Why have some states acquired nuclear weapons whereas others have abandoned their nuclear aspirations – although they possess the technical and economic capabilities needed to weaponize?1 Even today, six decades into the military application of nuclear technology, this question remains one of the central puzzles for the discipline of International Relations (IR). Based on the predominant theoretical thinking which is largely shaped by (neo-)realist conjectures, many IR scholars and policymakers alike tend to assume that nuclear proliferation is inevitable, given the anarchic nature of the international system. Nuclear proliferation is thus traced back to “a fallen humanity’s raw quest for power” or is explained by “vague references to security dilemmas and the capacity for evil that lurks within all of us,” as Hymans has put it so vividly.2 However, this reading obscures the fact that there are a large number of nuclear-capable states pitted against a still relatively small number of de-facto nuclear-weapon states. Without lapsing into naive nuclear optimism, it should in fact be acknowledged that of the approximately 50 states that are capable of building nuclear weapons at the beginning of the twenty-first century, fewer than ten currently command a nuclear force. Arguably, this figure still represents a grave danger to international peace and security; yet it also suggests that the picture of nuclear proliferation is more complex and less clear-cut than is often assumed. At the same time, traditional theoretical approaches to international security fail to offer a convincing explanation for this nebulous picture. While both realist and non-realist scholars alike have successfully interpreted some of the broader trends and generalized developments, their accounts often remain vague and underspecified – or fail to deal with important exceptions in crucial cases. Above all, neither the realist focus on security rationales, nor the non-realist focus on prestige, domestic-bureaucratic pressures and economic imperatives successfully explain why some states not only refrain from nuclear weapons acquisition, but even roll back programs they had previously started. All too often, hazy generalizations and grand, abstract theories inhibit a more profound and detailed knowledge of the very political processes that lead towards nuclearization or nuclear reversal. The absence of pandemic nuclearization or at least widespread proliferation is a puzzle that has hardly been solved by contemporary approaches to IR. Consequently, the aim of this study is to elaborate a theoretically innovative and

practically useful framework that eclectically draws on different IR approaches and thereby allows us to deepen and supplement our understanding of the “causes” – motives, reasons, objectives or purposes – behind nuclear restraint. It is argued that to realize this aim we must not rely on an analysis of external threats that are allegedly givens, or objectively measurable security dynamics or predetermined interests. Instead, we need to open the “black box” of the state and try to understand how – in a deeply political process – narratives and frames regarding its identity, threat perception, preferences or position in the international system emerge and shift.3 These narratives embody the cognitiveideational basis for state action. They offer, in other words, a coherent template for the understanding of a certain event and provide orientation for further action in a complex societal situation. Analyzing the dominant frames and narratives thus allows us to comprehend how actors perceive themselves and their environment and how they sketch possible courses of action. Accordingly, the underlying analytical question is which interpretations of the self, of others, of the security environment or of the “value” of nuclear weapons constrain and enable states to pursue or abandon nuclear weapons programs? Rather than treating the state as a discrete given entity that acts according to a clear set of (externally given) national interests and needs, the study thus pays attention to the multiple ways in which a state’s “place in the world” is discursively constructed. It is these very social-psychological representations that shape “who” and “what” states are, how actors perceive their environments and counterparts, and which policies are to be pursued in order to confront a challenging situation. The study is thus based on an understanding of international security that regards neither states’ nor individuals’ threat perceptions as given, but instead investigates the interpretive processes that both construe these very actors and make certain actions or practices possible in the first place. With respect to nuclear weapons, this implies that the search for “apparent” material or systemic causes of proliferation and nuclear armament is – if only temporarily! – suspended; the analysis of beliefs and narratives gains center stage. We need to investigate the way in which armament decisions are based on a political “story” that gives meaning to actors and objects – be they conventionally treated as either material (the international environment or system) or ideational (a state’s identity). After all, intersubjectively shared notions of self and others as well as ideas of the international environment enable and legitimize a state’s policies. This study does not, therefore, primarily seek to falsify and supplant existing theoretical explanations of nuclear proliferation and nuclear reversal. Rather, it aspires to supplement the conventional “toolbox” by adding an alternative analytical “instrument” to enable us to better understand how political beliefs or narratives emerge and how they constitute policies. To do so, I seek to bridge paradigmatic chasms and to draw upon and synthesize different existing approaches.4 The theoretical starting point for this study is a rereading of the rich yet partially buried sociological and social-psychological contributions made by American pragmatists. Their work helps us comprehend the underlying dynamics that

bring about the intersubjective establishment of shared meanings and identities. Unlike customary, predominantly cognitive approaches to international politics, which often consider rationally acting individuals to be the unquestioned core of their research, a pragmatist-inspired analysis provides a broader perspective on social phenomena. It transcends the idea that human beings merely act rationally on external stimuli. Rather, pragmatist scholars claim that human action is better understood as a process of continuous interaction between human actors and their environment – and as a permanent practice of interpretation, in which human beings give meaning to the components of their environment. With concrete reference to the proliferation puzzle, a pragmatist notion of political action leads us to assume that a state’s proliferation policy does not depend on purportedly given objective threats or on the given conditions of the international system. Rather, from a pragmatist point of view we should depict a state’s (non-) proliferation moves as the result of an ongoing process in which the state’s identity, its role and position in the environment are discursively established and in which potential courses of action are construed. Consequently, if it is correct to claim that a state’s armament policy is enabled by prevailing narratives of the state’s interests and preferences, its position in the world, by its attitude toward significant others and by its constructed identity, then we need to excavate these narratives in order to better appreciate a state’s policy decision. In other words, the key to understanding why states pursue or abandon nuclear weapons programs lies in understanding the key underlying beliefs that are evident in the national “nuclear discourses.”5 To corroborate this claim, the study focuses on a discourse analysis of two cases of nuclear “non-proliferation” or reversal – Switzerland and Libya. The study proceeds as follows. In a first step (Chapter 2) the study is situated in relation to the existing body of literature on nuclear proliferation and international security. It is argued that even supposedly non-theoretical, policyfocused studies are based on certain (often covert, implicit) theoretical assumptions that fundamentally shape the research outcome. I then describe how these theoretical assumptions define what is taken for granted in the study of nuclear (non-)proliferation. Over the course of several decades these rules of research have contributed to the consolidation of a realist orthodoxy in security studies that has left little room for alternative approaches. Mainstream presuppositions such as pre-defined interests, objectively identifiable threats or the presumably given anarchical condition of the international system, together with the positivist desire to find causal laws and objectively proven regularities within the social world, have shaped – and narrowed – our thinking about international security issues. Even many research projects that – at first sight – seem to follow a broader research agenda by paying attention to non-materialist aspects such as beliefs, ideas, identity or culture ultimately remain dedicated to a rationalist logic. As a consequence, they lack a sufficiently comprehensive understanding of the intersubjective processes that constitute social action. Moreover, the common adherence to a natural science-inspired model of cause finding sets strict boundaries for possible routes of enquiry and necessarily disregards the

broad array of social facts and intersubjective, shared meanings that enable or preclude a specific course of action and that constitute the reality in which further policymaking takes place. It is this gap that needs to be addressed if we are to gain a better and more substantial understanding of states’ nuclear policies. At the same time, a “pragmatist turn” enables us to bridge the gap that has emerged between metatheorizing on the one hand and problem-solving on the other. “Pragmatism starts with action,”6 as Kratochwil reminds us. It explicitly calls for more attention to “real-world” problems and concrete political challenges – but in a theoretically reflexive, conscious way. A review of the literature indicates that although a pragmatist-interactionist approach is innovative for security studies, it can nevertheless build upon and connect with many ideas and reflections already inherent in contemporary (IR theory) writings. Accordingly, the literature synopsis has two objectives within the overall structure of the study. On the one hand it explores shortcomings in the existing body of literature on nuclear proliferation in order to show that a pragmatist understanding of state action can fill a significant theoretical void and can supplement current approaches in the realm of foreign and security studies. On the other hand, it aims to highlight given links and junctions between some existing works and the suggested pragmatist-inspired approach. In Chapter 3 I outline the social-theoretical mainstays of a pragmatistinteractionist framework of analysis and its methodological implications. Drawing on the writings of authors such as George H. Mead, Herbert Blumer and Friedrich Kratochwil, I claim that the discursive practices of making sense of and interpreting the world act as the source and starting point for human action: they give both material and non-material objects of our environment an “identity,” i.e., a distinctiveness or meaningfulness to which we refer when we construe potential courses of action. A pragmatist understanding of the social world thus replaces the realist notion of a given, fixed reality; it presupposes instead that even core realist concepts such as “threats,” “national interests” or “security” need to be understood as socially created. Taking this presupposition seriously allows for an emphasis on the contingency and language dependency of meaning, and brings political agency – the creative, mindful behavior of social/political actors – back into our analysis. Moreover, it helps us to discern how state action – and a state’s security and nuclear policies – is shaped by socially shared political, historical and cultural imaginaries and narratives. In the second part of the chapter the methodological implications that accompany a pragmatist approach are illustrated. The two basic methodological steps on which each case study is based are outlined. The first step consists of a “historiographic reconstruction” of the events; it provides a process-tracing account of the historical undercurrents and of the political circumstances that underlie the specific procurement decision in question. This reconstruction is designed to increase familiarity with the case and to outline the broader socio-political and cultural environment in which the nuclear discourse is embedded. It also helps to discern the discursive realm, i.e., it identifies key participants and their major contributions to the debate. The concrete analysis of the documents – step two of

each case study – is based on a discourse analysis of publicly available government publications as well as non-governmental contributions to the debate. Hence, in order to grasp more fully the heterogeneous and multi-linear process of meaning-making, the analysis not only covers documents issued by state representatives or state agencies, but also includes articles, leaflets and other printed material published by societal groups and non-governmental actors. The underlying assumption is that basic policy beliefs and narratives are not merely imposed by one actor but evolve through a socially shared, intersubjective process. With regard to the practical terms of research, the analysis of each case does not start with a given, testable hypothesis but shifts repeatedly between data collection, analysis and concept formation or theory building. As a result, the procedure does not seek to verify or falsify pre-established theories. Its goal is rather the collection and continuous refinement of data that only eventually leads to the generation of new concepts, contingent generalizations and possibly theories of medium range. Chapters 4 and 5, the empirical core of the study, then put these theoretical and methodological considerations into practice. Chapter 4 contains an analysis of the “rise and fall” of Switzerland’s interest in nuclear weapons during the 1950s and 1960s. The analysis reveals that the decision-making process concerning the (non-)acquisition of nuclear weapons took place in a multi-layered and non-linear manner that cannot be reduced to a security rationale. Consequently, it is argued that it was not the emergence (and later disappearance) of an objectively given, concrete external threat or a threat perception which “caused” the Swiss interest in (and later dismissal of ) a military nuclear capability. Rather, instead of merely reacting to objectively given exogenous impulses or inputs, the decision to abandon the program followed from a fundamentally political exchange of perceptions and views and an ongoing renegotiation of key narratives and frames. More precisely, the study suggests that during the course of the debate there was a significant shift in the central underlying narratives of the country’s self-perception as a neutral state and on its role in the world, as well as the socially shared characterization of nuclear weapons. These narrative shifts ultimately precluded the acquisition of nuclear weapons. The analysis of the Libyan case which follows in Chapter 5 also calls for increased attention to be paid to underlying narratives and frames. On the grounds of both historiographical reconstruction and discourse analysis it appears misleading to reduce the emergence and abandonment of Libya’s nuclear program to security considerations. The study reveals that even fundamental political categories such as “threat,” “enemy,” “ally” and “security” were not fixed and settled concepts but exhibited oddities, inconsistencies and apertures for the renegotiation of meaning. As a consequence, the security narrative was not powerful and was thus not convincing enough to spur far-reaching, wholesale and sustained efforts to acquire nuclear weapons. An examination of socially prevalent beliefs reveals, on the other hand, that Libya’s initial interest in nuclear weapons was spurred by a deeply engrained desire for regional and global status and equality. At the same time, the case is striking for its clear lack

of explicit references to the nation state and the absence of a narrative of Libyan statehood or Libyan modernity. The idea of the Libyan nation remains vague and disembodied. Arguably, this lack of a state narrative contributed significantly to the eventual failure of Libya’s nuclear ventures: there was no compelling national frame to spur broad-based national efforts and to provide sufficient momentum in order to maintain a fully fledged research program over several decades. In other words, without a convincing and coherent narrative of “what Libya is,” all the efforts to establish a Libyan nuclear weapons program eventually suffered from insufficient political momentum. The final chapter (Conclusion) offers a summary of both the theoretical and empirical parts of this study. In addition, it outlines implications for the further theory development in IR as well as for our practical political dealings with the issue of nuclear (non-)proliferation. In theoretical terms the study has two major implications. First, a pragmatist approach to IR and international security calls for “theoretical modesty” and for the abandonment of large-scale, universal causal theories. We are urged to pay tribute to human agency, reflexivity and historical contingency, rather than parsimony, causality and universality. Consequently, instead of achieving systematic simplification and generalization, our research should attempt a better grasp of the complexities and intricacies of social phenomena. It appears necessary, therefore, to limit our efforts in explanation and theory building to the development of contingent generalizations and middle-range theories, such as are always preliminary and revisable. Second, pragmatism might hold the potential to overcome the long-standing dispute between meta-theoretically inclined IR scholars and advocates of more policyoriented research. Pragmatists advocate a “multi-perspective” research strategy that suspends unsolvable meta-theoretical contestations and instead calls for cross-paradigmatic dialogue and cooperation in IR. Yet they do so without relinquishing theoretically sound and stringent knowledge production and without promoting theoretical arbitrariness. Rather, pragmatism provides the chance to put on hold meta-theoretical quarrels and concentrate instead on eminent political challenges and “real-world” problems. At the same time, it encourages us to eclectically synthesize different analytic insights in order to develop more comprehensive, multifaceted explanatory accounts. In practical and empirical (i.e., proliferation-related) terms the study suggests that a pragmatist-inspired framework of analysis can deepen our understanding of the “nuclear trajectories” of both Switzerland and Libya. It directs our attention away from allegedly given motifs and causes and instead elucidates the role of narratives and beliefs in shaping both the origin and termination of each program. This has significant consequences for the further elaboration of global efforts to curb or even roll back nuclear proliferation. Instead of merely reducing the nuclear dynamic to a given security rationale, we need to increase our understanding of the political processes and the domestic nuclear discourses of potential “nuclear weaponizers.” If we accept that the nuclear decision-making of states follows from the intersubjectively shared beliefs and narratives regarding the country’s identity, its self-perception or its position vis-à-vis significant

others, then the international community should increase its efforts to intervene politically in these discursive processes. The case studies illustrate that even fundamental frames and concepts such as “enemy,” “ally” and “threat” are not fixed, but alterable and open for renegotiation: they are amenable to political discourse and “de-securitization” moves. In a similar vein, it is conceivable that a global “nuclear weapons convention,” while it would most likely not trigger immediate disarmament steps or nuclear reversal by states of concern, could, however, lead to a strengthened anti-nuclear weapons discourse and to an increased opposition to nuclear weapons. In the longer run, a (codified) condemnation of the possession and use of nuclear weapons might contribute to a negative framing of nuclear weapons and in turn reinforce the political obligation to act in accordance with the norm. While such instruments are not likely to trigger the immediate renunciation of the weapons ambitions of certain states of concern, they provide a fruitful side avenue for the strengthening of global nonproliferation efforts. In sum, this study aims to make three distinct contributions. First, it seeks to broaden our abstract understanding of the “causes” of state action in general and of states’ nuclear (non-)proliferation policies in particular. It does so by introducing a pragmatist-interactionist lens of analysis that helps us better understand the intersubjective discursive processes of meaning-making and interpretation which – it is argued – build the basis for ensuing political action. Second, by offering a fresh look at the cases of nuclear reversal in Switzerland and Libya, the study contributes to our concrete knowledge of nuclear proliferation and, above all, reversal. It reveals that an analysis of underlying “nuclear-related” narratives and beliefs provides crucial insights that have hitherto been neglected. Third, the study shows that a pragmatist framework of analysis offers an opportunity to overcome long-standing gridlocks in IR discipline and instead encourages IR scholars to reorient their efforts toward imminent “real-world” challenges.