While the recent IR “narrative turn” has greatly improved our understanding of how narratives influence state policy choices, we need to deepen our understanding of how narratives explain policy change. If state “autobiographies” provide such powerful explanations of why states do what they do, how can they change their policies and practices? To understand the relationship between policy change and state narrative continuity, I build on existing scholarship on narrative analysis and ontological security to examine ways in which state autobiographical narratives are used by political actors to confront state insecurities. My principal argument is that at times of great crises and threats to multiple state securities (physical, social, and ontological), narratives are selectively activated to provide a cognitive bridge between policy change that resolves the physical security challenge, while also preserving state ontological security through offering autobiographical continuity, a sense of routine, familiarity, and calm. I illustrate the argument with an analysis of Serbia's changing foreign policy behavior regarding the disputed status of Kosovo.
In March 2011, more than three years after Kosovo had declared independence from Serbia, Serbian foreign minister Vuk Jeremić tried to explain to the audience at Wheaton College, Chicago, what Kosovo continued to mean for Serbia:
Well, for us, Serbs, Kosovo is like the very air we breathe. It's the beating heart of our culture—and home to our most sacred shrines. Kosovo is the land where hundreds of thousands of Serbs gave their lives for their country and the cause of freedom. […] [Kosovo] is in our dreams at night, and in our prayers in church. It is the apple of our eye. It is our Jerusalem. (cited in Kuzmanovic Jovanovic 2011 :36)
What Serbian foreign minister was offering was a narrative—a story with meaning, characters, and a plotline. It is also a story with a very specific political purpose—to delegitimize the separation of the “heart” (Kosovo) from the “body” of the nation (Serbia), and to appeal to action that the heart be returned to its rightful and suffering owner. And yet, the heart was never returned, and Serbia somehow continued to exist. In April 2013, Serbia signed an agreement with Kosovo, accepting the authority of Kosovo's government over the entire disputed territory, in exchange for local autonomy for the Serb minority and without the official recognition of the new state.
This turnaround in foreign policy was striking, and it came about rather swiftly after de facto ultimatum by the European Union, which was holding Serbia's EU accession hopes hostage to the Kosovo deal. That Serbia in the end, under tremendous international pressure, had to relinquish its control over Kosovo is not surprising and is not the question that motivates this research. Instead, I am interested in how a state preserves its identity after its foundational narrative—in Serbia's case, the notion that Kosovo is what constitutes its identity—is fundamentally challenged. If Kosovo—the core of Serbian state identity—is gone, then whither Serbia itself? If a policy change undermines the foundational state narrative, then whither the narrative?
To understand the relationship between policy change and state narrative continuity, I bring together multiple strains of international relations (IR) and sociological theory to make a theoretical connection between identity politics and narrative. The principal question I want to answer is, if state “autobiographies”—stories states tell to and about themselves—provide such powerful explanations of why states do what they do and how can they change their policies and their practices? How can they escape the narrative straitjacket they themselves have created? My article addresses this very question by developing an analytical framework that helps us move away from too static views of state narratives. I build on existing theoretical interest in narrative analysis and on rapidly expanding scholarship on ontological security ( Kinnvall 2006 ; Mitzen 2006 ; Steele 2008 ; Zarakol 2010 ), to examine ways in which state autobiographical narratives and the domestic contestation over them are used by political actors at time of great external stress to confront state insecurities.
My principal argument is that at times of great crises and threats to multiple state securities (physical, social, as well as ontological), narratives are selectively activated to provide a cognitive bridge between policy change that resolves the physical security challenge (for example secession of territory), while also preserving state ontological security through providing autobiographical continuity, a sense of routine, familiarity, and calm.
While state narratives need to be coherent in order to become socially powerful, even the most dominant narratives contain inherent contradictions that different political actors exploit. Further, understanding narratives as schematic templates that contain different elements and layers allows us to follow how political actors strategically activate some elements of the narrative, while deactivating others. While the policy change proposed has to fit within the overall narrative schematic template to make sense to the public, it can be crafted in a way that emphasizes some parts of the story and conveniently forgets others. It is here that the strategic rhetorical action of political actors and narrative entrepreneurs comes into play.
While I understand language to be constitutive of political action ( Milliken 1999 ; Fierke 2002 ), I place my argument within the now well-established approach of “strategic social construction” ( Finnemore and Sikkink 1998 ), which demonstrates ways in which political actors strategically manipulate shared cognitive (narrative) frames for their own political ends ( Payne 2001 ). The fact that narratives are manipulated for political purposes does not make them any less important. In fact, it makes them critical to our understanding of what motivates political action in the first place. The approach that understands social construction as being strategic concedes that political actors make rational political calculations, but they do so within a dense normative social environment that constitutes their preferences and choices. In order to explain political action more completely, I apply this understanding of the dynamic of social action and social environment to demonstrate how political actors may pursue consequentialist behavior (policy change), but it is conditional on broad acceptance of shared narrative frames. Further, the political action these agents take then further constructs and maintains the social environment in which they are embedded, creating new narrative frames in the process.
The article proceeds as follows. I first offer a brief snapshot of the existing narrative research in sociology and political science, and its recent applications in IR, especially in research on state ontological security. I develop an argument that conceptualizes narratives as bridging the cognitive gap between policy change necessary for addressing state physical security and the need to maintain state biographical continuity at times of great external stress. I then illustrate the theoretical framework with an analysis of Serbia's changing foreign policy behavior in regard to the disputed status of Kosovo.
The Power of Narratives
Narratives are critical for political life. It is through narrative that we make sense of the world and create our own identities ( Somers 1994 :606). But what do narratives actually do , and how do they influence political behavior? In their systematic presentation of the importance of narrative to political science, Monroe and Patterson have shown that narratives influence how we observe political reality, and make us act differently in response. Narratives, therefore, “play a critical role in the construction of political behavior… We create and use narratives to interpret and understand the political realities around us” ( Patterson and Monroe 1998 :321). Narratives, of course, are highly selective and purposefully constructed. Any narrative will omit some parts of the story while emphasizing the other, a process of telling that grants ideological and emotional value to what we hear and how we choose to act on that knowledge ( Franzosi 1998 ).
Narratives are important because they give us a sense of space and a sense of place. All individuals build their own autobiographies by classifying particular events as successes or failures, critical junctures and “what ifs,” strokes of good luck and misfortune ( Hankiss 1981 ). And just like individuals need a sense of autobiography, so too groups need a narrative, a compelling story of where did “we” come from, how did we come to be who we are, what brings us together in a group, what purpose and aspirations does our group have. Having a secure autobiography, a firm grasp on our past and our history provides a sense of stability and allows us to move forward. These fundamental stories are “ontological narratives” ( Patterson and Monroe 1998 :325), and they help us function as social actors. These foundational narratives structure the image of the self and create a worldview where collective interests come from ( Hankiss 1981 ).
Narratives can also be mobilizational, created to establish and promote specific collective values, and encourage a sense of groupness and solidarity. Political actors always manipulate stories to convince their followers of a specific policy, in the process making political resources out of narratives ( Hart 1992 ). They seize on collectively remembered history to make specific political points of the present ( Roediger and Wertsch 2008 ).
Narratives, however, do not exist in isolation; they are not just tools of interpretation. They are embedded in everyday discourse, dialogue, and rhetoric, and they exist in a mutually contested relationship ( Wertsch 2000 :516). Most significantly for my argument, narratives establish the foundation not only for what once was, but for what ought to be ( Wertsch 2000 :518)—they are fundamentally normative in nature. They carry a desire for a particular social order and a particular set of social practices and policies. This is why we cannot understand state behavior if we do not understand what is the normative narrative underpinning of the policy choices actors make.
Narratives in IR
Scholars in other disciplines, notably history and psychology, have explored the implications of collective narratives on social groups in much detail (in history, see the path-breaking work on historical memory in Chile by Steve Stern in Stern 2004 ; in psychology, see the work of James Wertsch on collective remembering in Wertsch 2002 ). A movement in historical sociology has used narrative to challenge the meanings of causality and explanation in sociohistorical inquiry ( Gotham and Staples 1996 ). Questioning linear understandings of causal inference, narrative analysis was used to explain social phenomena by showing how a particular social act happens “because of the order and position it has in a story” ( Griffin 1993 :1099).
Within IR and foreign policy research, much of the work done on memory and narrative has explored how “analogous reasoning” from past events affects evaluations of current foreign policy issues ( Neustadt and May 1986 ; Khong 1992 ; Record 2002 ). Levy demonstrated ways in which leaders learn from the past and make foreign policy decisions accordingly ( Levy 1994 ). More recently, Saunders has argued that policymakers develop “internal” or “external” orientations toward threats, or they perceive threats as arising from regime types or other states’ foreign policy choices, based on the memory of their past policy experiences ( Saunders 2011 ). Wittes demonstrated how collective memories of past traumas impact ongoing negotiating styles between, for example, Israelis and Palestinians ( Wittes 2005 ).
We also know that policymakers construct security narratives to make sense of the strategic environment and perceived national security threats, but these narratives also shape how these actors think about policy challenges, enabling as well as constraining possibilities of policy shifts ( Homolar 2011 ). States over time create “national security cultures,” which are in part constructed by national mythologies of past events and relationships with historical friends or foes ( Katzenstein 1996 ; also Berger 1998 ; Soeya, Tadokoro, and Welch 2011 ; Welch 2002 ). Narratives, for example, of a “Western civilization,” served as boundaries to demarcate desirable security communities and exclude others ( Jackson 2003 ).
Explanatory power of narratives, therefore, is best understood as providing cultural cognitive boundaries which sanction or constrain activities of political actors ( Hart 1992 ). It is not in providing linear causality of political action, but instead of making action possible, allowing for some practices and policies, while foreclosing possibility for others. This further reproduces and entrenches dominant policies while marginalizing alternative ones ( Autesserre 2012 :6). Narratives are social structures that create expectations, interests, and, ultimately, behaviors ( Mattern 2001 ). They create opportunities for action, as well as taboos that make certain action unimaginable.
Most of the current narrative scholarship in IR, however, assumes a certain stasis to narratives. We can explain policy choices, but we have a harder time explaining policy change. If narratives are foundational cognitive frameworks that give meaning to political action, then how can the policies narratives had informed ever be rejected, amended, or replaced? Here is where the growing research agenda in ontological security studies can offer much help.
Narratives and Ontological Security
The rapidly developing scholarship on ontological security is too large to fully summarize here, but the main premise of this literature (which builds much on Giddens 1984 ) is that states care as much about their ontological security, the security of a consistent self, as about material, physical security, the traditional purview of IR inquiry. In fact, states may be willing to compromise some aspects of their physical security in order to maintain their identity, their view of self ( Wendt 1994 ). States need predictability and order; they thrive for routine and secure relationships with others ( Huysmans 1998 ; McSweeney 1999 ). It is through these routinized relationships with their significant others that states construct their identities ( Mitzen 2006 ).
The work on ontological security has already made much use of the concept of narrative. States construct their biographical continuity through internal efforts to maintain their self-reflexive narratives, their positive views of self, at times of crisis ( Steele 2008 ). When in the aftermath of the Asian tsunami in 2004 the United States was internationally shamed for being “stingy” in offering foreign aid, the ontological security rupture this accusation created in the American sense of self-worth led to rapid change in policy ( Steele 2007 ). Narratives are important for state ontological security seeking because they provide autobiographical justification and continuity with the “good past.” States need to feel good about themselves and about their past action in order to continue to function in international society. This is why, for example, states such as Turkey and Japan continue to deny their historical wrongs as a way of preserving both their biographical continuity and their place in the contemporary international system ( Zarakol 2010 ).
States, therefore, construct “autobiographical identity narratives” to make sense of their own behavior in the international system, to give their actions meaning ( Innes and Steele 2014 :17; also Berenskoetter 2014 ). Just as individuals exist and find out who they are and how they should act through the stories they tell about themselves, so do the states ( Ringmar 1996 ; Lang 2002 ). At its strongest, this literature argues that states not only use biographical narratives to pursue certain policies, but that states are, in fact , biographical narratives themselves ( Berenskoetter 2014 ).
Reconceptualizing states as existing through narratives then helps explain autobiographical change, as well as continuity. As states move through international society, enter new relationships with other states, and experience momentous events, their stories change and incorporate new elements. This does not necessarily mean that a state narrative, its autobiography, must be fundamentally altered. Instead, new events are interpreted in line with specific elements of the narrative, emphasizing some aspects of the narrative and disregarding others. Therefore, “a coherent narrative can include all sorts of change as long as a sensible link from “before” to “after” is maintained” ( Berenskoetter 2014 :279). Most significantly for my purposes here, “not all action is guided, or made meaningful, by one single narrative. A master narrative is sufficiently vague to exist alongside more specific, derivative narratives that can be either layered or interwoven, and that can be strategically employed without hurting the coherence of the basic discourse” ( Berenskoetter 2014 :280).
It is specifically at times of great stress or trauma that autobiographical narratives are needed to provide comfort and relief ( Kinnvall 2004 ). Traumas or profound ontological crises occur when external events cannot be neatly placed into the ontological security narrative because they represent a challenge to the state internal or external identity ( Innes and Steele 2014 ). And while Innes and Steele expect states to revise and rewrite their collective narratives in response to the external trauma, I show instead how at times of great crises the state autobiographical narrative can remain essentially the same, but the policy change brought on by the crisis is narratively explained by activating some elements of the broader narrative template and deactivating others. In the face of crisis, a familiar narrative then preserves a sense of well-being and provides a form of “ontological self-help” ( Ejdus 2014 ).
In unpacking this narrative process, I agree with the criticism of ontological security theory as being at risk of “homogenizing the national Self” ( Delehanty and Steele 2009 :526) by not paying due attention to competing autobiographical narratives, such as different gendered narratives that constitute the Self in very different and consequential ways. My interest, however, is not to present an explanation of narrative competition, but instead of the complexity and multidimensional character of the state narrative itself, which political actors selectively activate and deactivate at times of great ontological stress. My research therefore provides the missing link—the explanation of how this narrative gets reconstituted at times of trauma to discursively bridge the gap between solving a physical security challenge that requires a policy change, and the continuation of the biographical narrative necessary for preserving state ontological security. I offer a response to the invitation by ontological security scholars for new investigations into “the relationship between trauma and national narratives, how such narratives smooth out the opened traumatic space or rupture, or how they can serve to reopen past traumas” ( Innes and Steele 2014 :28). It is to this process that I now turn.
Narratives and Policy Change
Narratives come in varieties of form and depth and breadth of salience. Not all narratives will be equally important for understanding state action. In this project, I choose to look at “schematic narrative templates” ( Wertsch 2008 ), narratives of general patterns across space and time, reflecting a single general story line, instead of focusing on specific narratives of individual events. I want to show how multiple individual historical narratives can merge into one, larger, narrative, which then becomes a frame for understanding both the past and the present in a simplified, schematic, and linear fashion. These larger narratives are used in “unreflective, unanalytical, and unwitting manner” ( Bartlett 1995 :45 in Wertsch 2008 :124) and are particularly prone to state control, production, and consumption. They represent basic plotlines for the most significant events of a state's history (for example, the persistence of the narrative of “expulsion of foreign enemies” in Russian collective memory as analyzed by Wertsch 2008 ) and are cognitive molds that impose a shape on how people remember their past ( Bruner 2002 ). A focus on multilayered schematic narrative templates can help provide a more empirically useful understanding of what narratives actually do , politically speaking, and how they can be used to set up a policy in place, as well as justify its change at times of great ontological stress.
State narratives are constructed through an active and elaborate process that involves multiple political and cultural agents. Over time and with infinite iteration by narrative “entrepreneurs”—political leaders, elite intellectuals, education establishment, popular culture, the media—and everyday social practice, a particular state narrative template (of past events, or of the general place of the state in the international system) fixes the meaning of the past and limits the opportunity for further political contestation. A constructed narrative reaches a tipping point threshold when a critical mass of social actors accepts and buys into it as a social fact. This state narrative then becomes an uncontested “rhetorical commonplace” ( Jackson 2003 ). It becomes hegemonic. Using discursive coercion, akin to what Bially Mattern calls “representational force” ( Mattern 2005 ), political actors rhetorically trap each other into the existing narrative. Identity claims, then, become a form of “power enacted through the narrative gun” ( Mattern 2001 :352). Alternative narratives stop making sense; they do not sound coherent and are not compelling. This meaning making then reduces political space for debate and conflict ( Krebs and Lobasz 2007 :410). Narratives, in other words, produce their own “lock-in effects” ( Goddard 2006 ).
Narratives, however, are social constructs—which means they are activated and deactivated all the time. Political actors use narratives strategically and purposefully. Through the use of framing, agenda setting, and discursive practices, political actors activate narratives or specific messages within narratives, to justify policy shifts, and deactivate those elements that no longer serve the policy purpose. As stories change over time and space, the identities of social actors embedded in these stories also change, shifting their preferences and interests. They may reconstitute their identity as political leaders by offering a new political path that draws a clear boundary with past practice. Their use of narratives will then be strategic. A new policy will be justified by some elements of the hegemonic narrative, while other parts of the story will be deemphasized.
Traumatic events such as wars or other political disasters are particularly useful windows of opportunity for selective narrative activation. Crises are events that are not only disruptive and consequential, but are “intersubjectively interpreted as necessitating change” ( Widmaier, Blyth, and Seabrooke 2007 :748). Political actors then respond to such interpretations of crises by designing a policy change. Narratives play a critical role in providing a cognitive bridge between necessary policy change (which resolves the physical threat brought on by the crisis) and uninterrupted state autobiography by preserving a well-established sense of self-identity. Narratives provide intersubjective meaning to policy change. They make policy change comprehensible and acceptable.
Narratives therefore bridge the rupture created by the external trauma between multiple desired securities of the state. As they pursue ontological security, states also pursue social security, security of their membership in international society. It is not enough for states to feel secure in their view of self; they also need to feel secure in the company of other states (for example, by being considered “European,” “modern,” or “democratic”). Narratives help provide these securities during challenging times. They rhetorically link what needs to be done (policy change) with why it is done (because we are a “good state” or because we are “European” or because this is “a heroic sacrifice” that we have historically been asked to do over and over again) and why the policy change should be accepted.
But, how are state narratives “measured”? How do we know that they are broadly shared and deeply embedded in a group? We can begin by assessing the salience of a particular narrative by assessing the degree of diffusion and internalization among both the elite and the general population ( Langenbacher 2003 :49–50). I do so by conducting personal interviews with elite actors, but also relying on public opinion surveys of the general public, as well as conducting a careful textual analysis of selected media reports, as well as a thorough reading of historiography, memoirs, government documents, popular culture (novels and films). Using historiography as a source is always fraught with danger, particularly of selection bias, omissions, and errors of interpretation ( Thies 2002 ). While I take these concerns seriously, I have to accept that there will always be disputes about the truthfulness of a particular interpretation, especially since I carry out the interpretation of the interpretation. I am careful to use strategies to minimize selection bias (mostly be expanding the pool of data and texts to analyze), and to be upfront about potential inconsistencies and biases in the interpretation.
Serbia is selected here as a “hard case.” A hard case is a useful theory testing technique because it presents the most unexpected outcome in relation to existing theories, therefore increasing our confidence in the theory that can explain the case ( Bennett and Elman 2007 ). Considering Serbia's deep attachment to the idea that Kosovo represents its core territorial identity, the policy outcome—Serbia giving up effective control of the territory and accepting Kosovo's de facto secession in 2013—must be explained. A cultural explanation based solely on the fundamental importance of narratives in guiding policy choices cannot explain Serbia's policy about-face. Conversely, a purely incentive-based explanation that focuses on international coercion or material benefits Serbia received as part of the Kosovo deal cannot explain Serbia's continuing invoking of the Kosovo narrative even after the territory was, effectively, gone.
In what follows, I trace the process by which different elements of the dominant Serbian schematic narrative template were selectively activated and deactivated to provide a cognitive bridge between physical insecurity (the looming loss of Kosovo) and continuing biographical continuity of the state.
Kosovo: The Politics of Space and Place
Serbia's foreign policy choices since the 2000 ousting of Slobodan Milošević have been puzzling. Serbia's unwillingness to arrest war crimes suspects has led to much delay in its EU accession process. So has its refusal to accept Kosovo's declaration of independence in 2008. These foreign policy decisions make little sense. Why would Serbia hold on to Kosovo, at all cost, when the territory does not provide any material benefit to Serbia—it is extremely resource poor, is inhabited by 90% Albanian population that has grown increasingly hostile to the Serbian state, and has always been a drain on already limited Serbian resources? But however irrational and unstrategic Serbia's policies are, they continue to make sense to large majorities of Serbia's citizenry and to its elite policymakers. If we want to understand why, we need to unpack the state narrative that shapes Serbia's view of self, its purpose in the international community, and its contemporary policy choices.
In what follows, I first describe the Serbian state narrative template of victimization and historical injustice, which is derived from the foundational Kosovo myth. I then discuss the political activation of this narrative as it shaped both Serbia's rhetorical refusal to let go of Kosovo and its de facto acceptance of Kosovo's independence.
Serbian Narrative Template
The dominant Serbian state narrative template is built on self-identity of a victimized people, engaged in an honorable, but often futile self-defense against great-power enemies. The saturation of the public discourse with this national story has been overwhelming. If the salience of a group narrative is measured by the degree of diffusion and internalization among both the elite and the mass population ( Langenbacher 2003 ), then the Serbian narrative of the nation victimized by malevolent global powers has firmly penetrated all aspects of Serbian political life ( Subotić 2013 ).
As is well documented, much of this victimization narrative is constructed on historical memories of Serbian martyrdom, especially the constitutive myth of Serbian martyrdom at the Battle of Kosovo in 1389. 2 Much research exists on the development of the Kosovo myth in Serbian epic literature and historiography (for example Malcolm 1998 ; Judah 2008 ; Djokic 2009 ), so I will not dwell on this for too long.
For my purposes here, it is enough to remind the reader that on June 28, 1389, a battle took place on the Kosovo (“Black birds”) field between Serbian Christian forces and Ottoman Turks. The exact outcome of the battle has remained historically unclear, but it is known that among the heavy casualties both the Serbian prince and the Ottoman Sultan were killed. Significantly for the purposes of narrative creation, the Kosovo battle has come to be remembered as a momentous Serbian defeat and the beginning of 500 years of Turkish rule ( Djokic 2009 :6).
Central to the Kosovo narrative is the concept of sacrifice. As the legend has it, the night before the battle, a holy prophet visited Serbian prince Lazar and presented him with a choice: an empire in heaven or an empire on earth. Prince Lazar chose a “heavenly empire”—securing earthly loss in battle, but eternal life in heaven for the Serbian people. This plotline is important because this idea—that there is something noble in foregoing practical political victories for an imagined spiritual benefit—has become one of the central tropes in subsequent Serbian state narrative, a way of making sense of events and justifying political decisions that otherwise do not bring much material gain. It represents, from the very beginning, the idea that for Serbia its ontological security trumps its physical security needs.
The second significant trope of the myth establishes in the narrative an understanding of the Serbian people as victims of major foreign powers. What began with the defeat at the hands of the mighty Ottomans has then continued throughout history as Serbian defeat or control by the Habsburgs, Croats and Slovenes (in the first kingdom of Yugoslavia), Germans in World War II, then again Croats and other Yugoslav nations (in the communist Yugoslavia), followed by Albanians (in Kosovo), Bosniacs (in Bosnia), Croats (in Croatia), NATO (during 1999 war), and most recently the European Union (in its attempts to legitimize Kosovo's secession from Serbia). In all of these battles—real or imagined—Serbian narrative positions Serbs as victims of outside powers, losing in political life, but gaining in righteousness and morality.
The third element of the Kosovo story that is relevant for Serbian state narrative template is the message that “Serbs will rise again,” that what is lost will be vindicated, what has been mourned will be celebrated. In this sense, the Kosovo myth establishes “a historical continuity between the contemporary Serbian nation and the “Serbs” of the Middle Ages, suggesting a perennial nation” ( Bieber 2002 :96). There are multiple historical examples of the use of this aspect of the Kosovo myth to celebrate or motivate Serbian heroes to avenge the Kosovo loss and bring back Serbian glory, for example, the decorative medals “to the avengers of Kosovo” that Serbian soldiers received after marching into Kosovo in 1912 during the First Balkan War ( Djokic 2009 ). In fact, it was exactly this call for Serbs to “rise up” that Slobodan Milošević activated in his notorious speech at the Kosovo field on the 600th anniversary of the Kosovo battle on June 28, 1989, which propelled his consolidation of power and political use of Serbian nationalism that soon led to war ( Bieber 2002 ).
This narrative construction was, of course, not a coincidental development. After being relatively dormant during the communist era (especially 1945–1987, before Milošević's takeover of the ruling party), since the mid-1980s, as nationalism replaced communism as the principal ordering ideology, the story of Kosovo and the ideas it represented became more openly activated. Leading Serbian historians published hugely popular books linking the Kosovo myth to Serbia's contemporary political anxieties, such as the fundamental insecurity about Serbian national interests being threatened by other nations in the Yugoslav federation. Kosovo became tied quite directly to core Serbian national identity, as in the following excerpt from the book For a Heavenly Kingdom by a noted historian:
Nations have their metaphysical core, with some this is impulsive and with others it is hidden, sometimes even powerless…. The Kosovo orientation is not [only] a national idea, but also a trait of character which makes a Serb a Serb . (Samardžić 1991, 14, emphasis added)
The political purchase of the Kosovo myth for the Serbian nationalist endeavor of the 1990s was to eliminate the historical distance between past and present and, in effect, equate contemporary political leaders such as Slobodan Milošević, with historical Serbian figures (such as Prince Lazar), and also group contemporary enemies (such as Kosovo Albanians or Bosniacs) with historical enemies (such as the Ottomans). The vindication component of the Serbian state narrative—that Serbia will avenge its historical losses—to a great extent helps explain Serbia's war efforts in the 1990s. Historical vindication, as we know, is never a bloodless sport, and this pursuit of “historical justice” goes a long way in explaining the narrative meaning making behind Serbia's brutal wars in Croatia, Bosnia, and Kosovo.
In 1999, NATO intervention against Serbia further solidified Serbian feeling of victimhood and a great sense of injustice at the hands of great powers ( Jansen 2000 ). In fact, a succession of Serbian governments post-1999 consciously decided to promote the memory of Serbian victimhood during the 76-day NATO air war as if it were the central motif of the wars in the 1990s ( David 2013 :191).
This sense of collective victimhood persisted even after the Kosovo war, and even after Milošević was ousted from power in 2000. Serbian media relentlessly covered “revenge” attacks by Kosovo Albanians and the persecution of Serbs by the Kosovo Liberation Army, as well as real and documented destruction of Serbian cultural sites. The stories about renewed cycle of persecution of the Serb minority permeated Serbian public discourse, once again confirming the preexisting notions that Serbs were right to fear Albanians and that Serb suffering will continue in the new, non-Serb controlled Kosovo ( Bieber 2002 :105). The Serbian Orthodox Church continued as the primary incubator of this narrative.
This particular state autobiography of injustice, victimization, and need for vindication created a cognitive model for Serbia to understand and interpret its world. It designated political action by historical foes (Slovenes, Croats, Bosniacs, Kosovo Albanians) as threatening, a continuation of a timeless and perpetual struggle of the Serbian nation for survival. It also reconstituted elements of the Serbian national identity. This was clear already in the inaugural speech of president-elect Vojislav Koštunica on the eve of Milošević's defeat in 2000: “There are those who did us wrong, who bombed us. We cannot forget the damage or the crimes [against us]; Serbs will lose their identity if they forget those crimes” ( B92 , October 6, 2000, emphasis added). This narrative created a quite distinct sense of a “historical time in which national enemies fought over many decades and even centuries,” creating “direct, living connection with the past” ( Čolović 2002 :5; also Lieberman 2006 ).
Serbian State Narrative Activation
The construction and staying power of the narrative should by now be clear. The more important question, however, is how does this narrative continue to shape Serbia's contemporary policy choices and how does it explain policy change.
Serbia's position on Kosovo is not very well understood, while it is very broadly assumed and often mentioned as fact in international policymaking. Most analyses of Serbia's resistance to even negotiate, let alone recognize Kosovo's independence, explain it as an issue of territorial integrity (sovereignty claim) or ethnic chauvinism (nationalism claim). In his famous review of learning in foreign policy, Jack Levy mentions in passing and without problematizing that “Kosovo drives the Serbs” ( Levy 1994 :279), as an example of the importance of history to contemporary foreign policy. Even advocates of Kosovo's independence uncritically state that Kosovo is Serbian “ancestral homeland” ( Kupchan 2005 : 14), as if that is somehow a real and observable social fact. This idea that Serbian claim to Kosovo is “historical,” while Albanian claim is “romantic,” and therefore not equally legitimate, is also a frequent trope in Serbian contemporary historiography ( Di Lellio 2009 :375). To a great extent, therefore, the Serbian state narrative about Kosovo has been internationalized and has shaped international responses to the territorial dispute.
Ever since the end of the NATO war in 1999 and the placement of Kosovo under international protection, the international community primed Serbia for Kosovo's eventual independent status. The prospect of losing Kosovo was deeply felt and widely perceived as a profound blow to Serbian identity and the Serbian state. For most of Serbian elites and the Serbian people, this was an unacceptable political outcome. In a 2011 poll, 65% of the population declared that Serbian government should never recognize Kosovo because it would compromise Serbian “identity and national honor” ( Belgrade Center for Security Policy 2011 ). This strong opposition to Kosovo's independence holds even in the face of clear political costs. In the same poll, 54% of respondents were aware that Serbia could not join the EU without giving up its claims on Kosovo, but still overwhelmingly opposed Serbia's recognition of Kosovo's independence. This national sense of loss, tragedy, and betrayal was shared across the entire Serbian political landscape, and enveloped conservative nationalists as well as moderates and reformers. The crisis over Kosovo's declaration of independence created multiple insecurities for Serbia—a sense of physical insecurity (loss of significant part of territory) and a sense of ontological insecurity (loss of self-identity).
For example, Serbian reformist president Boris Tadić lamented:
Kosovo is where my nation's identity lies, where the roots of our culture are… Kosovo is the foundation of Serbia's history and this is why we cannot give it up. ( B92 , May 25, 2007)
Serbian foreign minister Vuk Jeremić, another self-declared political moderate, made an even more emotional statement:
I stand before you this afternoon as a proud European, and as an ashamed European… I am ashamed because if recognition of this ethnically motivated act of secession from a democratic, European state is not wrong—then nothing is wrong. ( B92 , February 20, 2008)
The Serbian government's claim to Kosovo from Serbia's position as a “European” state is not accidental. It builds on a constant trope in the Serbian state narrative, which delineates a clear opposition between the crescent and the cross, the Islamic East (of Kosovo, of Muslim Bosnia), and the Christian West (of Serbia and of Europe, or more precisely of Serbia in Europe). In this sense, the external crisis—secession of Kosovo—also threatened Serbia's social security, the security of its membership in the society of European states.
Serbia's former president and prime minister Vojislav Koštunica made this link between Serbian narrative and contemporary policy even more clear in a speech on February 15, 2008, 2 days before Kosovo declared its independence from Serbia:
Whoever has heard about Serbs and Serbia knows that Serbia is in Europe and the Serbs are a European nation. And two centuries ago it was not just Serbia who discovered Europe but Europe discovered Serbia as well. And when it did that, it found within Serbia European ideas and values and Kosovo as a synonym for the most valuable contribution made by Serbia to the Christian civilization. Therefore, no one can integrate Serbia into Europe or leave it out, and Serbia should enter the EU whole, just as all its other member countries did. ( Koštunica 2009 :207)
Koštunica's party, the Democratic Party of Serbia (DSS), then directly linked Kosovo's independence with Serbia's EU bid and called for Serbia's withdrawal from EU membership process. In its official party platform, DSS says:
The European orientation of Serbia should be called into question for a very simple reason: who in Serbia is ready to believe that someone who is part of the hostile context, notably the process of establishment of Kosovo status, may in any other matter have friendly intentions. Advice like “Let go of Kosovo, ahead of you is European future” is unacceptable for Serbs. ( NIN , February 8, 2007)
The narrative continuity in this policy choice is clear—the withdrawal from Europe may be a short-term political loss, but it will be a heroic sacrifice for the greater good—the preservation of Serbia's identity, its political soul. This theme of Kosovo's indivisibility, the unimaginable demands for Kosovo's separation from the Serbian body politic, is also evident in a prevalent theme in the Serbian Kosovo narrative that equates Serbia with Jerusalem ( Ejdus and Subotić 2014 ). The Kosovo/Jerusalem analogy first appeared in the Serbian public consciousness in 1985, when the novelist Vuk Drašković wrote, “Serbs are the thirteenth, lost and the most ill-fated tribe of Israel” (quoted in Živković 2011 :199). The Kosovo/Jerusalem meme has since become a common rhetorical heuristic by Serbian social actors. For example, Serbian foreign minister Jeremić used the “Kosovo as Serbian Jerusalem” metaphor in multiple speeches at the United Nations, interviews, and public events on numerous occasions ( Ejdus and Subotić 2014 ). The Church has been the persistent generator of this theme, as in the statement by a high-ranking bishop that Kosovo should stay “our spiritual and cultural cradle, our Serb Jerusalem. What Jerusalem is for the Jewish people, Kosovo and Metohija is for the Serbian People” (quoted in Di Lellio 2009 :380).
While Serbian emotional attachment to Kosovo is well documented, I suggest a better way to understand Serbia's position here is to consider that Serbia's claim to Kosovo is really not about Kosovo at all. Kosovo, the actual , real Kosovo, the land, the territory, and the people are not what “drives the Serbs.” It is the idea of Kosovo , what Kosovo represents in the collective memory of Serbia's past and present that is the driving force of Serbian political action. The idea of Kosovo is the memory of Serbia's past greatness, its victimization and suffering at the hands of great powers, and the desire to rectify the injustice and to make wrong right. The much-discussed “Kosovo myth” is, therefore, not a myth of territory, but a myth of future action. It is forward looking, not backward looking, and this is why it is directly relevant for Serbia's foreign policy choices.
Serbia could not “give up” Kosovo because Kosovo represents foundational myths that provide the state its continuing ontological security—the myth of Serbia's glorious past as a regional power, the myth of Serbia's cycle of defeat at the hand of empires, and, most important, the myth of Kosovo as the territorial epicenter of Serbia's identity. This is what Serbian politicians mean when they say, “there is no Serbia without Kosovo.” What they really mean is—the biographical narrative about what Serbia is, what its past was like, and what its future holds—cannot exist the same way without Kosovo, or better yet, without the idea of Kosovo.
The hegemonic state narrative that has been constructed over decades has made the Kosovo issue indivisible in the minds of Serbia's public as well as its elites, in much the same way as comparable narratives in Ulster and Jerusalem ( Goddard 2010 ). The narrative did not leave any room for policy compromise. Kosovo's independence and secession from Serbia have become unimaginable not just because of territorial claims, but also because of the tropes of the narrative that have become constitutive parts of Serbian national identity—victimization, injustice, and national revival.
Narrative Deactivation and Policy Change
The challenge to Serbian political leaders, therefore, was to preserve this idea of Kosovo while at the same time continuing with EU integration and reaping political, economic, and cultural benefits EU membership would bestow. Since 2010, the Serbian government began to gradually abandon its claim of effective territorial control over Kosovo through a series of EU-sponsored Serbian-Kosovo talks. At the same time, the official position that Serbia will never recognize Kosovo, the core of its national identity, and the source of its ontological security, remained unchanged.
In March 2013, Serbian Prime Minister Ivica Dačć displayed this contradiction by claiming, first, that the Serbian people have for years been “lied to that Kosovo is ours” ( NIN , March 7, 2013). Later that same day, in a different outlet, he stated that Serbia would never accept Kosovo's independence ( Balkan Inside , March 7, 2013). Clearly contradictory, this position however was representative of Serbian public opinion, as 63% of Serbian citizens recognized that Kosovo was, in fact, independent, while at the same time 65% demanded that the government keeps Kosovo as part of the Serbian state ( Balkan Insight , March 5, 2013).
Bridging this cognitive divide required some creative use of existing narrative tropes. The Kosovo narrative already contains the plot that Serbs lost in battle, but they sacrificed themselves for the greater good. As the EU ratcheted up pressure on Serbia to relinquish its territorial claims on Kosovo in exchange for EU accession negotiations, Serbian political actors began to activate the “sacrifice” element of the state narrative, while deactivating other narrative features (for example, that of imminent return). Vuk Drašković, the same novelist who was the first to activate the “Kosovo is Jerusalem” narrative trope, now called for Serbia's recognition of Kosovo's independence. However, the rhetorical pivot of this new argument is notable: “killing of Serbia for the sake of preserving something that only exists as a mirage must stop” ( B92, March 25, 2012). Again, Serbia is being “killed,” sacrificed, for a spiritual illusion ( Ejdus and Subotić 2014 ).
Serbia and Kosovo finally signed an agreement on April 19, 2013. The deal requires Serbia to accept Kosovo government's control over the entire territory, while Kosovo was to grant significant autonomy to Serbs concentrated in the country's north. The deal also allows Serbia to continue to refrain from officially recognizing Kosovo as a state ( Ejdus and Subotić 2014 ).
This remarkable policy change was now to be somehow presented to anxious and scared population, who was told for years that Kosovo was always to remain Serbian. Senior church leaders and a few thousand Kosovo Serbs organized a protest rally in Belgrade, where one of the bishops carried out a ceremonial “burial” of the Serbian government, and another accused Serbian leaders of treason, because “there is no Serbia without heavenly Serbia” ( B92 , May 10, 2013).
Serbian politicians responded mostly by minimizing the magnitude of Serbian concessions ( Ejdus and Subotić 2014 ). They used a variety of rhetorical tools to discursively deny that any real policy change had in fact taken place. Serbian President Tomislav Nikolić pledged to the public, “we would never cut our wrists and commit suicide by giving up Kosovo” ( B92 , May 17, 2013). More significantly, the Serbian deputy Prime Minister Aleksandar Vučić again activated the sacrifice narrative trope, when he said,
The agreement is the only way for Serbia to survive, for us to stay united and solve our problems together in the future… This is a difficult agreement, causing many problems for the Serbian people, but it was the only possible solution at the moment… Sometimes we must make difficult decisions, but a state cannot survive without its people, and the people cannot survive without its state. ( Radio Free Europe , May 12, 2013)
Serbian political actors therefore linked a dramatic policy shift, a change that is rhetorically unimaginable, with the existing and broadly understood elements of the Serbian state narrative. Again, they implied, Serbs are sacrificing their earthly kingdom for spiritual rewards such as national unity, justice, and love.
The second narrative trope Serbian leaders activated in the run up to and in the immediate aftermath of the 2013 Kosovo deal was the idea of great-power control. As presented earlier in the article, that Serbia has been historically victim of great-power manipulation and injustice is one of the constitutive elements of the Serbian state narrative with broad reach and massive popular acceptance. In addition to the sacrifice trope, the great-power element of the narrative was particularly useful in justifying Serbian foreign policy change to the confused public.
The idea that it was the great powers that made Serbia “give up” Kosovo was everywhere in the public discourse. Serbian Prime Minister Dačić, for example, said,
It was the great power meddling that contributed to the disagreements in the Balkans. So, which great power has the most influence in the Balkans? Not just one, but each one a little, because each great power has “its” people in the Balkans. We do the others’ bidding instead of working for ourselves. ( B92 , December 5, 2012)
Serbian president Tomislav Nikolić similarly argued, “the international community doesn't have a plan [for Kosovo], it just follows the great powers” ( B92 , March 25, 2013). Serbian Patriarch Irinej called upon Russia, “as a great Slavic power,” to help resolve the Kosovo problem in Serbia's favor ( B92 , July 23, 2013). During this period, a great deal was made in the Serbian media about “famous foreign experts” declaring the Kosovo problem was the consequence of great-power politics. Some of the usual suspects were called upon to comment. So, Noam Chomsky, whom the Serbian interviewer introduced as “the most important and most influential world intellectual and the most famous United States dissident” declared that Serbia is forced to give up Kosovo under Western pressure because “the world is fundamentally ruled by force, not by law” ( Prelević 2013 ). This interview got a tremendous amount of play in the Serbian media because it directly confirmed the “great powers” narrative trope that was so important for the justification of the Serbian Kosovo policy shift.
Similarly, a foreign policy expert from the Moscow Institute for International Relations was broadly cited in the Serbian media for arguing that the state of Kosovo was the creation of great-power interests who needed a proxy to control international financial flows ( Mondo , February 18, 2013). Finally, the obsession with great powers so saturated the media discourse that even unrelated news items, such as US Secretary of State John Kerry's statement, “ the international community is ready to talk to Iran” ( US Department of State 2013 ) got translated by the official Serbian news agency and then reported by every single media outlet as “ the great powers are ready to talk to Iran” ( B92 , February 8, 2013). Great powers—they truly were everywhere.
It is during this process of difficult policy change and the recognition that some form of “loss” of Kosovo is inevitable that Serbian political and social actors strategically activated some elements of the Serbian master narrative (“sacrifice” and “great power injustice”) and deactivated others (“inevitable return”). Putting the two strands of the narrative together, the loss of Kosovo was narrated as a major Serbian sacrifice, the result of another deep historical injustice at the hands of great powers, this time the European “capture” of Serbian territorial and spiritual core. The narrative lives on, but its political implications change.
At times of great external stress, such as the secession of territory, familiar narratives help bridge the gap between a difficult policy change that is necessary for state physical security, and the need to maintain state biographical continuity and ontological security. During the period of high anxiety, political actors do not create new narratives from scratch, nor do they significantly rewrite the existing ones. Instead, they use the wealth of multiple narrative tropes that exist in dominant state narratives to activate some elements of the narrative and deactivate others. This process manages to preserve the larger narrative template, thus maintaining a sense of order, stability, and ontological peace.
But, if actors have constructed the narrative differently, or used different parts of it, the outcome would have been different. For example, if the Serbian political actors activated the “inevitable return” part of the Kosovo narrative, the aftermath of the Kosovo war would have looked quite different than if they understood and projected outward the narrative of “sacrifice” and “historical injustice.” This dynamic process of narrative activation and deactivation should not imply that these narratives are fickle and malleable to the point of becoming meaningless. Instead, this proves once more that “there are always more stories to tell and more ways of telling them” ( Gotham and Staples 1996 :493).
But an explanation based on narrative also needs to address possible alternative explanations. In the Serbian case, could the policy shift be explained as simply Serbian political defeat under tremendous international pressure and consequent face saving in front of an angry public? While I do not depart from the material explanation that Serbia had to make the Kosovo deal if it was to continue its EU accession, this explanation only gets us so far. If this was really such a straightforward deal for Serbia to make, then why not simply go out and frame it as a victory of Serbian foreign policy that created a path toward Serbia's EU membership? Why did this policy change take so long—Kosovo has been under international protection since 1999 when Serbia's territorial control effectively dissolved? Why did the Serbian government not frame the Kosovo deal as “good riddance” of a resource poor and hostile territory and population? None of these outcomes occurred. Instead, political actors faced a continuing need to maintain the idea of Kosovo being constitutive of Serbian identity even after the infected appendage has already been amputated. A rationalist explanation cannot account for these discursive choices. A focus on biographical continuity through narrative activation/deactivation I presented here can. The persistence of the Kosovo narrative in the Serbian national consciousness, therefore, did not cause either Serbian rejection of Kosovo's independence or its ultimate de facto acceptance. Rather, what the Kosovo narrative meant to political actors shaped the particular form that the policy change took, and consequent ways to interpret it within the intersubjectively shared narrative frame.
While the empirical focus of this article was on Serbia, this analysis can be applied to great benefit in other cases. For example, the ongoing narrative battleground between Russia and Ukraine would be a particularly fruitful case for analyzing the process by which multiple tropes of the Russian schematic narrative template (such as, foreign invasion, reconstitution of the lost empire, or pursuit of international status) were activated or deactivated as was necessary to pursue Russian foreign policy at times of clear and profound ontological stress the Russian state is experiencing in early twenty-first century. This analysis, therefore, is not limited to only small or weak states. All states have narratives and all states face threats to various aspects of their security. My argument then, viewed more broadly, provides an illustration of the necessary analytical synthesis of consequentialist behavior of political actors and normative social frames that guide political action. It also helps move the scholarship on ontological security forward, by presenting another possible mechanism for integrating narratives more directly into the study of state ontological security-seeking behavior.