender issues, and today we see instances – primarily in the US – of gender being securitised (as when some feminists see rape as a sign of men‘s threat to women as such, and when homosexuals become a key element in a perceived threat to a certain white, male, middle class definitions of the American way of life).37 The distinguishing feature of securitisation is a specific rhetorical structure (survival, priority of action and urgency, ―because if the problem is not handled now it will be too late, and we will not exist to remedy our failure‖). This can function as a tool for finding security action in sectors other than the military-political one. As witnessed by the debate on ‗environmental security‘ the criteria for including ‗new‘ sectors is often either whether it has effects on the military one (which logically means not to see the new sector as security in itself) or whether it is ‗as dangerous‘ as war (usually discussed without any precise measure for such comparison across issues). With the securitisation approach one has a clear idea of the chief quality of security and can thus assess whether some question has become a security issue. To register the act of something being securitised, the task is not to assess some objective threats that ‗really‘ endanger some object, rather it is to understand the processes of constructing a shared understanding of what is to be considered and collectively responded to as a threat. The process of securitisation is a speech act. It is not interesting as a sign referring to something more real: it is the utterance itself that is the act. By saying the words, something is done (like giving a promise, betting, naming a ship). By uttering ‖I apologise for my behaviour‖ the speaker actually makes an apology; he does not describe himself apologising for his behaviour. A sentence like ―X is a security question‖ is not a ‗constative‘ 35 This interpretation was originally based mainly on analysis of contemporary security discourse, but it is now backed up by an (emerging) history of the concept: Ole Wæver, ‖Security: A Conceptual History for International Relations‖, paper presented at the annual meetings of ISA (New Orleans) and BISA (London) 2002. This re-tracing of the evolution of the concept since Roman times, shows among other things that there is no stable, self-evident core meaning to security – the concept has changed several times along important axes such as positive/negative, subjective/objective, individual/state and through various changes of delineation vis-àvis neighboring concepts. A particularly important lesson is that it was not a key concept in international relations until the 20th century, and a major change happened in the US in the 1940s whereby the concept gained the particular function of being a call for priority, a conceptual shift partly caused by the erosion of classical concepts like raison d‘état and ‗necessity‘. 36 Quentin Skinner, Visions of Politics 1: Questions on Method, Cambridge: Cambridge University Press 2002, chapters 5 and 6. 37 See Buzan et al, Security, pp. 129-131. 10 but a ‗performative‘ and therefore it does not have truth conditions but felicity conditions (to which I return below). Security is a self-referential practice. It is by labelling something a security issue that it becomes one – not that issues are security issues in themselves and then afterwards possibly talked about in terms of security.38 Thus the exact definition and criteria of securitisation is the intersubjective establishment of an existential threat with a saliency sufficient to have substantial political effects. Securitisation can be studied directly through the study of discourse and political constellations. The study of securitisation does not aim at something ‗hidden‘ like motives or culture – it studies politics directly. When does an argument with this particular rhetorical and semiotic structure achieve sufficient effect to make an audience tolerate violations of rules that would otherwise have been obeyed? This shows that the audience is actually crucial. There are numerous securitising actors who stand up and make securitising moves with reference to some referent object, but a successful securitisation has only happened when the relevant audience accepts the security argument to an extent where this could be used as a basis for using extra-ordinary means at fending off the alleged threat. Key concepts are:  Referent object: what is posited as having a demand on necessary survival and as currently threatened. Note that societies differ in terms of what is generally assumed to have to survive. Most often state sovereignty and national identity are taken to be sufficiently ‘necessary‘ that it is a powerful move to claim they are threatened (which often implies defining them in a particular way). In contrast, it is more varied whether a national film industry is seen as necessary for national identity, or whether the survival of a particular species is necessary and thus a reason for taking extraordinary steps if it is threatened. Securitising actor is the one that makes the argument about a threat to the referent object. Traditionally, the distinction between securitising actor and referent object was not always made, usually because the state was seen as both object and actor. With a wider concept of security, the distinction becomes obviously necessary. It is not the whales who claim that extinction of the whales is unacceptable – it is typically some environmental movement. And when this is noticed, it becomes clear that also the classical cases are about some actor making claims about threats to some other group or principle. Audience is those who have to be convinced in order for the securitising move to be successful. Although one often tends to think in terms of ‗the population‘ or citizenry   J L Austin, How to do things with words, 2nd ed. Oxford and New York: Oxford University Press 1975 (1962). The theory of speech act has since then been developed in roughly two directions. On the one hand, it has been developed in somewhat parallel ways by John R. Searle (Speech Acts: ...) and Jürgen Habermas (Universal Pragmatics ...) as a theory specifying (with elaborate differentiation) the conditions for correct usage of different kinds of speech acts. On the other hand, Jacques Derrida‘s critique of Austin and particularly Searle (Limited Inc ....) started an exploration of the other side of Austin; see Judith Butler, Excitable Speech: A Politics of the Performative, Routledge 1997; B. Honig, ―Declarations of independence: Arendt and Derrida on the problem of founding a republic‖, American Political Science Review, vol. 85:1, 1991, pp. 97-113; and John Forrester, The Seductions of Psychoanalysis: Freud, Lacan and Derrida, Cambridge University Press 1990, especially ch. 7. The second strand emphasises the paradoxes of performativity, the act of creating something out of nothing, the ‗unfoundedness‘ of speech acts and therefore the symmetrical status of failure and success in speech acts, and the necessary and impossible role of essentialism in performatives. Pierre Bourdieu is in some ways a third case to which I return in a later footnote. 38 11 being the audience (the ideal situation regarding ‗national security‘ in a democratic society), it actually varies according to the political system and the nature of the issue.  Functional actors are central actors in a sector who are not involved in securitisation but greatly influence the dynamics of the sector, e.g. an polluting company in the environmental sector. The central role of the audience underlines that the theory is basically about security in an inter-subjective sense. Since Arnold Wolfers‘ seminal piece on ‗security as an ambiguous symbol‘ in 195239, it has been common to discuss ‗subjective‘ and ‗objective‘ senses of security, meaning perceptions of being free from threats versus actually being so. With this baggage, security analysts are inclined to hear the Copenhagen School argument as being one about shifting from objective to subjective. This is, however, not the case. Whether an issue is securitised is not decided by individual perceptions – it is an inter-subjective, political process of negotiating the possible acceptance of a specific kind of argument, a securitising move by a securitising actor. Securitisation and De-securitisation Although securitisation might at first be seen as merely a further intensification of politicisation (often making an even stronger role for the state), in an important sense it is opposed to politicisation. Politicisation means to make an issue appear to be open, a matter of choice, something that is decided upon and that therefore entails responsibility, in contrast to issues that either could not have been different (laws of nature) or should not be put under political control (eg. a free economy, the private sphere, and matters for expert decision). Securitisation means to present an issue as urgent and existential, as so important that it should not be exposed to the normal haggling of politics but should be dealt with decisively by top leaders prior to other issues. Therefore, ‗security‘ should clearly not be idealised. It has often anti-democratic implications. The ‗critical‘ strategy of widening security is quite problematic. It accepts the underlying assumption of the mainstream approach that ‗the more security the better‘ and extends this to still more areas. In contrast, security should be seen as a negative, as a failure to deal with issues as normal politics. Ideally, politics should be able to unfold according to normal procedures without this extraordinary elevation of specific ‗threats‘ to a prepolitical immediacy. Even the securitisation analyst can judge that a situation is so critical that the prioritising imperative of the securitising operation should be used, but this is then a trade-off against other concerns. De-securitisation is the optimal long-range option, since it means not to have issues phrased as ‗threats against which we have countermeasures‘ but to move them out of this threat-defence sequence and into the ordinary public sphere (or the economy, or letting religion be religion40, or what other mechanisms it is then left to). The concept of ‗de-securitisation‘ (and a-security) naturally follows from the securitisation approach. In the traditional perspective, security and insecurity are seen as exhaustive and polar options. The more security, the less insecurity and vice-versa. However, from the securitisation perspective, both are constituted by security framing: Both are answers 39 Arnold Wolfers, ‖National Security as an Ambiguous Symbol‖, Political Science Quarterly, vol. 67:4 (December 1952), pp. 481-502; reprinted in Wolfers Discord and Collaboration. Essays on International Politics, Baltimore and London: The Johns Hopkins University Press 1962, pp. 147-166. 40 Carsten Bagge Laustsen and Ole Wæver, ―In Defence of Religion: Sacred Referent Objects for Securitization", Millennium: Journal of International Studies, vol.29 no. 3, 2000, 705-739. 12 to a question put in terms of security and threats. In the securitisation perspective, the dichotomy turns into a triad: Insecurity is the situation when there is a threat and no defence against it; security is a situation with a threat and a defence against it; but a-security (a situation that has been desecuritised or never securitised) is simply not phrased in these terms, it is not a question of being secure or not and there is not a perception of existential threats being present. Two: security insecurity opposition – but in terms of what? => As all opposites, security and insecurity are unified by that which divides them. => They share the security problematique. Three: Insecurity: + Threat Security: + Threat - Defence + Defence Asecurity: Not seen in terms of threat and defence Figure 1: Security concepts from two to three In a conflict resolution perspective, the way forward is often de-securitisation rather than the production of more security. A striking example of desecuritisation is European integration. The Jean Monnet strategy for Europe was motivated by a worry that Europe might return to a pattern of power balancing and rivalry, but the tool to counter this was not a ‗security system‘ in any traditional sense, i.e. a system that took care of problems defined as security problems. On the contrary, the strategy was to a large extent to shift attention away from security concerns to other issues (especially welfare) and thereby de-securitise relations among the core countries of Western Europe. ―If the countries of Europe once more protect themselves against each other, it will once more be necessary to build up vast armies. (…) Europe will be reborn yet again under the shadow of fear‖, Monnet wrote in 194341. A similarly indirect strategy has worked on border/minority conflicts in Europe both in the mid-2oth Century on conflicts like the Danish-German one, and currently in EastCentral Europe (Hungarian minorities, Baltic-Russian relations, etc). In the latter case, it works via the disciplining power of the EU (and to a lesser extent NATO) setting explicit and implicit criteria for good behaviour. From the applicant side the non-conflict strategy on 41 According to his own Memoirs, Garden City, NY: Doubleday & Company 1978 (French original 1976), pp. 221-222. 13 minorities is in the first phase a question of rational strategy – of for instance Hungary downplaying minority issues in order to become a member (in some versions even: become a member in order to be stronger when finally raising the minority issue in relation to nonmembers Slovakia and Romania). But when playing the EU game of focusing on welfare, the old Monnet strategy works, identities change, and the old issues gradually become non-issues. However, since the mechanism starts as rational policy, it only works for those who see themselves as realistic future members and therefore it works most strongly for the geographically nearest. Thus, important processes of de-securitisation can be observed, but they have so far not the object of much systematic analysis.42 Facilitating conditions of the security speech act Two of the main criticisms of the securitisation approach are that it ignores or underestimates respectively the question of power and the role of ‗objective‘ threats. Both are handled under the category of ‗facilitating conditions‘. This part of the theory dealing with conditions is highly sensitive (because it can lead to a re-introduction of objectivism into the theory) and it is necessary to be very precise about the exact status of the different elements of the theory. Securitisation is ultimately constituted in the inter-subjective realm and therefore even very important conditions for successful securitisation can never replace the political act as such.43 No condition (any number of tanks at the border) or underlying cause (motivation of leaders), not even a solid position of authority of the speaker of security, can make for a securitisation – they can only influence a political interaction which ultimately takes place among actors in a realm of politics with the historical openness this entails. This, however, does not make the conditions uninteresting. On the basis of theories of speech acts, we can say that there are three (in Austin‘s terminology) ―felicity conditions‖ of a successful security speech act – or in more accessible language, three facilitating conditions for securitisation: 1) the demand internal to the speech act of following the grammar of security and constructing a plot with existential threat, point of no return and a possible way out; 2) the social capital of the enunciator, the securitising actor, who has to be in a position of authority, although this should neither be defined as official authority nor taken to guarantee success with the speech act;44 and 42 A promising exception is Isil Kazan, Regionalisation of Security and Securitisation of a Region:Turkish Security Policy After the Cold War, Copenhagen: PhD dissertation 2003. Kazan shows systematically when and why Turkey uses de-securitisation strategies on certain conflicts and not on others. 43 Or at the risk of moving very close to a Schmittian dezisionismus: the very choice about whether to make something a case of security (as in his studies: of friend/foe or of sovereignty) is irreducible. And – in order to avoid simply moving from objective to subjective – it should be stressed that since securitisation is never (in contrast to Schmitt) decided by one sovereign subject but in a constellation of decisions it is ultimately inter-subjective (and truly political in an Arendtian sense). In any case (Schmitt or post-Schmitt) it is irreducible to causal back ground factors. Causes always have to pass a number of decisions of a political nature and the status of ‗security‘ is therefore as socially and politically constructed. On Schmitt and securitisation, see Williams, ―Words, Images‖ op.cit.; Jef Huysmans, ―Security and The Political: The Political Significance of The Figure of The Enemy‖ , forthcoming. [And other articles by Huysmans, Laustsen and others.] 44 The importance of cultural capital for the ability to perform a speech act has been argued by Pierre Bourdieu, Language and Symbolic Power, Cambridge: Harvard University Press 1991 [1982]. A speech act is not only linguistic, it is also social, and depends on the social position of the enunciator, and thus in a wider sense is inscribed in a social field. Later, however, Bourdieu has explained that this was an argument he made in order to 14 3) conditions historically associated with a threat: it is the more likely that one can conjure a security threat if there are certain objects to refer to which are generally held to be threatening-be they tanks, hostile sentiments, or polluted waters. In themselves they never make for necessary securitisation, but they are definitely facilitating conditions. [Maybe insert later an example of each, or rather two of for each of the three factors: one example of success and one of failure for each. Or is this text too clear for that to be necessary?] Especially the third facilitating condition might invite different interpretations. Does this mean to include a kind of ‗rump materialism‘ (Alexander Wendt‘s argument that his theory is not ―ideas all the way down‖ and thus a ‗rump‘ materialism is left as a concession to realism that nevertheless is intended to have little effect because it always enters actual processes together with ideas45)? No, I believe that whereas this argument might be compatible with the kind of theory Wendt develops, it would create logical inconsistency to introduce such limited materialism in the theory of securitisation, because the core of this theory is a political act and this is categorically distinct from material conditions, even if these influence the act. Therefore, the facilitating condition about the designated threat has to be seen in terms of conventions: tanks have a different threat status than pamphlets because of the historical record (which in turn might be more or less closely aligned with physical attributes and capabilities). Deepening our understanding of security dynamics through sectors Does security then mean the same to these different units? Yes and no. Yes, security is a generic term with a distinct meaning, but its form varies. Security means survival, it means ‗this is an existential threat with a point of no return; if we do not handle this in time, if we do not give it full priority, then we will not be here to tackle the other more mundane matters‘. But survival does not mean the same to different kinds of units. Survival to a state means sovereignty. If it is no longer sovereign, it is no longer a state, no longer a subject in the international political system. To a society, survival spells identity. If a society is no longer itself, it has not survived. As with the state: the individuals might have survived as individuals, but the state as state has only survived if it is sovereign, and the community asks whether ‗we‘ are here as ‗us‘ - an identity question. In the economic counter a tendency of some post-structuralists and philosophers of everyday language to make the purely linguistic, internal features of a speech act completely determining. Thus, he accepts the critique by Judith Butler (1996, 199?) that since the speech act needs to include Bourdieu‘s own idea of ‗social magic‘ it has to be indeterminate, open for surprises. The magic whereby some are accepted as holding authority, others not, is not purely a question of a formal position of authority (Austin‘s example where ―I declare you man and wife‖ is an effective speech act only when performed by a properly authorised authority). There is a performative force of the speech act; to use Bourdieu‘s own concepts: it has a magical efficiency, it makes what it says. A speech act is interesting exactly because it holds the insurrecting potential to break the ordinary, to establish meaning that is not already in the context. It reworks or produces a context by the performative success of the act. While it is important to study social conditions of successful speech acts, it is necessary always to keep open the possibility of failure of an act that previously succeeded and where the formal resources and position are in place (the break down of communist regimes in Eastern Europe) and conversely that new actors can perform a speech act they previously where not expected to (the environmental movement) (cf. Butler 1996; Derrida 1977a[1972], 1977b, 1988; Wæver 1995); Jacques Derrida, "Signature Event Context", Glyph. 1 (1977): 172-197. Therefore, the issue of ―who can do security?‖ and ―was this a case of securitisation?‖ can ultimately only be judged in hindsight. (Wæver, et al, Identity, Migration, op.cit., p. 188) It can not be closed off by finite criteria for success. 45 Alexander Wendt, Social Theory of International Politics, Cambridge University Press 1999. 15 field, one should think, the firm would be the primary unit - and survival would mean avoiding bankruptcy. The instructive question, however, is whether security discourse has much meaning here. Firms have no legitimate claim for ‗a right to survival‘, and they can therefore not carry out appeals in the grammatical form of security: Our survival is threatened, therefore we had to break the law ! It is inherent in the logic of capitalism that bankruptcy is possible. Security arguments might be used internally e.g. to justify salary cuts, but they carry no force as such in the society at large. A firm then has to link its fate to issues that do count as security such as secure supply in wartime or cultural reproduction (European film industry). Security policy in the economic field is found in relation to the state: when the state claims that a specific firm or sector is indispensable, and therefore the state is allowed to break rules it should otherwise adhere to, for example, WTO rules against protectionism.46 46 There are more to the economic sector than firms and state/WTO issues, notably questions of globalisation, the liberal international economic order, and development: Buzan et al Security: A New Framework, op.cit., ch. 5. See also: James A. Caporaso, ‗False Divisions: Security Studies and International Political Economy‘, Mershon International Studies Review, 39:1 (1995), 117-22; David Goldblatt, David Held, Anthony McGrew, Jonathan Perraton, " Economic Globalisation and the Nation-State" Alternatives, 22:3 (1997), 269-85; Voctor D. Cha, ‗Globalization and the Study of International Security‘, Journal of Peace Research, 37:3 (2000), 391-403; Jean-Marie Guehenno, ‗The Impact of Globalization on Strategy‘, Survival, 40:4 (1998-9), 5-19; Barry Buzan, 'Economic Structure and International Security: the Limits of the Liberal Case,' International Organization, 38:4 (1984) 597-24; G. Luciani, 'The Economic Content of Security', Jnl of Public Policy, 8:2(1988); Theodore H. Moran, 'International Economics and National Security', Foreign Affairs 69:5 (1990); Aaron L Friedberg, ‗The Changing Relationship between Economics and National Security‘, PSQ, 106:2 (1991); David Baldwin, Economic Statecraft, 1985; Stuart Harris, ‗The economic aspects of Pacific security‘, Adelphi Paper 275, 1993; Tor Egil Førland, ‗The History of Economic Warfare‘, Journal of Peace Research, 30:2 (1993); Theodor Moran, ‗ An Economics Agenda for Neorealists‘, International Security, 18:2 (1993); Joanne Gowa and Edward Mansfield, ‗Power Politics and International Trade‘, American Political Science Review, 87:2 (1993); Beverly Crawford, Economic Vulnerability in International Relations, 1993; Vincent Cable, ‗What is international economic security?‘ International Affairs 71:2 (1995); Beverly Crawford, ‗The New Security Dilemma Under International Economic Interdependence‘, Millennium, 23:1 (1994); Colin Gray, ‗ Global Security and Economic Well-Being: A Strategic Perspective‘, Political Studies, 42:1 (1994); John G. Ruggie, ‗Protectionism and Welfare Capitalism‘, Journal of International Affairs, 48:1 (1994) 1-12; Erich Weede, ‗Economic Policy and International Security‘, European Journal of International Relations,1:4 (1995) 519-37; Beverly Crawford, ‗Hawks, Doves, but no Owls: International Economic Interdependence and Construction of the New Security Dilemma‘, in Ronnie Lipschutz ed. On Security, 1995, ch. ; Robert Latham, ‗Liberalism‘s Order/Liberalism‘s Other: A Genealogy of Threat‘, Alternatives, 20:1 (1995) 111-46; Peter Liberman, ‗Trading with the Enemy: Security and Relative Economic Gains‘, International Security, 21:1 (1994) 147-75; Dale Copeland, ‗Economic Interdependence and War: A Theory of Trade Expectations‘, International Security, 20:4 (1996) 5-41; Sperling, James, and Emil Kirchner (1998), ‗Economic security and the problem of cooperation in post Cold War Europe‘, Review of International Studies, 24:2, 221-37; Jonathan Kirshner, ‗Political Economy in Security Studies after the Cold War‘, Review of International Political Economy, 5:1 (1998) 64-91; Michael Mastanduno, ‗Economics and Security in Statecraft and Scholarship‘, International Organization, 52:4 (1998), 825-54; Beeson, Mark (1999) ‗States, Markets, and Economic Security in Post-Crisis East Asia‘, Asian Perspective, 23:3, 33-52; Kun Young Park and Wang Hwi Lee (1999) ‗The Financial Crisis of 1997-1998 and Its Impact on Security Relations in East Asia‘, Asian Perspective, 23:3, 129-51; Special Issue on ‗Economic Statecraft, Interdependence and National Security‘, Security Studies, (1999-2000) 9:1/2; Special Issue on Trade and Conflict, Journal of Peace Research, 36:4 (1999); Kaplinsky, Raphael, ‗Globalisation and Economic Insecurity‘, IDS Bulletin, 32:2 (2001) 13-24. Thanks to Barry Buzan who has collected most of these references; the same goes for the next footnote. 16 Sector Societal Political/military Economic Environmental Referent Object Nation State Firm 'Nature' Survival Identity Sovereignty No entitlement! Sustainability Figure 1: Dialects of security (main examples) As a final illustration, the environment as security issue shows how this approach offers a new angle on security questions. In most attempts to consider whether ‗environmental security‘ should be taken seriously, it is discussed either whether environmental breakdowns have become an ‗equally big‘ threat as military threats (however that is to be measured), or whether environment and resources in the future can be expected to become a source of military conflict.47 Instead we could focus on the form. Are environmental issues addressed in the form characteristic of security? This is exactly what they are. Environmental activists claim that we face irreparable disasters: if we do not give absolute priority to this, it will soon be too late, and exactly therefore, we (Greenpeace, Earth First, etc.) have a right to use extraordinary means, to depart from the usual political rules of the game. (For instance place metal 47 More extensively on environmental security, see Buzan et al, Security: A New Framework, ch. 4; see also Ullman, "Redefining Security", op.cit., Jessica Mathews,'Redefining Security', op.cit.; Norman Myers, 'Environment and Security' Foreign Policy, 74 (1989); Daniel Deudney,'The Case Against‘, op.cit.; Mark Imber,'Environmental Security', Review of International Studies 17:2 (1991); T.F. Homer-Dixon, ‗On the threshold: environmental changes as causes of acute conflict‘, International Security 16:2 (1991); Shaukat Hassan, Environmental Issues and Security in South Asia, Adelphi 262 (1991); Neville Brown, ‗Climate, ecology and international security‘, Survival, 31:6 (1989); Philippe Sands, ‗Enforcing Environmental Security‘, Journal of International Affairs, 46:2 (1993); Marvin S. Soroos, ‗Global Change, Environmental Security and the Prisoner‘s Dilemma‘, Journal of Peace Research 31:3 (1994); Thomas Homer-Dixon, ‗Environmental Scarcities and Violent Conflict: Evidence from Cases‘, International Security, 19:1 (1994); Andrew Hurrell, ‗A Crisis of Ecological Viability? Global Environmental Change and the Nation State‘, Political Studies, 42: special issue, (1994) 146-166; Simon Dalby ‗Security, Modernity, Ecology: the Dilemmas of Post Cold War Security Discourses, Alternatives, 17:1 (1992) 95-134; Patricia Mische, ‗Ecological Security and the Need to Reconceptualize Sovereignty‘, Alternatives, 14:4 (1989) 389-28; Nina Græger, ‗Environmental Security?‘, Journal of Peace Research, 33:1 (1996) 109-16; Deepak Lal, ‗Eco-fundamentalism‘, International Affairs, 71:3 (1995) 515-28; Marc Levy, ‗Is the Environment a National Security Issue?‘, International Security, 20:2 (1995) 35-62; Eric Stern, ‗Bringing the Environment In: The Case for Comprehensive Security‘, Cooperation and Conflict, 30:3 (1995) 211-37; Monica Tennberg, ‗Risky Business: Defining the Concept of Environmental Security‘, Cooperation and Conflict, 30:3 (1995) 239-58; Carsten Rønnenfeldt (1977) ‗Three Generations of Environment and Security Research‘, Journal of Peace Research, 34:4, 473-82; Shaun Breslin, "The China Challenge: Development, Environment and National Security", Security Dialogue, 28:4 (1997), 497-509; Alan Dupont, (1998) ‗The Environment and Security in Pacific Asia‘ Adelphi 319; Raimo Väyrynen, ‗Environmental Security and Conflicts: Concepts and Policies‘, International Studies, 35:1 (1998), 3-22; Simon Dalby, ‗Ecological Metaphors of Security: World Politics in the Biosphere‘, Alternatives, 23:3 (1998), 291-319; Jon Barnett, Jon ‗Destabilising the environment-conflict thesis‘, Review of International Studies, 26:2 (2000), 27188; Karin Dokken, ‗Environmental security and regionalism in the Asia-Pacific: is environmental security a useful concept?‘, The Pacific Review, 14:4 (2001) 509-30; Narottam Gaan, ‗Rethinking Security: The Environmental Approach‘, International Studies, 38:3 (2001) 299-310. 17 spikes in precious forest trees killing the forest workers – not usually an appropriate procedure in democratic politics, but then this is a security issue, the argument goes.) It is implied in the argument that when something is a security question, there is a ‗point of no return‘, and we have to make sure it is not crossed - therefore our issue cannot enter the usual balancing of interests in relation to other concerns. Why make these sectoral distinctions, why judge, for example, whether the nation or the state is the referent object? Often it is the state elite that uses references to ‗the survival of the nation‘, so the actor and the purpose can be the same whether that elite makes references to state or nation. The crucial factor here is, however, that these references each involve a specific language game, each have their characteristic logic. Having said ‗identity‘, certain moves are more easily justified than others - other steps follow when it is claimed that the environment or the sovereignty of the state is at stake. Of course, there will be concrete cases where several are combined, but there is reason to expect, that security action has different attributes if conducted with different referent objects. A major importance of the theory is to show the effects of securitisation. When an issue is securitised, this has implications both ‗internally‘ (for instance by inhibiting debate and democracy) and ‗externally‘ by often stimulating conflict, security dilemmas and escalation. However, different referent objects engender different dynamics. The clearest example of this in the writings of the Copenhagen School so far is probably the later (postIdentity, Migration ..) writings on ‗societal security‘. The Copenhagen school has been criticised for singling out ‗identity‘ questions as a separate ‗sector‘ when identity construction is obviously involved in the constitution of any referent object for security.48 The point, however, is that it makes a difference if a threat is defined in terms of identity, i.e. it is articulated as a threat to ‗identity‘. Due to the paradoxical nature of identity, a defence of identity sets off complicated and often self-defeating processes. Identity is never something one simply has or is; identity discourse is about what one could be or should be if only …; what I shall have been. The identity is thus a fantasy of a completeness that is not present, and the lack is constitutive of a subject and its identity.49 Therefore, to raise an issue about a threat to one‘s identity, in one sense is a way to stabilise one‘s identity, by producing a threat, an Other and thereby the explanation for the lack; however, it also points to the problematic character of one‘s identity, and thus produces more insecurity. Therefore, attempts to ‗defend security‘ in the societal sector (i.e. where security is directly articulated in identity terms) tend to be self-defeating but thereby also self-reproducing. There are no direct instruments of societal security policy, and the analysis of societal security should first of all serve to avoid the triggering of these problems in the first instance.50 48 Iver B. Neumann, "Identity and the Outbreak of War : Or Why the Copenhagen School of Security Studies Should Include the Idea of 'Violisation' in Its Framework of Analysis", International Journal of Peace Studies, 3:1 (January 1998), pp. 7-22. 49 Ole Wæver, Concepts of Security, University of Copenhagen 1997, pp. 328ff, 344ff and 360. 50 Ole Wæver, ‗Insécurité, Identité : une dialectique sans fin‘ in Entre Union et Nations: L’État en Europe, ed. by Anne-Marie Le Gloannec, Paris: Presses de Sciences Po 1998, pp. 88-137. (English version as ‗Identity and Insecurity Unlimited‘ as chapter 10 in Wæver, Concepts of Security, University of Copenhagen 1997.) On the question of security dilemmas and security systems in the societal sector, see also Wæver, ‗Identity, Integration and Security: Solving the Sovereignty Puzzle in E.U. Studies‘, in Journal of International Affairs, vol. 48:2, Winter 1995, pp. 389-431, and ‗European Security Identities‘ in Journal of Common Market Studies, vol. 34:1, March 1996, pp. 103-32. See also: Paul Roe, "The Intrastate Security Dilemma: Ethnic Conflict as a 'Tragedy'", Journal of Peace Research, vol. 36:2, 1999, s. 183-202; Paul Roe, ―Misperception and ethnic conflict: Transylvania‘s societal security dilemma‖, Review of International Studies, vol. 28:1, 2002, pp. 57-74; Pierre Hassner, ‗Beyond Nationalism and Internationalism‘, Survival, 35:2 (1993); Gidon Gottlieb, Nations without 18 Such insights into the particular dynamics of one form of security – in this case societal security – is of practical importance because the purpose of the theory is not only a general ‗critique‘ of securitisation , but to be able to analyse political situations and (in the classical tradition of security studies) to contribute to the ‗management‘ of difficult situations. Ultimately, the interest is not so much in single instances of securitisation but their interrelatedness, ie. relational security such as security dilemmas, security complexes and other security constellations (cf. the ‗soft realist‘ tradition of Herz, Wolfers, Jervis, Buzan and early Booth). To be able to contribute to avoidance of escalation and security dilemmas and possibly create constructive dynamics, the theory needs to be able to understand and to some extent predict patterns securitisations as shaped by the different kinds of referent objects and the different sectors. Understanding the nature of identity helps to find characteristic patterns in the societal sector. Why then do we take the world apart into sectors? To get at the characteristic dynamics, referent objects, securitisation actors and security dilemmas in the different sectors.51 It is then necessary to put the picture together again, and here the clout is that in some sense the actors do this for us, because it is inherent in the logic of security to weigh threats against each other (it is about prioritising, and you can not claim that everything is most urgent) and about interpreting one threat in the light of another (this water issue is a security threat because they also threaten us militarily).52 The cutting up into sectors does therefore not mean that one can study a concrete situation in one sector only. It is crucial when one wants to capture the dynamics of a situation – the vicious circles, dilemmas and spirals – to see also security dilemmas across sectors. When putting the picture together, however, it is useful to have the clearest possible understanding of the characteristic dynamics of the different kinds of security issues. But the real threats ….? Things exist. And what happens happens. This often makes critics say, that when you see something nasty happen, this proves that those hit were previously threatened. However, it is not possible to conclude from later events to the appropriate threat assessment at an earlier stage. In the traditional theories, this is acknowledged and discussed in terms of probabilities. What does the September 11 terror attacks imply for an accurate threat assessment? That this had for a long time been a greater threat than understood? That approaching September 11 it moved towards 100%? But if this had been the threat assessment, then it had maybe not happened (because it could then have been prevented). What had the threat then been? The whole idea of measuring threats in themselves is problematic, because the assessment itself influences the threat – to the extent that it is acted upon, and it always is to some extent because even if no particular steps are taken, the simple alertness will make a difference. States‘, Foreign Affairs, 73:3 (1994); Ted Gurr, ‗Peoples Against States: Ethnopolitical Conflict and the Changing World System‘, International Studies Quarterly, 38:3 (1994); Kristian Gerner, ‗From the Black Sea to the Adriatic: Ethnicity, Territoriality and International Security‘, Security Dialogue, 24:1 (1993); Roxanne Lynn Doty, ‗Immigration and National Identity: constructing the nation‘, Review of International Studies, 22:3 (1996) 235-55. 51 This sectoral differentiation also means that quite a lot of traditional security theory can be re-imported as contingent dynamics that hold true under given conditions; this is most clearly visible in the ‗military sector‘ chapter in Buzan et al, Security: A New Framework, op.cit. (ch. 3). 52 Buzan et al Security: A New Framework, op.cit. ch. 8. 19 Things exist, but they do not come with labels on. It is a political choice whether something should be handled the security way or with ‗ordinary means‘. German Nazism in the 1930s was a threat of a magnitude that if other countries assigned it to normal procedures it was not countered, and securitising it might therefore have meant a more efficient response. But in other cases where securitisation is possible, it is not always necessary or preferable: Although some industrial development might have adverse environmental impacts or globalisation a challenge for traditional religion, it might be that both the environment and faith can be saved by using in the one instance ordinary environmental policy making with economic incentives and in the other a purely spiritual re-wakening. Both could have been countered also by securitising them and taking drastic and enforced measures in the one case and maybe fundamentalist extremism in the other, but it is impossible from this to say whether these were security threats – they could have been handled so, or not. Whether one is the correct labelling and the other wrong can not be decided because history can not be played out several times. It is politically chosen which of the possible futures are realised. Security is always about the future. It is often discussed as if it is an attribute of the present: are we secure? What are the threats? Of course, this is a preferable terminology for positivist, empiricist and materialist approaches, because the future is less tangible than the present. Actually security is a dual statement about the future. What happens: 1. 2. if we do not act (the threat). if we act as I recommend (defence). Whether the issue is securitised or not has implications for how it is handled and how other actors react. Therefore, it is important whether an issue is securitised or not. But this also implies that it is impossible to ‗test‘ whether the securitisation was ‗correct‘ – to securitise or not creates different realities. What we can study is this practice: who can ‗do‘ or ‗speak‘ security successfully, on what issues, under what conditions, and with what effects? The more conceptually aware parts of the mainstream literature defines security in terms of probability. Often with some variation on the basic formula of probability times impact, i.e. a threat is bigger the more likely but also the ‗larger‘ it is. A similar conception is used in the traditional literature on risk – not the recent ‗risk society‘ sociology, but the more conventional works on e.g. road safety, food safety, etc. They have the advantage over international security studies to be able to work from statistics on past events and even comparisons before and after some measure is taken, e.g. road accidents with or without seat belt legislation. However, as convincingly shown by critics from within this tradition such as John Adams in the excellent book ‗Risk‘53, these statistics are flawed in numerous ways of which the most serious is that people react to the counter-measures. And since most people do not aim at zero risk, they often compensate – eg. safer cars make people drive faster, and when the adrenalin factor is included: if traffic becomes too safe more might take to dangerous sports. The idea of measuring risks independently of actions taken – and thus independently of measuring – simply does not work, even in these much more benign contexts. In IR and strategic studies, it has rarely been suggested that threats could literally be quantified, but a lot of the mainstream literature actually operates as if it could be. One argues in an ad hoc way about more or less security, about threats that are large, small or nonexistent, and alternative approaches are criticised for not delivering the measures of the new 53 John Adams, Risk, London: Routledge 1995. 20 forms of security, that the traditional approaches actually never have devised themselves for the old security. This is, however, not as big a loss as it might sound. Often, it is not politically very helpful to try to define ‗real security‘ outside of the world of politics and then teach the actors to understand the term correctly. Classical realism was often attentive to this. Universalist prescriptions for how one ought to think, and what the other side should or should not be provoked by, is politically dangerous. If one wants to avoid security dilemmas and unnecessary conflicts, one has to understand how the other side actually thinks and what is likely to trigger drastic reactions.54 This art of diplomacy was always based on principles very close to securitisation theory. What research questions can the theory deal with? Quite a lot it seems! There is by now a surprising amount of empirical studies done with the full or partial use of the securitisation theory. These do not follow a standardised format and it is possible to focus on different phases of the process and different levels of aggregation. Thus, my list here will not be exhaustive but only point to some characteristic forms of investigation from this starting point. The theory does not point to one particular type of study as the right one. The theory rather operates as a conceptual apparatus and with this a number of different kinds of questions can be asked. Optimistically, the diversity is a sign that the theory has a relatively clear core idea and sufficiently explicit conceptualisation, that it can generate/structure different kinds of usage and even produce anomalies for itself in interesting ways. The many critiques of the theory are in my view (mainly!) a sign of strength. You don‘t criticise a theory that is so vague that it does not do much. If the theory is distinct enough, you can produce precise problems and these are then interesting too – and only possible to get to by starting from this theory. 1. General trends in the transformation of security.  Is there a shift among sectors, eg. with less military securitization and more environmental? The comparative examination of all regions in Regions and Powers for instance shows the great variation from the low saliency of military security in North America and Europe to the much higher in eg. the Middle East and East Asia (not surprising in itself, but it can not be detected by an approach that equals security with political-military security). These changes can not be quantified in a mathematical sense, but when using the categories of the theory systematically, the procedures are tight enough that at least we could force surprising and un-wanted results on ourselves when we wrote Security: A New Framework for Analysis: We had e.g. expected much less economic security (due to the above argument about competition being part of the game in capitalism, but we found in the economic (and the political) sector much more widespread securitisation on behalf of principles and thereby typically system-level referents like ‗the liberal international economic order‘, ‗free trade‘ and ‗human rights‘.55 54 Henry A. Kissinger, A World Restored, Boston 1957; Herbert Butterfield, ―The Tragic Element in Modern International Conflict‖ (1950) reprinted in Butterfield, History and Human Relations, London: Collins 1951, pp. 9-36. 55 Buzan et al Security, op. cit., see e.g. pp. -- - -- - - -- 21  What is the trend in terms of referent objects? Is less security with reference to the state and more for instance about the nation or say the EU or ‗the liberal international economic order‘? If so, this might carry implications beyond security analysis and be a way of mapping the political lay-out of a region: what exists with the saliency of the ability to become the referent object for security, ie. with an effective demand for survival? It is a familiar IR argument that the criteria for existence as an international political unit is ability to defend oneself, and in increasingly post-sovereign conditions especially in Europe, a map of securitisation thus serves to measure degrees of ‗unitness‘ and thus the nature of the emerging political landscape.56 Securitising actors – are these to an increasing or decreasing degree state representatives? (In stead of deciding that theoretically as it is most often done in the wide/narrow debate, it can be studied empirically with the securitisation approach.) Not many convincing empirical studies have nevertheless been conducted to my knowledge – probably because most scholars fall back on the convenience of focusing on state leaders in macro studies and in micro studies the relative and comparative aspect is lost. However, it would be interesting to see case studies in fields where state and non-state actors compete for the power to securitise and de-securitise. Are threats territorial or de-territorialized, and thereby: do geographic patterns reproduce consistently across sectors or do ‗regions‘ (and thereby our own theory of regional security complexes) fall apart? The regional patterning of security – the basis for regional security complex theory (RSCT) – is a product of geography. Despite globalisation, it is still the case that most threats travel more easily over short than long distances. This is clearly valid for military threats, but often also for identity, migration, many political threats but not nearly as much for economic security. From the perspective of RSCT, deterritorialisation is the most important possible long-term effect of globalisation. On the one hand, a number of leading scholars of globalisation agree that de-territorialisation is the defining feature of globalisation57, and on the other hand territoriality is a necessary assumption of RSCT. Thus, far-reaching globalisation will eventually undermine RSCT, but it does not seem reasonable to depict the situation today or tomorrow or in 20 years in terms of that much de-territorialisation. In lieu of this radical change, globalisation will nevertheless have effects that can be studied within RSCT. The effects defined in terms of relative de-territorialisation will show themselves partly as an increase in non-territorially defined referent objects, partly as an increasing importance of great powers with regiontranscending military power and thereby the formation of more super-complexes (like today the Asian one). Much security theory implicitly assumes radical de-territorialisation (despite a resistance to other sweeping effects of globalisation). However, the current degree of globalisation is enough to add non-territorial patterns to the territorial ones, but territorial dynamics are far from gone.   2. Analyse constellations. As in traditional security studies – notably the ‗soft realist‘ tradition from Wolfers, Herz and Kissinger over Jervis to early Booth and Buzan – a major aim of studying security is to grasp the interactions and vicious circles among actors. In this respect, secutitisation study is simply a sophisticated security dilemma analysis. As the classical realists, it does not ask when 56 57 Ole Wæver, ―European Security Identities‖, op.cit.; Wæver, Concepts of Security, ch. 11. Scholte, Globalization, 2000; Held et al, Global Transformations, 2000 22 people should feel secure but when they do, and it tries in the classical diplomatic tradition to locate the thresholds and ‗soft spots‘ that differs from state to state (to other actor) and should not be universalised away. A specific advantage of securitisation analysis is to capture such dynamics across sectors (eg Syria-Turkey on water and secessionism). When all major security connections have been drawn up and the zones of relative indifference located in order to separate regions from each other, the relational analysis ends up with security complexes, see 4. 3. Early warning. Securitisation is often the step before violence (or in Iver Neumann‘s term: violisation58). Extreme action is often justified with existential threats and thus attention to securitisation could warn us where violent conflicts could loom in the future. When a conflict has turned violent – and has been deeply ideologised or ethnified – it is often quite limited what can be done until it has run its course. If the study of securitisation can help us to act earlier in the process, this could potentially be important. 4. Detailed study of securitising moves and processes of de-securitisation. There is a need to – but obviously also a framework ready for – study in more detail of specific securitising moves including some that fail, and possibly some that succeed only partially. Also processes of de-securitisation have not been sufficiently studied, but most of the categories for such study should be available in the framework. [I am aware of a number of ph.d. students currently working on such detailed projects on very different types of cases from the Chinese communist party to the political situation in Colombia and vigilante groups in South Africa. Some of their findings should eventually be included in this paragraph.] 5. Silence and non-securitisation. As shown recently by Lene Hansen, the approach can analyse situations where securitisation is surprisingly absent, maybe systematically absent because the victims can not securitise a threat without increasing their problems (eg. violence against women in Pakistan or rape in the Balkan wars).59 At first her argument might seem to be (and probably is intended as) a criticism of the approach because the one thing a securitisation analyst cannot say is ‗this really is a security issue‘ (or at least he or she can only do this as a securitising actor, not as an analyst). The big contrast to Critical Security Studies is that the analyst cannot step in on behalf of actors who do not speak security and tell that really their main security problem is this or that, only they suffer from false consciousness. Speaking from some general emancipatory ideal, Critical Security Studies can deal with exactly the blind spot of the Copenhagen School and thus the two might be complementary. Thus, pointing to ‗silence‘ is a critique of the Copenhagen School. However, it would seem that the finding of a conspicuous absence of securitisation is possible to locate exactly by starting from the securitisation approach. There is nothing in the theory against studying practices and 58 Iver B. Neumann, "Identity and the Outbreak of War : Or Why the Copenhagen School of Security Studies Should Include the Idea of 'Violisation' in Its Framework of Analysis", in International Journal of Peace Studies, 3:1, pp. 7-22, 1998. 59 Lene Hansen, ―The little mermaid‘s silent security dilemma and the absence of gender in the Copenhagen school‖, Millennium, vol. 29, no. 2 (2000), pp. 285-306 23 processes of silencing. Lene Hansen has other critical points in the article and I return briefly to these below. 6. Regional security complex theory. This is really an elaborate version of type 2, the analysis of constellations. Imbedded in more general arguments about international history and international relations theory, a specific method and vocabulary has been devised for the analysis of security with the regional level in focus.60 Recently, a comparative analysis of all the regional security complexes has been completed61, but there is still much room for unfolding such analysis through book length studies of individual regional security complexes (or individiual insulator states placed inbetween these62). More immediate political usage As already hinted at, one of the main worries often raised about the securitisation approach is the abdication from speaking objective security.63 Traditionalists are worried that the core of the (sub-)discipline is and should be the supply to decision makers of an objective analysis of threats and strategies. Critics traditionally have taken the position of criticising the official view of security with reference to what really are the security problems – the Soviet threat is exaggerated, the risk of environmental breakdown has to be given first priority, etc. The securitisation approach in contrast studies the processes whereby issues are turned into security concerns or purged of this quality. However, this does not mean that the securitisation approach is without the possibility of criticising existing or suggested policy or in other ways advising, warning or suggesting. First, the main explicitly political application of securitisation studies is (as suggested above) in relation to the management of security constellations (the updated form of classical realist diplomacy). This produces an inclination against too aggressively questioning the existing securitisations, because a major task is to assist in manoeuvring in relation to security worries as they be. In one respect, this makes the Copenhagen School less ‗critical‘ but also makes it less universalist. Again the classical realist impulse shows in the concern not to prescribe others their ‗correct‘ thresholds for security concerns which could easily amount to yet another self-interested idealism.64 (We will return below to the need in certain situations to engage self-definitions more critically.) Second, the most general political relevance of the approach is in heightening a political sense of responsibility in practitioners and analysts alike. It is not innocent to talk security – you are never only commenting on something that ‗is‘ a security issue, it is your choice to securitise and this is a political act. What it asks of anyone deciding whether to 60 In practice, there are problems with using fully the securitisation approach on large-scale analysis. It is not possible to do the detailed mapping of all instances of securitisation. This has to remain the principled metatheory and the tool for investigating the main tricky cases, whereas aggregate analysis of a whole security complex (or the world) has to operate on the received images of what are the main security issues on the agenda – not study the formation of each. 61 Buzan and Wæver, Regions and Powers, op.cit. 62 Kazan, op.cit. 63 Even Bill McSweeney who generally criticises the School for not being sufficiently constructivist presents also a longing for traditional security, where ―the state would intervene and speak objective security for the society‖, McSweeney, ―Identity and Security‖, op.cit., p. 88. 64 E. H. Carr, The Twenty Years’ Crisis …. 24 approach a question in security terms is not to find out whether this is ‗correct‘ or not, but to try to think through the consequences of securitising or not. This means both the effects on the matter at hand – is say this particular environmental problem best and maybe fastest dealt with under a heading of environmental security or as for instance sustainable development; and the in-direct effects on social and political structures, notably democracy and identity formation. Thirdly, it is possible to use the approach to contest a speech act directly. Not by telling what the objective situation is, but by holding the securitising actor to the conditions for securitisation, particularly the criteria of ‗existential threat‘. In a sense this role means to act as a part of the audience. From this position one can ask the securitiser to spell out: how is this an existential threat? Why is there a point of no return, why urgency, why is it necessary to do X?65 Finally, it is possible – but never un-problematic – for a securitisation analyst to securitise. One acts then in the role of securitising actor, not as an observer of security practices. This seemingly simple option of exchanging the role of analyst with that of actor is however ridden by a paradox: If the securitisation analyst decides to participate in the securitisation of something – on the basis of the combined effects and possibly genuinely worried about some development – should he or she then include the essentialisation and depoliticisation usually implied? It is part of the defining quality of securitisation to claim the necessity of some action, the unavailability of choice. But the securitisation student who chooses reluctantly to engage in securitisation will do it with an awareness that it is not necessary in the full and absolute sense of the word; it is not because the issue is a security threat, only that he/she tries to define it as one. Thus, the securitisation analyst who acts as securitiser is most likely not terribly good at it. The one thing the securitisation approach cannot do is to say ―this is not a security issue – that is‖. Because the designation is a political, intersubjective quality, it is not something held by the object. Any given issue can be discussed including the likely effects of handling it one way or another. The security label as such, however, belongs to the political realm. In American security studies, it seems that the question of political relevance is posed – for better and worse – in one distinct format: Security is a goal that society wants. Therefore, this aim should be defined as clearly as possible, and public debate as well as decision making processes should be as rational as possible, and this includes being informed by the knowledge generated by security experts. The conceptual analysis part of research can help to clarify logical and definitional matters, but the main task for security experts is to supply relevant knowledge of cause-effect relationships that enable efficient and successful policies. The pressure for political relevance that the Copenhagen School is in a European context exposed to raises a much broader agenda that includes much more the role of public intellectual and participant in societal discussions about ends not only means. Some of these discussions ultimately become choices about ‗who we want to be‘ because of the self-effects of securitisation/de-securitisation and thus the full range of ethical issues are raised. Strengths, weaknesses and further development of the theory 65 See Williams, ‘Words, Images‘ for an interesting attempt to develop this aspect with the help of Habermasian Diskursetik. 25 In addition to those critical points raised in the previous section by way of what can and can not be done politically within the theory, there are criticisms of three kinds: internal problems discovered by employing the theory, critical comments and suggestions that can relatively easily be accommodated, and finally truly external criticism that claims to raise fatal objections and to which the theory must try to answer. First there are some internal problems to be worked on. Particularly, the concepts of ‗audience‘ and ‗extra-ordinary measures‘ need better definition and probably differentiation. For this purpose, more thinking could be helpful, but not least more case studies, and especially studies from more diverse cultural settings in order to develop a more general formulation of these concepts. The present version risks being too tied to a Western democratic form of politics. In principle, the audience does not have to be ‗the public‘ or ‗the people‘, it can be the relevant group that needs to be convinced – that be the party elite in East Germany in 1989 or a particular subculture in the US, but the broader concept is not well enough developed. Similarly, political style varies between countries and cultures, and therefore it is probably more common to invoke ‗existential threats‘ in some places than others, and empirical study must show how to handle this.66 Finally, there are in practice often several audiences involved in the same operation. For instance, the American case for missile defence has to be made both domestically to justify huge spendings and controversial policy decisions but also internationally not least to justify the revoking of an international arms control treaty (the ABM treaty).67 When for a while, the US largely succeeded domestically but not internationally to make the case that a threat from ‗rogue states‘ justified these acts, this had important effects on internation dynamics, because the US administration was freed from domestic constraints but did not have acceptance internationally for its unusual behaviour. A more sophisticated construction of categories of audience would make the theory much stronger. In relation to extra-ordinary measures, the challenge of the movable line between ‗normal‘ and extra-ordinary between countries has been addressed in the presentation of the theory68, but the operationalisation remains insufficient. Furthermore, there is a need for more work on de-securitisation and failed acts of securitisation, and finally partial securitisation could be worth exploring. This is probably increasingly relevant in Western Europe. For instance the Danish involvement in the security of the Baltic states is not about existential threats to Denmark, but the issue has been given some security rationale. Still, this is certainly not enough to lift it out of normal politics – quite the contrary, much of this co-operation is rather de-politicised and ‗technical‘ due to its low urgency.69 The currently quite absolutist concept of ‗securitised or not‘, might be differentiated through empirical studies of mixed and partial situations. In the application on concrete cases, quite hopeless debates often emerge on whether something is ‗ordinary‘ or ‗extra-ordinary‘. The theory operates with a category of institutionalised securitisation that is able to handle many of these cases in theory, but in practice it becomes too elastic. Much of defence policy for instance is basically rule following 66 These criticisms, I have mainly learned from my many astute students – several participants in the course on ‗the concept of security‘, Spring 1999 and Fall 2002 and individual projects by Andreas .., Tobias .., Asser .., Thomas C. [references to be completed] 67 Pia Pinkowsky, Securitization som process, Copenhagen: MA thesis, Department of Political Science 2001. Simon Bonde, USA og slyngelstaterne – en konstruktivistisk analyse af amerikansk sikkerhedspolitik i 1990erne, Copenhagen: MA thesis, Department of Political Science 2001. 68 Buzan et al, Security, op.cit., pp. 24f. 69 Ref to MA thesis from Aalborg (?+ student from the military academy?) 26 and not very dramatic, but deep down in the history of the institution, some huge security arguments have established the military institution and it rests on a security rationale that needs not be spelled out very often. Probably, it will be helpful in the future if the criteria to apply in specific instances is less the extra-ordinary nature of particular measures (because hard to make precise) but rather the threat construction as such and the argument about necessity. The extra-ordinary measures cannot be left out of the theory, but the focus of empirical investigations should be on the rhetorical structure of statements more than an institutional history of particular measures. Wouter G. Werner has pointed out that the original formulation does not clearly distinguish between a regulated, exceptional situation and a radical exception that includes violation of all regulation including rules for exceptions.70 This needs to be dealt with and might even contain some elements to the solution of the above problem. If the concept of ‗extraordinary measures‘ contain distinct sub-categories, it might be easier to develop degrees of securitisation without watering out the theory.71 The quite extensive debate on the Copenhagen School contains a number of suggestions that can easily be accommodated and simply leading to extensions of the theory in ways that do not violate the core. Johan Eriksson has pointed to the affinity to theories of agenda setting.72 He presents it with a critical edge, because he wants to develop a broader concept of threat politics of which securitisation is only one particular type. If, however, one wants to focus on securitisation – due to the assumption that it has distinct features that allow for more of a theory – there is nothing in the Eriksson suggestion that prevents this. Iver B. Neumann has suggested the addition of a concept of violisation,73 and this in itself is unproblematic, although some of his suggestions for the implications on securitisation of this might be more doubtful.74 Michael Williams has made an interesting analysis of the whole problematique of securitisation and desecuritisation according to which de-securitisation of identity was part of the general modern ‗package‘ with state, sovereignty, ontological materialism and epistemological positivism which all constituted a security strategy to end the religious wars and civil wars of the 16th and 17th Centuries: If security could be confined to military issues defined by the state on the basis of observable threats, there would be much less war and violence than if all actors were allowed to securitise religion, identity and various other issues.75 Thus, the new theorists of security should think twice before un-packing this order because more violence might threaten. This interesting argument poses a reverse challenge to the mainstream: it offers the best argument so far against widening (much better than the circular argument that security is about the state and military and should not be extended beyond this, because security is about the state and military). However, this offer to the traditional approach is double-edged because it can only be received if one thereby also 70 71 Wouter Wener, ‖Securitisation and Legal Theory‖, COPRI Working Paper 27/1998. This in turn might be linked to the meta-debate around Carl Schmitt that has emerged in relation to the school (Michael Williams, Jef Huysmans) and on performativity (Lene Hansen drawing on Judith Butler). See below. 72 Eriksson, ‘Introduction‘, op.cit. (+ more texts by him and others around him including other chapters in Threat Politics, op.cit.). 73 Neumann, ‖identity‖, op.cit. 74 Wæver, ‖Krigens Sikkerhed‖, Politologiske Studier, vol. --75 Michael Williams, "Comment on the 'Copenhagen Controversy'", Review of International Studies, 24:3 (1998), pp.435-441; Michael C. Williams, "Security and the Politics of Identity", European Journal of International Relations, 4:2 (1998), pp. 204-225. 27 admits that the narrow concept of security (as well as positivism and materialism) are political (well motivated) choices, and this is problematic because the discourse upholding these positions are usually not one about choice, but one about non-choice, about this being the way it is and has to be.76 Mike Williams as well as Lene Hansen have argued the importance of going beyond speech acts in the narrow sense and thinking about the politics of security in an age of visuals, media and images.77 This again is probably a non-radical extension rather than a challenge to the core: it is possible to think about meaning effects in other media than language. This should not be done by levelling differences and just extending speech act logic to all communication and create one un-differentiated space, but parallel logics can be developed while respecting the importance (i.e. non-neutrality) of the different media. Turning now to the external criticism, the most famous is of course Bill McSweeney‘s. The core of his critique is an alleged reification by the Copenhagen School of referent objects, not least in the societal sector, which deals with identity and thus nations and other communities. Up to a point, McSweeney‘s criticism can be re-defined as an overly dichotomous opposition between the study of constitution and operation of identities and other social categories78: It is totally valid and relevant to study the origins of national identities, states, companies, environmental units or any other referents for security action and thereby expose their contingency and possibly try to re-articulate them. But in many cases, they are likely to remain strong referents for a while – be they the idea of a Palestinian people or the French state – and there is a real task for security analysis in trying to analyse the interaction of security action on behalf of these quite solidly sedimented social phenomena. McSweeney obscures this by presenting somehow identity as very constructed and other referents – such as states – as more innocent ‗things‘. For a constructivist, it is – or should be – impossible to create a dichotomy between the fully constructed and thus totally fluid on the one hand and the given on the other hand. In the words of Derrida: ―You cannot object to a unity simply because it is the result of a process of unification (…). [T]here are no natural unities, only more or less stable processes of unification, some of them solidly established over a long period of time.‖79 When phrased like this, much of the criticism is mis-guided. However, the challenge is real for two reasons. One is that there are many conflict situations where improvement – not least de-securitisation – will demand a re-definition of identity and self-conception among the participants, and thus McSweeney‘s strategy becomes relevant to the Copenhagen School agenda. Secondly, constitution and defence can not be kept separate because often the practice of securitising and defending something is strongly involved in re-producing or even producing referents.80 Therefore, this issue has to be explored and a more dynamic and recursive conception of referent objects included into the theory. 81 76 77 Richard K. Ashley, -- chapter in Booth, Smith, Zalewski ? and/or unpublished paper from 1996-97? Williams, ‖Words, Images‖, op.cit., and Hansen, ‖The Little Mermaid‖, op.cit.. 78 Buzan and Wæver, ‖Slippery, …‖, op.cit. 79 Interview with Jacques Derrida: ‖The Deconstruction of Actuality‖, Radical Philosophy, vol. 68, 1994, p. 41. 80 200 references to post-structuralist studies of identity in international relations --- Campbell etc etc etc 81 Much of the debate between McSweeney and the Copenhagen School addresses the question of methodological collectivism (and Durkheim). The Copenhagen School is insistent in avoiding the ontological prioritising of individual security by much of Critical Security Studies, and the Copenhagen School insists on the importance of being able to operate with collectives (like state and nation) to understand the inter-unit dynamics that are not possible to derive from individuals. McSweeney (‖Durkheim‖, op.cit.) raises some relevant questions to the exact formulations of this doctrine, and this part of the debate deserves more attention. 28 Secondly, Jef Huysmans and Janus Mortensen have raised the critique that the specific speech act form of security is formalised into a stable core meaning of security.82 Even if it is correct that ‗securitisation‘ captures an important social mechanism in the postwar period and it has been a useful analytical instrument, the theory has to remain open to the possibility that the concept changes again and leaves the securitisation speech act behind. This is correct, and I try currently to explore this as part of a larger investigation of the history of the concept of security including its possible current transformations due mainly to the logic of ‗risk society‘.83 An important criticism has not been much pushed by the critics, but it ought to: Especially in relation to ‗societal security‘ a political dilemma has emerged. Whenever you let out a concept, you have to be prepared for it to take on a life of its own. The concept of ‗societal security‘ (i.e. identity security) easily lends itself to anti-immigrant, anti-EU and similar exploitation. One can they counter that this is a mis-understanding of the concept, but to some extent this misses the point. One has to accept the power of words, including their power to liberate themselves from their creator. Thus, the act of creating theory is as risky as any other political practice – you never know what harm you end up doing. ‗Societal security‘ has particular problems due to its possible legitimation of any defence of ‗our identity‘, but all this writing about security is in itself a problem.84 Even when one writes about security with the aim of achieving de-securitisation or to sensitise everybody to the problems of securitisation, one securitises by way of putting these issues in security terms.85 An illustration of such ironic effects from our very own IR world can be Richard Ashley‘s seminal attack on neo-realism, ―The Poverty of Neo-realism‖.86 This article surely contributed a lot to developing the critique of neo-realism, but in order to make the critique, Ashley invented the term neo-realism and in general constructed a coherent school. In doing this – not least in giving it a name – he might have helped and strengthened neo-realism more than he weakened it by his intellectual attack.87 All this amounts to an argument for not writing too much about security and in particular an argument against terms like ‗societal security‘. These problematic effects, however, have to be balanced against the potentially 82 Huysmans, ‖On the creation …‖, op.cit.; Janus Mortensen, Fred og sikkerhed i et politisk-filosofisk perspektiv, Copenhagen: Politiske Studier 1998. 83 Wæver, ‖Security: A Conceptual History …‖, op cit; see also Mikkel Vedby Rasmussen, , ‖Reflexive Security: NATO and International Risk Society‖ i Millennium, vol. 30:2, 2001, pp. 285-309; Shlomo Griner, ‖Living in a World Risk Society: A Reply to Mikkel V. Rasmussen‖, i Millennium, vol. 31:1, 2002, pp. 149160; Mikkel Vedby Rassmussen – reply to Shlomo Griner, forthcoming. See also on risk society amongst many other sources: Niklas Luhmann: ―Risiko und Gefahr‖ i Soziologische Aufklärung 5. Konstruktivistische Perspektiven. Westdeutscher Verlag, Obladen 1990, pp. 131 – 169, Luhmann, Risk: a sociological theory, 1993; Ulrick Beck, Risk Society. Towards a New Modernity, London: Sage 1992; Zygmunt Bauman, Community: Seeking Safety in an Insecure World, Cambridge Polity Press 2001; Anthony Giddens, Modernity and SelfIdentity: Self and Society in the Late Modern Age, Cambridge: Polity Press 1991; François Ewald, ―Insurance and Risk‖ in The Foucault Effect: Studies in Governmentality, Graham Burchell, Colin Gordon and Peter Miller eds., Chicago: The University of Chicago Press 1991, pp. 197-210; François Ewald, ‖Two Infinities of Risk‖ in Brian Massumi (ed.) The Politics of Everyday Fear, Minneapolis and London: University of Minnesota Press 1991, pp. 221-8; Claudia Aradau, ‖Beyond Good and Evil: Ethics and Securitization /Desecuritization Techniques‖, Rubikon e-journal, Dec 2001: http://venus.ci.uw.edu.pl/~rubikon/forum/claudia2.htm 84 85 Jef Huysmans, ‖Security, what do you mean? …‖, op. cit The strongest version of this argument is Huysmans … Alternatives … I have discussed this at somewhat greater length in ‗Securitising sectors‘, op.cit. 86 Richard K. Ashley, "The Poverty of Neorealism," International Organization 38 [Spring 1984]: 225-86. 87 Like ‘neo-realism‘, also ‘the English School‘, ‘the Copenhagen School‘, and ‘neo-liberal institutionalism‘ are names coined by critics – and ultimately immensely helpful to the object of attack. 29 positive effects of such writings. It is the aspiration of the theorists of security to contribute to an increased understanding of security dynamics and thus increase the ability to handle risks in this area. Ultimately, this comes down to a race between on the one side the rational development of intellectual instruments for understanding and intervening in a field, and on the other the unconscious of language where the word travels independently of the logical structures one wanted to contain it in. Possibly, it is an arrogant optimism when we in the Copenhagen School so far continue to believe that our analyses do enough on the first account to risk the hideous dynamics of the latter. (If we understand better the thresholds and dynamics that drive securitisation and thereby security dilemmas and other escalations, we should be better prepared for pre-empting some conflicts and managing other.) Another way to present the on-going debates and signal the kinds of critique, the Copenhagen School meets, is to briefly survey the main ‗neighbouring‘ approaches (or competing critical perspectives, if one so wishes): 1. The Bourdieu inspired works by Didier Bigo and his collaborators (The Paris School?) [+ sometimes Jef Huysmans] is close to the Copenhagen School in many ways but differ in a few.88 Bigo shows how internal and external security merge as agencies compete for the gradually de-territorialised tasks of traditional police, military and customs. Also they jointly produce a new threat image by constantly connecting immigration, organised crime and terror. In-security is largely a product of security discourses and security policy (which contrary to Bigo‘s claims is actually very close to the Copenhagen School conception). The main difference is that this Bigo school starts from a conception of a ‗field‘ and its actors and ask what they do, whereas the Copenhagen schools starts from a particular form, a rhetorical structure, and asks who are doing this. If done simplistically, the actor based approach becomes easily something close to conspiracy theory. But by now this work has evolved into a very elaborate and unusually well documented mapping of practices notably also at the micro level by the various agencies involved on the security field. An important advantage of this approach is that it better includes routine practices and even deviation from official policy, i.e. it is less oriented to discourse and more to all practices of agencies. In some ways, this is simply complementary to the Copenhagen School and the two approaches can fruitfully be brought together in concrete contexts, but partly it constitutes a challenge for the Copenhagen school in relation to the study of the practices of relevant actors beyond their rhetorical moves in public. The Copenhagen school in turn poses a challenge to Bigo & Co. regarding the processes whereby actors create the symbolic power which their analysis takes as starting point. 2. Critical Security Studies is commonly seen as having two wings: a more Frankfurt School inspired emancipatory side represented by Ken Booth and Richard Wyn Jones, and a post-structuralist side represented by Mike Williams, Rob Walker, David Campbell and others. For the present purpose, this will be sub-divided even further. First the emancipatory school. This has been touched upon already and can simply be summarised as the challenge 88 Didier Bigo, Polices en résaux, l’expérience européenne, Paris: Presses de Sciences Po 1996,, "Security(s): Internal and External, the möbius ribbon", paper presented at the annual convention of ISA, Toronto March 199? (is published, but where?), ―Security and Immigration: Toward a Critique of the Governmentality of Unease‖, Alternatives, vol. 27: supplement, Feb. 2002, pp. 63-92; Bigo chapter in Kelstrup/Williams book, ‗L‘Europe de la sécurité intérieure: penser autrement la sécurité‘, in Anne-Marie Le Gloannec (ed.) Entre Union et Nations, Paris: Presses de la Fondation Nationale des Sciences Politiques 1998, pp. 55-90; Ayse Ceyhan, ‘Analyser la sécurité: Dillon, Waever, Williams et les autres‘, in Culture et Conflits no. 31-32, Automne-hiver 1998; + Ceyhan and others in Alternatives, vol. 27: supplement, Feb. 2002; Huysmans, ―Defining Social Constructivism‖, op.cit. 30 from CSS to the Copenhagen School what to do with actual threats to the security of individuals that are not articulated politically and therefore not securitised and the question about the ultimate purpose of it all (if not emancipation, then what?)89. The Copenhagen School in turn asks difficult questions to CSS about the dangers of privileging individual security which both excludes a full understanding of the functioning of collectives90 with sui generis reality and risks repeating a contractarian Hobbesian logic.91 3. Within Critical Security Studies, especially RBJ Walker, Michael Williams and (parts of the work by) Keith Krause emphasise political theory questions in relation to the early modern (largely Hobbesian) ‗solution‘ around the concept of sovereignty which implied a narrow concept of security and especially the exclusion of culture from security.92 This argument raises some troubling questions both for its own work and for ours, because it means that un-packing this modern order by opening the concept of security and reconnecting identity and security implies to risk some gains in terms of peace and order associated with the modern state. However, these processes are unfolding and it is not a real option to put them back in the old order but more realistic to work on the involved deep questions of political theory to develop new political conceptions. Therefore, it is necessary for the Copenhagen School to deal with the question posed by Walker, Williams and others about security and the political: The Copenhagen School artificially isolates questions about ‗who can speak security‘ while this necessary raises broader question about ―who we think we are and how we might act together‖93, that is the subject of politics and of security. 4. Many post-structuralists (whether called Critical Security Studies or not) often focus their work on showing how dominant constructions of security were brought about, e.g. studies of self-other relations, of collectives like ‗the West‘, and of particular threats.94 Often this comes surprising close to the American constructivists because it does not deal with the concept of security as such, but the construction of this or that, and/or it shares with the emancipatory part of CSS a contrast between dominant construction of threat and security (typically by the state) and ‗real‘ security interests of the people. However, it does have a critical case against the Copenhagen School regarding the problem of projecting identities in a too stable way, where the post-structuralists want to emphasise the fluid and constantly performative character of identity, cf. the discussion above in relation to McSweeney. 5. However, a third and final form of post-structuralist work is the radical postmodernism of scholars like James Der Derian, Mike Dillon, Constantin Constantinou and 89 90 Cf. Booth [several texts] op.cit. and especially Wyn Jones who discusses the Copenhagen School explicitly. Wæver, ‖Securitization and Desecuritization‖, op.cit. and Concepts of Security, chapter 2; Buzan et al, Security: A New Framework, chapters 1 and 9. 91 Krause and Williams, ‖Broadening‖, op. cit. 92 Keith Krause and Michael C. Williams, ―Broadening‖ op cit.; Krause and Williams, ―From Strategy to Security: Foundations of Critical Security Studies‖ in Krause & Williams (eds.) Critical Security Studies: Concepts and Cases, London: UCL Press 1997, pp. 33-60; Keith Krause "Critical Theory and Security Studies: The Research Programme of 'Critical Security Studies'", Cooperation and Conflict, 33:3 (1998),pp.298-333; Williams, ―Comment‖, op.cit and ―Security‖ op cit.; R B J Walker, ―The Subject of Security‖, in Krause/Williams, pp. 61-82. 93 Walker, ‘The Subject‘, p. 66. 94 Bradley S. Klein, Strategic Studies and World Order, Cambrdige Univesity Press 1994.; Simon Dalby, Creating the Second Cold War, Londno: Pinter 1990; Bradely Klein, ‗Politics by Design. Remapping Security Landscapes‘, European Journal of International Relations 4 (3) (1998), 327-346; David Campbell, Writing Security: United States Foreign Policy and the Politics of Identity, University of Minnesota Press 1992 and not least the epilogue to the second, revised edition (1998, pp. 207-227); chapters in Krause/Williams by Dalby and Klein; articles by Jim George, etc etc etc. 31 Andreas Behnke.95 By linking more directly to Nietzsche and Heidegger, they question why at all we are concerned about security. Only the timid and unambitious strive for security – better is to live interestingly and less predictably. Also the point by Derrida and others about the typically modern longing for fixity and predictability can be used here. Better would be to face the Other and what is truly different as an exciting challenge. However, this is often a somewhat difficult message to bring to both citizens and policy makers concerned about e.g. weapons of mass destruction – have fun and enjoy the challenge. Also, it is paradoxically a somewhat apolitical and secure (!) theoretical position detached from actual political struggles and withdrawn to meta-politics. Thus, it might be argued that although their position is at first the orthodox post-structuralist one, the less ‗correct‘ one of the Copenhagen School actually involves itself in on-going politics and difficult dilemmas in a way that better lives up to the ethical injunctions from post-structuralism. However, these ‗hardcore postmodernists‘ do pose an important question about ‗living with in-security‘, although it is also raised in other contexts including the discussion on conflicts between liberty and security and parts of the risk society literature. 4. Feminists have done much work on security thinking.96 Often, the usual divisions within feminism (standpoint, Marxist, liberal, post-modern, etc) show up in this work too, and accordingly different parts line up close to some of the already mentioned schools, while other parts are more distinct. Quite a bit of the feminist work comes close to Booth type CSS (which is not surprising given the strong influence of feminism on Booth‘s thinking): individual security should be given priority, state security is over-emphasised by traditional, masculinist scholarship, and sometimes it is also stressed how the very forms of theory and study is male science.97 Other parts of feminist work is more post-structuralist and stress the articulation of concepts of gender, nation and security (often inspired by Elshtain)98, some combine this with securitisation theory.99 Feminists have also contributed importantly to the literature on performativity. The Copenhagen School needs to do some work at a relatively abstract level of theory in relation to the question of performativity. Currently, this is defining/distinguishing for the theory in relation to more traditional constructivist approaches: Even with the concept of facilitating conditions, the theory is not causal in a traditional sense, because securitisation is conceptualised as a performative act never exhaustively explained by its conditions. It does more than realise already given potentials and produces genuine novelty, in what Bourdieu calls an act of ‗social magic‘ something happens at this exact point and therefore the act can never be reduced to a transmission belt in causal chains.100 However, the way the Copenhagen School uses this concept is problematic because the school has paid insufficient attention to the constitution of subjects. Especially in Butler‘s formulation, there is no given 95 James Der Derian, ‖The Value of Security: Hobbes, Marx, Nietzsche, and Baudrillard‖ i Ronnie D. Lipschutz, ed. On Security, New York: Columbia University Press 1995, pp. 24-45; Michael Dillon, Politics of Security: Towards a Political Philosophy of Continental Thought, London: Routledge 1996; Costas Constantinou, ―Poetics of Security‖, Alternatives, vol. 25:3, 2000, pp. 287-306; Anthony Burke, ―Aporias of Security‖, Alternatives, vol. 27:1 (2002), pp. 1-27; Andreas Behnke, ―‘Postmodernizing‘ Security‖, paper presented at ECPR Joint Session of Workshops, Mannheim, 26-31 March 1999. 96 J. Ann Tickner, Gernder and International Relations: Feminist Perspectives on Achieving global Security, New York: Columbia University Press 1992. Add many more ref. 97 Cf. Booth, ‖Security and Self‖, op.cit. 98 Lene Hansen, Rape/Bosnia article in Feminist Journal, several articles in special issue of Cooperation and Conflict especially the Swedish one on conscription, etc etc. 99 Karen Lund Petersen, traficking in women, Cooperation and Conflict. 100 (Derrida 1977, 1992; Weber 1995; Butler 1997; Campbell 1998:25-8; Wæver 2000b) 32 subject that acts – it is acting that constitutes the subject. Also there is no singular and deliberate act but a nexus of power and discourse. I am inclined here to defend the singular act somewhat more, but the constitution of agents in the act has to be accounted for much better in securitisation theory. 7. Finally, there is the question what critical comments are raised by more mainstream, traditional or rationalist scholars. This is difficult because most of it is based on mis-representation of securitisation theory, typically through interpretation through other theories that the critics are reminded of.101 It usually goes the indirect way through general attacks on constructivism, on post-structuralism, on allegedly ‗ignoring real threats‘, on the wider concept of security, and even – quite astonishingly – for dropping the state. Hopefully, meaningful debates will emerge in the future also on this axis. Conclusion The Copenhagen School is often seen mostly as a theory of ‗the politics of security‘, i.e. of ‗agenda setting‘ and thus a possible addendum to more traditional security studies that study the issues as such. However, this is too limited for two reasons. First, the theory has implications of a more radical nature and while it explicitly aims to re-integrate the traditional agenda, this has to happen in a re-formulated way in order to achieve theoretical consistency. Secondly, it offers novel insights to the handling of security issues as such and it is therefore not external to ‗substantial‘ issues. It can be combined with a number of different types of research where it by offering criteria for what to count as security helps to delineate empirically the security issues that are dealt with in different ways by different other theories. But it contains in itself also an understanding of how security operates. It makes an important difference when an issue is securitised, i.e. turned into a security issue. Therefore, it is an important element in understanding the management of threats, escalations and to achieve conflict resolution and also holds implications for the societal organisation/institutional structures for security. A focus on security as ‗labelling‘ should not belittle the importance of security (―it is only a matter of words – security is just a label‖) – it is highly consequential when something is securitised. Securitisation/desecuritisation is an important element when attempting to deal with the challenges that other theories will declare to be the ‗threats‘ and the ‗real security issues‘. What has been done above to assess the theory and its relevance, I will finally try to structure in terms of the answer to four questions: 1) Is it a fruitful research program?, 2) How much of what is generally seen as security study does it cover?, 3) What are its implications for other theories? 4) What kind of theory is it? 1. A fruitful research program? It has stimulated a lot of activity and it is widely seen as a useful framework for empirical work. However, it is striking how differently it is used: it can be made to structure empirical work of very different types and with a focus on different elements of the theoretical network of inter-related concepts. Is this a weakness? Maybe, in relation to expectations about the emergence of ‗normal science‘ it is, because then one needs ‗paradigms‘ in the sense of ideal cases of ‗how to do it‘. But in terms of theory as here conceived ―a picture, mentally formed, of a bounded realm or domain of activity‖, one that 101 Olav F. Knudsen, ‖Post-Copenhagen Security Studies: Desecuritizing Securitization‖, Security Dialogue, 32(3), September 2001, pp. --; Raimo Väyrynen …; etc. 33 ―depicts the organization of a realm and the connections among its parts‖ (Waltz), it is unproblematic that one can ask different kinds of questions guided by this theory and do different kinds of empirical work – it might even be seen as an advantage, because the studies of different kind (micro and macro, securitisation and desecuritisation, etc) can be added up to a fuller picture of the whole. More problematic are those weaknesses in the basic conceptual apparatus (extraordinary measures, audience[s?]) mentioned above. It is crucial that the emerging waves of work actually contribute to sharpening the theory on these points. 2. How much does it cover then of what is traditionally seen as security studies? One thing (question 1) is to ask, how well the theory does what it does, another is how much of ‗our work‘ can be handled this way. As mentioned earlier, the theory has a certain ability to subsume much of traditional security studies such as theories of arms races, deterrence, economic sanctions, as well as not-so-traditional security studies with theories of nationalism and environmental conflict as long as these are re-formulated on more constructivist terms. However, the questions are then first whether there are parts of traditional studies that do not enter the securitisation picture, and second whether there are theories or knowledge that so to say ‗don‘t work‘ when reformulated. To the first question, a preliminary answer would be that securitisation theory can only get a grip on security dynamics that explicitly include an element of articulation of threats and/or security, whereas eg. economic theories about the impact of defence spending on growth do not operate via such mechanisms. The calculations as such will not gain (or lose) anything by being put in the new framework, and only sometimes will a part of the analysis be about the security policy conclusions drawn from arguments of this nature and then securitisation theory will help. The second question asks whether there are elements of traditional security theory that only works on objectivist assumptions and therefore get lost if transferred to a securitisation framework. So far, I am not convinced that this is the case, rather that there are arguments about cause-effect relationships of various kinds that are un-affected. Thus, it seems so far that it is possible to create an encompassing framework that puts the security specific element in securitisation theory, but allows for lots of relevant knowledge that enter unchanged from traditional theory. Or to put it differently: there are big parts of the theoretical and empirical knowledge in security studies – about military technology, economy, environment, and other questions not including the politics hereof – that are not covered by securitisation theory, and where it can not in fairness claim to be an alternative to traditional theories. 3. What are its implications for other theories? This partly follows from the previous point. The answer can be put in terms of three groups of theories. First those that are compatible with the theory because un-affected, second those that are compatible because the theories can engage constructively with each other (cf. the preceding section of debate among various critical ‗schools‘), and third those that are questioned by the theory of securitisation. In other words: ‗who should feel threatened by the Copenhagen School?‘ – who are likely to securitise securitisation theory? The most radical effect follows from the argument made about the basic character of security as inter-subjective and not objective (or subjective). This implies the impossibility of talking about what is and is not a security issue and with any degree of precision about how secure someone or something is. To recapitulate briefly: this impossibility has two sources. One is the problem that all assessments of risk influence behaviour and thereby invalidate themselves (cf Adams‘s critique of risk research and the use of probabilities), the other is the more specific problem that ‗security‘ is a label with specific effects and therefore it is a political move to employ this term (or something equivalent). This argument is a problem to much policy research because it makes exactly this kind of arguments: that this is a security problem, that is a bigger security problem than something 34 else, and we would be more secure if we did so. However, the problematisation of objective security is not nearly as much of a challenge to most of contemporary, academic security studies, because as noticed in the introduction, it often does not use the concept of security much, only as a name for an issue area. (However, an interesting exercise remains in thinking through to what extent neo-realism as a theory is affected, because it claims security to the driving motive of states. Does it therefore create bigger problems for defensive realism that sticks to security all the way than for offensive realism that derives a logic of power from a basic security motive? Or is this problem fictitious because neo-realism – offensive, defensive or any other kind – is not about motives or domestic politics but about structural pressures and thus it is not really ‗security‘ in this specific mid-20th Century sense we are talking about here but rather a concept of unit survival which is defined internally to the theory, not derived from public usage of the term security?) 4. What kind of theory is securitisation theory? 102 It is not a general theory with a candidacy at ‗great debate‘ level, nor is it an issue specific theory because it has implications beyond any narrowly conceived field. It is an ‗idea theory‘ that unfolds potentially far into other areas. It is built from one core idea, that security should be conceived as a speech act, and works out from here towards meta-theory, empirical studies, etc. In contrast to the standard image of theory development in IR, it does not start by meta-theory as a religious question about conviction and personal stand-point. That is actually what has kept the Copenhagen School core team together against all odds. The core team was the structural realist Barry Buzan, the post-structuralist Ole Wæver and usually some liberalist co-authors, Pierre Lemaitre, Morten Kelstrup and/or Jaap de Wilde. We did not start by trying to settle meta-theoretical issues or work out a compromise position – we decided on what theory to use in any given study and then followed through on that. At some kind of pragmatic level we share an attitude to theory and science: each theory is always one big experiment. What Niels Bohr argued in nuclear physics is true for social sciences: you can not conduct an experiment and then end up with a statement about reality isolated from the experiment – you can only understand that which you learn about together with the experiment. Similarly, any theory creates certain stories, makes some things stand out, and even its failures and difficulties are enlightening, but only if you understand how they are a product of this theory, not external and somehow ‗belonging‘ to a rival theory. If e.g. the theory of regional security complexes runs into particular problems with Turkey as an unusual kind of insulator, the conclusion is not necessarily that this ruins the theory but that it could be because Turkey is a very interesting and unique country and this is brought out. It is this reality that has a problem, not the theory. Such insights are only gained by trying to cultivate the inner logic of a given theory. The Copenhagen School aims not at the consistency of individuals, but of theories. In our 1990 book European Security Order Recast, we used basically neo-realist theory, and whatever developments we made were discussed in terms of their compatibility with and soundness for neo-realism. Similarly in Identity, Migration from 1993, we worked on how to develop traditional security theory to encompass identity issues, not by using the new challenges as an argument for skipping the old.103 102 In addition to the ‘idea‘ or ‘inner logic over meta-theory‘ feature of the theory, it is also unusual in an American context for at least one more reason: It approaches its key concept – ‗security‘ – in the unusual way of drawing it from practice, not abstract definitions and even imbedding the whole analysis in a kind of conceptual history or genealogy by seeing the concept as evolving over time and the concept not only as an analytical tool but also an important practice to be studied. 103 See Lapid and Kratochwil, op.cit., for a discussion of the different strategies for reacting to the challenge 35 Returning to the arguments made in the opening paragraphs about how to discuss theories and research programmes. With this view of theories as experiments, it is not too helpful to use general, disciplin-wide meta-theoretical debates as starting point and settle these first. Rather the challenge is to cultivate the inner logic of a theory, refine it, apply it and finally evaluate how interesting and relevant the insights are – both the positive ones that it speaks proudly and the in-direct lessons from its failures. Or maybe the two are not that easily differentitated – or in the words of Niels Bohr:‖If you have a correct statement, then the opposite of a correct statement is of course an incorrect statement, a wrong statement. But when you have a deep truth, then the opposite of a deep truth may again be a deep truth".104 [Word count 15.614 (including footnotes)] from culture and identity when security theory in the 1990s faced nationalism, minority conflicts and migration. 104 Quoted by Heisenberg "Theory, criticism and a philosophy",p.31-46 in: "From a Life of Physics. Evening lectures at the International Centre for Theoretical Physics, Trieste, Italy" Wien:IAEA. 36