Variances in democratic conflict behaviour Societal role theory as a new approach to democratic war

Transcription

1 Variances in democratic conflict behaviour Societal role theory as a new approach to democratic war Lena Clausen, Berlin Abstract The democratic peace, i.e. peace among democracies, is a thoroughly analysed and studied phenomenon in peace and conflict research. Since the 1980s, a special focus was given to the particularity of peace among democracies. Assuming a general peaceful impact of democracy, numerous studies were conducted in order to prove a specific peace-proneness of democracies. After the end of the cold war, however, the war-proneness of democracies became more and more evident and research focused on the so-called democratic wars, meaning the wars that democracies lead against non-democracies. In the conflict behaviour of democracies increasing peculiarities and irregularities emerged. They cannot be accounted for by the traditional theoretical approaches to the democratic peace. Especially during the Iraq war in 2003, these peculiarities, or so-called variances, became evident. They are based on two empirical observations: the behaviour of different democratic states varies: there are democratic states that never or very seldom go to war and others that are often involved in wars (cross-democratic variance). Furthermore, there are variances in the conflict behaviour of one democratic state over time, either in one conflict or in several following conflicts (temporal variance). This paper aims at grasping these twofold variances, using the theoretical approach of role theory. In order to achieve this aim, role theory and its concepts are re-conceptualised on a new societal-constructivist basis and applied to case studies. Both dimensions of variance can be observed in the conflict behaviour of the Scandinavian states Sweden, Denmark and Norway in the Iraq war 2003, which are chosen as examples in this paper. Keywords Democratic war, variances, social constructivism, role theory, foreign policy, Iraq War

2 1. Introduction Democratic peace has been thoroughly analysed for 30 years and yet new puzzles keep emerging in the matter. During the 1990s, the war-proneness of democracies could not be overlooked any longer. Thus, the separate peace -assumption stating that democracies are peace-prone towards each other but not less war-prone in general was widely accepted (Geis/Brock/Müller 2006). 1 Research started to focus on the dark side of the democratic peace (Geis/Brock/Müller 2006), striving to identify specific democratic reasons for war against non-democracies. 2 Regarding these democratic wars, a new puzzle emerged: In theory, all democracies were supposed to act alike due to their common regime type be it peace-prone or war-prone. 3 In practice, there is significant variance among them. Variances appear along two dimensions: (1) not all democracies participate in the same wars (Geis/Müller/Schörnig 2010; Clark/Nordstrom 2005; Müller 2003). There are democratic states that very seldom or never go to war, e.g. Sweden, Switzerland, Germany for 50 years after 1945, and others, that are often involved in wars, e.g. the US, Great Britain, France, Australia. This variance is called cross-democratic variance. (Carlsen/Mouritzen 2004) Furthermore, there are variances in the conflict behaviour of one democratic state over time, either in one conflict or in several following conflicts (Müller 2003; Brock 2006). This dimension of variance can be described as temporal variance. The observed variances put an enormous strain on the traditional thinking about democratic peace and war and must be analysed further. Traditional approaches to the democratic peace suggest that all democracies act alike, thus disregarding any differences in democratic structure, process, practice and/or time. The only promising outlook to grasp variances is offered by social constructivist approaches focusing on identity and roles (Risse-Kappen 1995; Peceny 1997). These allow for the possibility to differentiate between democracies according to their identity and roles. Most of them already draw albeit implicitly on role 1 The assumption of the general peace-proneness of democracies can however still be found with Rummel (1997), Russett/Oneal (1999) and MacMillan (2003). Yet, empirical data do not confirm this monadic assumption (Chojnacki 2006a, 2006b). 2 Wars that can be labeled democratic wars are: The Gulf War 1991, Kosovo 1999, Afghanistan 2001, Iraq 2003 and Libya Characteristics of democratic wars are: the specific moral and normative legitimations, that lead to humanitarian or military interventions ; the explicit goal of an implementation of UN-resolutions often in the absence of a UN-mandate (exceptions are Afghanistan and Libya); the exercise of military force in a coalition or alliance - often NATO (Geis/Brock/Müller 2006). 3 The traditional explanatory approaches of democratic peace are institutional (Russett 1993; Bueno De Mesquita et al. 1999), rational (Doyle 1983a, 1983b) and normative approaches (Risse-Kappen 1996) - (cf. Maoz/Russett 1993). 2

3 theory (Müller 2004; Brock/Geis/Müller 2006). 4 A central assumption of these approaches is that roles of democracies are deeply influenced by their being democracies: a democratic selfawareness (Müller 2004). This paper addresses one theoretical question: can role theory grasp variances in democratic conflict behaviour? Furthermore, it addresses one empirical question: can role theory grasp the variances in the conflict behaviour of the Scandinavian States during the Iraq War 2003? Unlike existing approaches would suggest, the Nordic states didn t exhibit uniform behaviour during the Iraq War On the contrary: while Sweden stayed out of the war altogether, Denmark participated as a member of the coalition of the willing. Norway, moreover, displayed periodic variance in its behaviour: showing hesitance at first, there was no original participation in the war. In July 2003, however, 150 Norwegian soldiers were sent to Iraq. They were later withdrawn (from July 2004). This made Norway one of the first countries to leave Iraq behind. The Scandinavian States were chosen as examples because of the evident variance in their behaviour. Furthermore, they exhibit broad similarities that help control for a series of alternative explanations: location and geopolitical situation, size (small/medium powers), regime type and political system (parliamentary monarchy), language and culture, alliances (EU: Sweden and Denmark; NATO: Denmark and Norway). 2. Theoretical frame This article utilizes a theoretical framework building on role theory and societal constructivism. 5 It aims to exemplify role theory s contribution to the understanding of democratic peace and war. When we look at the variances displayed by the Nordic states, we get interesting results using a slightly modified approach to role theory. As in traditional role theory, in this new approach, too, roles are conceived as consisting of an ego part (identity) and an alter part (expectations) (Holsti 1970; Gaupp 1983). 6 Traditional role theory, however, is a systemic IR-approach focusing on interactions among states. 7 A shortcoming with this standard systemic approach is its inability to open up the black box state and capture 4 All of the theoretically approaches to democratic peace and war are, however, rooted in the liberal agenda of the democratic peace (Müller 2003). 5 Constructivism is not a classical theory of IR but more an analytical frame looking closely at norms, social interactions and identities. Different types of constructivism focus on different levels of analysis, depending on which interactions are considered as central for the co-constitution of identities: inter-state, societal or individual interactions. Role theory, too, is not a complete theory, but a theoretical framework to analyse international relations and foreign policy. 6 Furthermore, roles consist of multiple role elements (Clausen 2012). 7 Role theory can - ex post - be categorized as a constructivist approach, it must then be attributed to Alexander Wendt s state-centric constructivism (Wendt 1992, 1994, 2004). 3

4 aspects on the inside, the societal level. In classical role theory, states as a whole act out roles. 8 Several problems linked to this perspective arise in traditional role analysis: The state which is treated as an individual itself cannot speak and act for itself (cf. Wendt 2004; Gaupp 1983).In traditional role theory a short cut is chosen to avoid this problem: it is the state s representatives who state and act out roles for the state itself (cf. Holsti 1970; Gaupp 1983; Chafetz/Abramson/Grillot 1996). This is a mixture of considerations at the systemic and individual level of analysis that is unaccounted for. Another problem arises because most earlier role analyses focus exclusively on decision makers statements (Holsti 1970; Gaupp 1983; Wish 1987; Shih 1988; Chafetz/Abramson/Grillot 1996; Kirste 1998). As such they only take the ego part of the role into account. The alter part is reduced as well: If it is at all taken into consideration, it is the external alter part, i.e. expectations from others states and their representatives (cf.holsti 1970, 1987; Harnisch/Maull 2001). To sum up, the systemic perspective does not allow capturing the role s collective character in all respects: the ego part is reduced to decision makers role conceptions, the alter part is reduced to external expectations, it is unclear in what way individual representatives speak for the state and act out its role while holding their individual role conceptions that can differ from those of the state. Moreover, since the systemic perspective does not allow a look into the state, traditional role theory neglects how roles developed historically within the state and how broadly society or societal groups share them. Therefore, it is hardly apt to analyse specific democratic roles (Clausen 2012). To specify further, different role conceptions between state s representatives (government) and the democratic public cannot be captured (Breuning 2011). Fortunately, there is a more apt alternative: role theory can be put on a new social constructivist basis that allows for a more profound role analysis. This social constructivist perspective entails several advantages: the black box of the state is opened up and the societal development and negotiations are taken into account (Le Prestre 1997). This perspective entails several advantages in order to grasp variances in democratic conflict behaviour: the society as a collective is the role carrier - the state is not seen as a person. The ego part, hence, reflects the societal self-image (Le Prestre 1997) not the role conception of its decision 8 Wendt sees states as persons (Wendt 2004) - this reflects in traditional role theory, where the state is conceived as a sole role carrier. The international system then is a community where states hold roles and interact with each other as individuals (cf. Gaupp 1983; cf. Singer/Hudson 1987; cf. Boekle/Rittberger/Wagner 1999). 4

5 makers. 9 By taking internal aspects of the state into account, the impact of the regime type on its identity or role can be measured through public role statements. To sum up, there is no mixture of levels of analysis. 10 Last but not least, the alter part is expanded by societal expectations towards political representatives, i.e. expectations on the social level. In sum, while a state centric constructivist view makes sense to analyse states roles in general, it cannot account for the specific role of a democracy neither theoretically nor analytically. If role theory, however, is put on a new basis, using a societal constructivist point of view, it should be able to distinguish roles of democratic states from those of other states. Thus, it holds the key to grasp democratic variances. While the traditional role theory approach already encompasses historic role development, role conflicts and role change, these concepts have been refined for the present analysis. These dynamic aspects of role theory and their possible impact on variances will be elaborated below. The role In traditional role approaches, the role already consists of an ego- and an alter part. Most analyses, however, focused exclusively on the ego part (identity) of a state s role. Nevertheless, the alter part has to be included in the analysis in order to attain a full role analysis. Due to the proposed societal perspective the alter part is no longer simply made up of external expectations of other states, but also of expectations of the society towards the government. This is especially important for a democratic perspective where the people is the sovereign. Thus their role conceptions should be crucial. Important questions for societal role theory are: Does society share the representatives role conceptions and vice versa? Whose role conceptions prevail? (cf. Breuning 2011). Historical development of roles The historical development of roles has been widely neglected in existing role analyses (Holsti 1970; critical already Wish 1987). It is, however, essential to have knowledge of the 9 It is, however, not suggested that society as a whole holds one common role conception. Different groups in society rather hold different role conceptions (Shih 1988: 602), i.e. there are different role elements in one society. The groups negotiate publicly the different elements and the options for appropriate action they entail (cf. Gaupp 1983: 102). 10 While society itself is as speechless as the state, the public negotiation of societal role elements can be tracked through role statements of different speakers with different role-conceptions within society. 5

6 formative incidents that lead to the emergence and development of role elements. Furthermore, it is important to know prior role performance and the experience a state has had acting out certain role elements. A historical role analysis should be conducted as a pre-step to actual role analysis. This can be done using secondary literature. In any case, it will help not only to identify existing role elements but also to identify underlying role conflicts and past role change(s) that may influence role performance in a given situation/conflict. Role conflicts Role conflicts are inherent in multidimensional roles (Gaupp 1983). 11 To simplify and merge existing categorisations of role conflict, I propose to distinguish between endogenous and exogenous role conflicts. Endogenous role conflicts emerge on the inside of the state, i.e. between society and government. Exogenous role conflicts emerge on the outside of the state, i.e. between the external expectations and the states government and/or the society. Furthermore, role conflicts can be sensibly categorised according to their causes: they can occur as identity related conflicts (identity conflict) and as conflicts resulting from internal or external expectations (expectation conflict). Role change Role change can be triggered by role conflicts. Role change, then, is a means of acute conflict resolution and its course is rather rapid. Slower role change can emerge when there are underlying tensions concerning the states role (Clausen 2012). 3. Analytical framework This paper introduces an analytical framework, based on social constructivism and role theory, for the purpose of evaluating the impact of collective, social roles on conflict behaviour. It illustrates how the variances among the Scandinavian States can be understood on the basis of their historic roles, previous role change and occurring role conflicts. Analytically, a collective role can be captured through speech acts, i.e. role statements that reveal role elements and are linked to the ego- or alter part. Former role analyses were conducted analysing statements of high ranking diplomats and state representatives (Holsti 1970; Gaupp 1983; Wish 1987; Shih 1988; Chafetz/Abramson/Grillot 1996; Kirste 1998). 11 A multidimensional role consists of ego- and alter part as well as different role elements it is not a unitary phenomenon (Clausen 2012). 6

7 This approach does not meet the requirements of a societal role analysis. In the societal version of role theory, the impact of public opinion has to be taken into account in order to identify and analyse role elements at the societal level. To specify further, this analytic approach has been modified to include public opinion in the analysis in order to capture the democratic public s role elements. Public opinion statements betraying role conceptions can easily be identified in newspaper articles. The role analysis undertaken consists of two steps. In a first step the historical analysis of the three cases Sweden, Denmark and Norway was conducted using secondary literature. At this stage the emergence and development of role elements were retraced, thus revealing possible underlying role conflicts and role change. In a second step the current role analysis was undertaken. At this stage the newspaper articles of the research period (January-March 2003) were analysed using qualitative content analysis, thus identifying role statements and linking them to respective role elements while conserving openness for unexpected role elements that were not historically identified. The newspaper articles were assembled on online databases: factiva.com, artikelsök, presstext und mediearkivet. All material was analysed in original language. 4. Findings: Nordic variances in the Iraq War Sweden Historical analysis identified several central role elements for Sweden: neutrality/nonalignment, the relevance of international law and the United Nations (UNO) as well as international activism, i.e. holding a view on international affairs and pronouncing it openly. The Swedish conflict behaviour in the months leading up to the Iraq War 2003 was mainly influenced by the role element neutrality which was interpreted by society as Swedish peace tradition. This notion of neutrality was unexpected given the historical meaning of neutrality (which does not include pacifism) and the change in the neutrality notion after 1989 narrowing it down to military non-alignment with its core of staying out of NATO. 12 This shows that the changes after 1989 while conducted in elite discourses have not shaped societal identity and role lastingly. In society the notion of neutrality of the Cold War is still deeply rooted. Role conflicts in Sweden evolved around the societal role element of neutrality/peace tradition which eventually forced the Persson-government, that was initially clinging to strict 12 Despite close cooperation with Nato in the Partnership for Peace programme. 7

8 UN-loyalty 13, to refuse a participation in a possible military action even if it was mandated by the UN Security Council. This was, however, only an option because the UN eventually did not explicitly mandate the use of military force. Thus, the refusal to use military means could be combined with loyalty to the UNO. In the Swedish case, no traces of significant external expectations were found in role statements. This is probably due to the fact that no role partner actually expected Sweden to participate in the Iraq War because of its traditional neutrality. 4.2 Denmark Denmark participated in the Iraq War as a member of the coalition of the willing and sent 150 soldiers to Iraq. The withdrawal of the personnel took place only in Denmark s willingness to participate in the Iraq War 2003 contrasted sharply from the Swedish and Norwegian behaviour. The decisive role element was that of atlanticism interpreted as super atlanticism by the Danish prime minister at the time Anders Fogh Rasmussen (Mouritzen 2006, 2007). In Denmark the atlanticist orientation only developed as a role element after the end of the Cold War and the crumbling of the security system Denmark was part of ( Nordic Balance ). Subsequently, Denmark experienced fundamental role change so that the emergence of atlanticism as a role element went along with the emergence of a new more active foreign policy doctrine and the degradation of traditional antimilitarism. Atlanticism itself was hence a rather new role element that underwent further change after 9/11 towards Super-Atlanticism. This unconditional willingness to align Denmark with the US was represented by Danish Prime Minister Anders Fogh Rasmussen (now Secretary General of NATO). The Danish public noticed the obvious role conflict that occurred between traditional loyalty to the UN and the more recent loyalty to the US, whereas the Danish government point-blank refused to acknowledge this. Instead it insisted on the compatibility of both role elements, i.e. being loyal to the UN as well as to the US at the same time. Only role performance did eventually betray that US-loyalty was acted out over UNO-loyalty. In a last attempt to convince the public the Danish government and the Prime Minister in particular tried to turn the decision to go to war into a battle of good vs. bad or between democracy and dictatorship. 13 Strict UN-loyalty means that any decision of the UN Security Council would be supported, it does not include a decision with regards to content (peace/war) but is a formalistic attitude. 8

9 The moral justifications for the use of force, however, did not resound well with the Danish society: there was no identitary reaction to this argumentation. 4.3 Norway Norway s conflict behaviour was influenced by several central role elements that were weighed against each other during the course of the conflict: Norwegian independence from others, the loyalty to the UN, bridge building, traditional atlanticism and the importance of NATO in security matters. While independence, bridge building and atlanticism were rather underlying tendencies, one of the elements outweighed the others: UN-loyalty. It was declared the absolute principle of Norwegian foreign policy: no war without UN-mandate. 14 The central role conflict erupted between the importance of the UN on the one hand and the close friendship with the US on the other hand. Apart from UN-loyalty another decisive role element is that of loyalty to NATO. Similar to Denmark, Norway relied on NATO as its sole security guarantee during the Cold War. After 1989, NATO was still seen as essential for Norwegian territorial security. Unlike Denmark, however, Norway has developed a selfimage of an emancipated partner. 15 Therefore, it was able to maintain a stronger standpoint vis-à-vis the United States. This was essential because Norway came under significant role pressure from US-officials. This pressure led to a promise made by the government of Prime Minister Bondevik to help rebuild Iraq. This promise, in turn, paved the way for the temporal variance in the Norwegian conflict behaviour during the course of the Iraq War. Nevertheless, the historical role element of UNO-loyalty stayed crucial during the course of the temporal variance. Norway clung to this role element and tried to accommodate it with the external expectations. When the UN Security Council adopted Resolution 1483 on May 22 th 2003, Norwegian troops could be deployed in Iraq without breaking the principle of UN-loyalty while at the same time accommodating the US. Thus, in July Norwegian soldiers were sent to the region around Basra, first under British command, later under Polish command. Temporal variance continued in July 2004 when the large majority of Norwegian troops was withdrawn from Iraq (140 soldiers). This was initiated by the Bondevik administration. The only explanation that was given was that Norway as a small country could not be 14 This was also linked to the relatively new role element of Norway as a peace nation which was not often mentioned explicitely in role statements but was closely linked to the role of the UN. 15 Historically, independence is of utmost importance to Norway, having spent hundreds of centuries under foreign rule. 9

10 everywhere (Foreign Minister Jan Petersen). When the government changed in September 2005 to the red-green coalition of Jens Stoltenberg only 20 Norwegian officers remained in Iraq. The ultimate withdrawal that was announced by the new Prime Minister was thus primarily of symbolic nature. Yet, in doing so, Stoltenberg and his foreign minister, continued to emphasise the continuity in Norwegian conflict behaviour by evoking traditional role elements such as UN-loyalty, Norway as a peace nation, the close relationship to its allies. i.e. US-loyalty. The temporal variance can therefore be traced back to Norway s holding on to central role elements (UNO- and US-loyalty) while at the same time emphasising Norwegian independence. The variance was facilitated by the role conflict resulting from external expectations. Short or long-term role change, on the other hand, seems to have played a lesser role. 5. Conclusion This analysis pursued the question whether variances in the conflict behaviour of democracies can be theoretically captured using a societal role approach. The answer is yes: overall, societal roles can account for variances in states and more specifically democracies conflict behaviour. The analysis departed from the puzzle of democratic variance which is part of the research agenda of the democratic war. It could not be accounted for by existing approaches to democratic conflict behaviour. Therefore, I proposed the use of a societal role approach that could grasp the collective role of a democratic state/society and thus differentiate between different democracies. This new approach drew on existing social constructivist approaches to democratic peace/war, refining them and putting them on a new basis. Using a theoretical frame based on role theory and social constructivism has proven a fruitful approach to democratic variances. Societal role theory offers a few central advantages to grasp democratic variances compared to the traditional systemic perspective. Its focus on societal interactions within the state makes the role a truly collective phenomenon unlike the unitary concept that existed on the interstate level. Instead the black box role was opened up to include the development of role elements and their negotiation on the societal level. These elements and the appropriate actions they indicated could satisfactorily be traced in public role statements. Completing this new theoretical perspective to democratic conflict behaviour is the analytical framework this paper has introduced. It was designed to not reduce roles to role statements of 10

11 high ranking representatives, but to include the role elements of the democratic public an essential factor when it comes to analysing democratic conflict behaviour. The framework treats roles as multidimensional social constructions that can be traced through historic analysis identifying existing role elements and their enactment in former role performance. Furthermore, role elements can be identified via public role statements by society as well as state s representatives. The analysis undertaken suggests that the Scandinavian states acted according to their respective roles during the Iraq War 2003, experiencing role conflict and role change. To sum up, the Scandinavian States while similar enough in other regards hold idiosyncratic role elements and roles which lead to varying conflict behaviour. Temporal variance in the Norwegian case could be traced back to underlying role conflicts. Unlike existing approaches to democratic peace and war would suggest, however, regime type seems to matter little when it comes to social roles. To specify further, neither a role element democracy nor a democratic self-image as anticipated in earlier role approaches to democratic peace and war could be traced in role statements. This casts important doubts on the previous thinking about specific democratic conflict behaviour. Thus the question, how democratic roles are influenced by their democraticness is left unanswered for the moment. More efforts have to be made to find out how regime type influences identities and roles of states. Another aspect casts doubts on the link between regime type and social roles: Endogenous role conflicts between the democratic public and government - could be observed in all three of the analysed cases. These differences are especially crucial for conflict research focusing on the regime type. Traditional approaches to the democratic peace and war assume that society s attitude towards peace and war translates into foreign policy. In societal role theory this translates: society s role elements translate into foreign policy. While this proved right in two of the cases (Sweden and Norway), the third challenges this assumption. The Danish government and more specifically the Prime Minister, however, acted according to his own role conceptions which he prioritised over societal role elements and which resulted in war participation. It is therefore worth mentioning that individual decision makers role conceptions have an impact on role behaviour. It is these role conceptions of individual decision makers that are of particular interest to researchers analysing democratic conflict behaviour. If their role conceptions are indeed decisive as suggested by earlier approaches 11

12 to role theory (e.g. Holsti 1970) what would distinguish role performance of democracies from that of other states? 6. Bibliography Boekle, Henning/Rittberger, Volker/Wagner, Wolfgang (1999): Normen und Außenpolitik: Konstruktivistische Außenpolitiktheorie, Tübinger Arbeitspapiere zur Internationalen Politik und Friedensforschung Nr. 34, Tübingen. Breuning, Marijke (2011): Role theory research in international relations. State of the art and blind spots, in: Harnisch, Sebastian/Frank, Cornelia/Maull, Hanns W. (Ed.) (2011): Role Theory in International Relations. Approaches and analyses, London and New York, Brock, Lothar (2006): Triangulating War: the Use of Force by Democracies as a Variant of Democratic Peace, in: Geis, Anna/Brock, Lothar/Müller, Harald (Ed.) (2006): Democratic Wars. Looking at the Dark Side of Democratic Peace, Houndmills, Brock, Lothar/Geis, Anna/Müller, Harald (2006): The Case for a New Research Agenda: Explaining Democratic Wars, in: Geis, Anna/Brock, Lothar/Müller, Harald (Ed.) (2006): Democratic Wars. Looking at the Dark Side of Democratic Peace, Houndmills, Bueno De Mesquita, Bruce/Morrow, James D./Siverson, Randolph M./Smith, Alastair (1999): An Institutional Explanation of the Democratic Peace, in: American Political Science Review, Vol. 93, No. 4: Carlsen, Per/Mouritzen, Hans (2004): Danish Foreign Policy Yearbook 2004, Diis Book, Danish Institute for International Studies,, Kopenhagen, (2004): Danish Foreign Policy Yearbook 2004, DIIS Book, Studies, Danish Institute for International, Copenhagen. Chafetz, Glenn/Abramson, Hillel/Grillot, Suzette (1996): Role Theory and Foreign Policy: Belarussian and Ukrainian Compliance with the Nuclear Nonproliferation Regime, in: Political Psychology, Vol. 17, No. 4: Chojnacki, Sven (2006a): Democratic Wars and Military Interventions, : the Monadic Level Reconsidered, in: Geis, Anna/Brock, Lothar/Müller, Harald (Ed.) (2006a): Democratic Wars. Looking at the Dark Side of Democratic Peace, Houndmills, Chojnacki, Sven (2006b): Anything New or More of the Same? Wars and Military Interventions in the International System, , in: Global Society, Vol. 20, No. 1: Clark, David H./Nordstrom, Timothy (2005): Democratic Variants and Democratic Variance: How Domestic Constraints Shape Interstate Conflict, in: The Journal of Politics, Vol. 67, No. 1: Clausen, Lena (2012): Twofold variance in democratic conflict behaviour - the relevance of societal role theory by the example of the Nordic states. Doctoral thesis (forthcoming, in German language), Berlin. Doyle, Michael (1983a): Kant, liberal legacies, and foreign affairs, Part 1, in: Philosophy and public affairs, Vol. No. Summer: Doyle, Michael (1983b): Kant, liberal legacies, and foreign affairs, Part 2, in: Philosophy and public affairs, Vol. No. fall:

13 Gaupp, Peter (1983): Staaten als Rollenträger. Die Rollentheorie als Analyse-Instrument von Aussenpolitik und internationalen Beziehungen, Zürcher Beiträge zur Politischen Wissenschaft, Klöti, Herausgegeben von Prof. Dr. Daniel Frei und Prof. Dr. Ulrich, Band 6, Liebefeld/Bern. Geis, Anna/Brock, Lothar/Müller, Harald (Ed.) (2006): Democratic Wars. Looking at the Dark Side of Democratic Peace, Houndmills. Geis, Anna/Müller, Harald/Schörnig, Niklas (2010): Liberale Demokratien im Krieg. Warum manche kämpfen und andere nicht- Ergebnisse einer vergleichenden Inhaltsanalyse von Parlamentsdebatten, in: Zeitschrift für Internationale Beziehungen, Vol. 17, No. 2: Harnisch, Sebastian/Maull, Hanns W. (2001): Introduction, in: Harnisch, Sebastian/Maull, Hanns W. (Ed.) (2001): Germany as a Civilian Power? The foreign policy of the Berlin Republic, Manchester, 1-9. Holsti, Kalevi J. (1987): National Role Conceptions in the Study of Foreign Policy, in: Walker, Stephen G. (Ed.) (1987): Role Theory and Foreign Policy Analysis, Durham, Holsti, Kalevi J. (1970): National Role Conceptions in the Study of Foreign Policy, in: International Studies Quarterly, Vol. 14, No. 3: Kirste, Knut (1998): Rollentheorie und Außenpolitikanalyse. Die USA und Deutschland als Zivilmächte, Frankfurt am Main. Le Prestre, Philippe G. (1997): Role Quests in the Post-Cold War Era. Foreign Policies in Transition, Montreal & Kingston. Macmillan, John (2003): Liberalism and the democratic peace, in: Review of International Studies, Vol. 30, No Maoz, Zeev/Russett, Bruce (1993): Normative and Structural Causes of Democratic Peace, , in: The American Political Science Review, Vol. 87, No. 3: Mouritzen, Hans (2007): Denmark's Super Atlanticism, in: Journal of Transatlantic Studies, Vol. 5, No. 2: Mouritzen, Hans (2006): Choosing Sides in the European Iraq Conflict: A Test of New Geopolitical Theory, in: European Security, Vol. 15, No. 2: Müller, Harald (2004): Demokratien im Krieg - Antinomien des demokratischen Friedens, in: Schweitzer, Christine/Aust, Björn/Schlotter, Peter (Ed.) (2004): Demokratien im Krieg, Baden-Baden, Müller, Harald (2003): The Antinomy of Democratic Peace, Paper prepared for the ECPR Meeting, Marburg, September 2003, Oneal, John R./Russett, Bruce M. (1999): The Kantian Peace: The Pacific Benefits of Democracy, Interdependence, and International Organizations, , in: World Politics, Vol. 52, No. 1: Peceny, Mark (1997): A Constructivist Interpretation of the Liberal Peace: The Ambiguous Case of the Spanish-American War, in: Journal of Peace Research, Vol. 34, No. 4: Risse-Kappen, Thomas (1996): Collective identity in a democratic community: The case of NATO, in: Katzenstein, Peter J. (Ed.) (1996): Culture of National Security, Risse-Kappen, Thomas (1995): Democratic Peace - Warlike Democracies? A Social Constructivist Interpretation of the Liberal Argument, in: European Journal of International Relations, Vol. 1, No. 4: Rummel, Rudolph J. (1997): Power Kills: Democracy as a Method of Nonviolence, New Brunswick, N.J. 13

14 Russett, Bruce (1993): Grasping the Democratic Peace. Principles for a Post Cold War World, Princeton. Shih, Chih-Yu (1988): National Role Conception as Foreign Policy Motivation: The Psychocultural Bases of Chinese Diplomacy, in: Political Psychology, Vol. 9, No. 4: Singer, Eric G./Hudson, Valerie M. (1987): Role Sets in African Foreign Policy Behaviour: Testing an External Predisposition Model, in: Walker, Stephen G. (Ed.) (1987): Role Theory and Foreign Policy Analysis, Durham, Wendt, Alexander (2004): The state as person in international theory, in: Review of International Studies, Vol. 30, No Wendt, Alexander (1994): Collective Identity Formation and the International State, in: American Political Science Review, Vol. 88, No. 2: Wendt, Alexander (1992): Anarchy is what states make of it: the social construction of power politics, in: International Organization, Vol. 46, No. 2: Wish, Naomi Bailin (1987): National Attributes as Sources of National Role Conceptions: A Capability-Motivation Model, in: Walker, Stephen G. (Ed.) (1987): Role Theory and Foreign Policy Analysis, Durham,