‘Not a heap of stones’: material environments and ontological security in international relations

I thank Jelena Subotić, Iver Neumann, Brent Steele and three anonymous reviewers for their helpful comments. An earlier draft of this paper was presented at the International Studies Association Annual Conference, New Orleans, 2015.

Abstract

Extant scholarship on ontological security in international relations has focused on the significance of social environments for state identity. In this article, I argue that material environments also provide an important source of ontological security for states. In order to assume this role material environments need to be discursively linked to state identity through either projection or introjection. Once incorporated into state identity narratives, material environments become ‘ontic spaces’: spatial extensions of the collective self that cause state identities to appear more firm and continuous. However, ontic spaces are inherently unstable and require maintenance, especially during periods of crisis or transition. States bear agency in this process but they never achieve full control, as identity discourses are continuously contested both domestically and internationally. I illustrate these claims by looking at the role of the General Staff Headquarters in Belgrade, destroyed by the North Atlantic Treaty Organization in 1999, in the ontological security of Serbia.

Introduction

In the heart of Belgrade, astride the main urban artery and right in front of the seat of the Government of Serbia stands a colossal ruin. Built in the 1960s, this gem of socialist architecture housed the Yugoslav General Staff for more than three decades before North Atlantic Treaty Organization (NATO) missiles destroyed it in 1999. Seventeen years on, Yugoslavia is long gone, Serbia is a NATO partner and a candidate for European Union (EU) membership, but the ruined building remains in existence. Over the years, the ruin has become an integral part of the surrounding cityscape and most Belgradians take it for granted. However, anyone who has visited the city in the past 17 years and noticed this uncanny edifice will be bewildered by its puzzling persistence. Why has a ruin so big, so central and so historically important been left as it is for such a long period of time? I argue that the continued presence of the ruined General Staff Headquarters (HQ) is not the result of economic hardship or bureaucratic gridlock, but a way for Serbia to satisfy its ontological security needs. The ruined General Staff building has been introjected into Serbia’s state identity narrative and has thus become an ontic space that needs to be maintained.

The extant international relations (IR) literature has so far overlooked the role of the material environment, either natural or built, in the ontological security of states. In this paper I demonstrate that trust in the constancy of the material environment, including familiar cityscapes or significant landscapes, is a source of ontological security in world politics just as important as a state’s relationships with its significant others. However important for the grounding of the self they may be, routinized international interactions are never fully controllable or predictable. States therefore need the material environment as an additional anchor of their self-identity narrative. In order to assume this role, natural landscapes, ancient or contemporary cityscapes, landmark monuments and publically recognizable cultural sites need to be discursively linked to the project of the state self.

The linking of identity to material environments can take the form of either the projection of state identity narratives onto material environments or the introjection of material environments into state identity narratives. Once incorporated into the project of the self through projection or introjection, material environments become ontic spaces—spatial extensions of the collective self that help states ‘bracket out’ the inherently fragmented, contested and contingent nature of their identity narratives and achieve the sense of continuity in the world which is necessary if a state is to have purposeful agency. Ontic spaces are inherently unstable and the link between the self and the material environment requires continuous maintenance, especially during times of crisis and transition. While states bear agency in this process of grounding the self in the material environment, they never achieve full control. The state identity narrative is never absolutely accepted by the entire society and other international actors, and it is particularly prone to strong contestations in times of uncertainty. When societal or international actors destabilize their narrative of the self, states may resort to the reappropriation of ontic spaces in order to fend off threats to their biographical continuity.

The article proceeds in the following way. In the first section I outline the main facets of the ontological security literature in general and the treatment of the material environment in particular. In the second section I discuss the role of the material environment in the ontological security of states and develop the concept of ontic space in IR. In the third section I illustrate these theoretical arguments with the case of the General Staff HQ in Belgrade and its role in Serbia’s ontological security.

Ontological security: literature review

The concept of ‘ontological security’ was first coined within psychology, where it denotes a need of individuals for trust in and constancy of relationships with significant others, especially during infancy (Erikson 1968; Laing 2010). According to Giddens, who translated the concept into sociology (1984; 1990; 1991), ontological security is a ‘confidence that most human beings have in the continuity of their self-identity and in the constancy of the surrounding social and material environments of action’ (Giddens 1991, 92, emphasis added). The continuity is maintained through a routinization of relationships with significant others which helps actors create a ‘protective cocoon’, ‘bracket out’ existential anxieties (Giddens 1991, 44) and achieve continuity of self-identity or ‘the persistence of feelings of personhood in a continuous self and body’ (Giddens 1991, 55).

For Giddens, the material environment of action is of utmost importance for understanding ontological security processes. The routinization of relationships, so central to the sense of biographical continuity, is always situated in certain locales or, as Giddens calls them, ‘settings of interaction’. Locales, he writes, ‘may range from a room in a house, a street corner, the shop floor of a factory, towns and cities to the territorially demarcated areas occupied by nation-states’ (Giddens 1984, 118). They are never defined by their physical property alone but by meanings attributed to them. This ‘sense of place’ is an important source of ontological security because it provides, according to Giddens, ‘a psychological tie between the biography of the individual and the locales that are the settings of the time-space paths through which that individual moves’ (Giddens 1984, 367).

The role of the material environment of action for ontological security was further explored in studies focusing on the role of home and dwelling (Dupuis and Thorns 1998; Padgett 2007; Newton 2008; Hiscock et al. 2001). These studies have shown that, as sites of constancy, homes shield people from the unreliability and impermanence of the outside world. As such, homes allow people to achieve a sense of continuity, maintain self-identity and realize full agency. But, however crucial it may be for ontological security, as Giddens has rightly pointed out, home is not the only locale that provides a sense of constancy.

Other studies have looked at ‘homely spaces’ such as national borders, which, as Skey has pointed out in his study on travelling, ‘also matter in an ontological sense because they are fundamental in representing the nation (and indeed, the world) as a known and knowable entity that can be seen, experienced and relied upon’ (Skey 2011, 240). Similarly, Margareta Rämgård has shown how important childhood places are for what she calls ‘place security’. During transition periods, people are especially aware of this need for ‘place security’. Rämgård shows how, for example, pregnant women try to resurrect the feeling of ontological security by revisiting childhood places for therapeutic reasons (Rämgård 2006).

Recent studies have suggested that communities too can draw their sense of ontological security from their built environment. Jane Grenville, who translated the ontological security argument into the field of urban conservation, compared how different cities and communities cope with conservation and urban development (Grenville 2015; 2007). Her findings suggest that, in the aftermath of political turmoil, ontological anxieties may result in the desire either to completely change one’s environment or to retain it unaltered ‘as a bulwark against a transient and untrustworthy external world’ (Grenville 2007, 451). Communities with a more stable sense of self-identity seem to be more prone to moderate conservationist solutions, with a creative combination of old and new. In contrast to this, ontologically insecure communities, according to Grenville, incline towards radical solutions of either completely restoring the old or building everything anew (Grenville 2007, 458).

In recent years, the concept of ontological security has also travelled into the field of IR. Ever since Alexander Wendt made a distinction between physical security, or ‘differentiation from other actors’, and ontological security, which has to do with the freedom from anxiety resulting from a predictable relationship with the world (Wendt 1994, 385), the literature on ontological security in world politics has grown exponentially (Huysmans 1998; McSweeney 1999; Mitzen 2006; Steele 2007; 2008; Kinnvall 2004, 2006; Krolikowski 2008; Delehanty and Steele 2009; Zarakol 2010; Lupovici 2012; Youde 2014; Subotić 2016; Rumelili 2014, 2015).

One of the central debates within the IR literature on ontological security is about the unit of analysis.¹ It revolves around the question of which entities can be analytically treated as ontological security seekers. While originally the theory was developed in social psychology and sociology in relation to individuals, most of the authors using the concept in IR extrapolate the logic to states (Mitzen 2006; Steele 2008; Zarakol 2010). Taking cues from the state-as-actor debate in IR more generally (Wendt 2004), Mitzen and Steele have offered four defences of such anthropomorphization of states: (1) everyone is doing it; (2) states provide ontological security to their members; (3) ontological security explains macro patterns in decision-making; and (4) state representatives act ‘as if’ states are ontological security-seekers (Mitzen 2006, 351–353; Steele 2008, 15–20).

This translation of an individual-centred concept from social psychology and sociology into a state-centred concept in IR has been critiqued by a number of authors. Alana Krolikowski for example posits that ‘resorting to the assumption of state personhood obscures important aspects of how the state, as an evolving institution, affects individuals’ sense of ontological security’ (Krolikowski 2008, 111). Paul Roe agrees with Mitzen’s assumption that states are providers of individual ontological security, but argues that it doesn’t follow that states, like persons, can also have the need to be ontologically secure (Roe 2008, 785). In his view, ontological security-seeking is an emotional preference of an individual, whereas the state or any other social group is no more than a larger material and discursive framework within which individuals build their self-identities.

While it is impossible to give justice to all the nuances and richness of this rapidly growing research agenda in IR here, several general remarks are in order. The extant studies on ontological security in IR have significantly advanced our understanding of different levels of agency in ontological security processes, from individuals, through nation-states to international organizations. Also, the research agenda on ontological security in world politics has shed a new light on a variety of other concepts in security studies, such as securitization, security communities, security dilemmas and conflict resolution.

Most importantly for this article, virtually all of the existing studies in IR have focused exclusively on the role of the social environment in ontological security in world politics.² In other words, they have been investigating how relationships with significant others, be they friends, partners, competitors or enemies in world politics, affect the ability of states to achieve biographical continuity. Unfortunately, the role of material environments, such as architecture, natural landscapes or other locales from which states can draw their sense of continuity in the world, has remained largely unaddressed in IR. In the next section I develop some ideas on how to start filling this gap.

Ontological security and material environment in world politics

Previous works have established that material environments serve as an important source of ontological security for individuals by shielding their everyday routines from transience of the outside world. Drawing on the state-as-actor approach discussed above, I posit that states also require constancy in their material environment in order to have a sense of continuity in the world. Stable and predictable social relationships, either internally within states or externally with other international actors, are indeed an important source of the ontological security of states. However stable and routinized social relationships may appear to be, they are never fully predictable. Consequently, states need an additional, material anchor for their collective self-identity narratives which will stabilize their sense of self and conceal or mend its essentially contested, fragmentary and plural nature.

Not all material environments, however, are of equal relevance to ontological security of states. While individuals draw their sense of constancy from ‘homely places’ through embodied routines of everyday life, collective actors such as states need to discursively link their self-identities into their material environments. Some locales that have little functional utility or material value can be imbued with higher, even sacred meaning while others that seem to be of much greater practical value may bear little importance for collective identities. The relevance of natural or built environments to the ontological security of states, therefore, stems not from their inherent properties but from the meanings with which they are imbued.

When familiar and symbolically important material environments are incorporated into state identity narratives they become ‘ontic spaces’, spatial extensions of the collective self that cause state identities to appear more firm and continuous. In the face of transient social relationships ontic spaces serve as the material bedrock that stabilizes state identity and provides a material anchor of its agency. As visible, tangible and durable extensions of the state self, ontic spaces enable collective identities to appear more firm, continuous and real. By anchoring themselves in the material environment, states 'bracket out' fragmentations and contestations of their identity narratives. This allows them to maintain biographical continuity and develop a sense of collective agency.

Ontic spaces are constructed through a process of discursive linking which can take two distinct forms. The first form of linking material environments to collective identity narratives is projection, which involves extrapolation of the self onto the material environment as if it were a screen. As Neil Leach writes, ‘the nation, in effect, needs to read itself into objects in the environment in order to articulate that identity’ (Leach 2006, 85). When it comes to the built environment, projection usually starts in the planning phase and involves the design of an object so that it can directly represent a collective identity. This is the case with sites of great symbolic importance such as seats of governments, religious centres or historic monuments that represent polities and serve as their repositories of memory. In contrast to structures erected during the period of romantic effusion with the nation-state, contemporary landmark cityscapes use more subtle narration of the self. This is the case, for instance, with the One World Trade Centre built at Ground Zero in New York. Here, the national script is subtly projected onto the structure through its 1776 feet tall Freedom Tower, obviously alluding to the signing of the United States (US) Declaration of Independence (Jones 2006, 558).

Polities can also read their identity narratives into the natural environment. For example, an important component of the Zionist enterprise, especially since 1948, has been to return the Israeli landscape to its biblical-era shape. The principal way of projecting the Zionist self-identity narrative onto the land has been through agricultural practice. By reintroducing plant species mentioned in the Bible, some of which disappeared from the area centuries ago, Zionist agriculture has had the role of securing the continuity between the golden age, present times and the promised future. How important this was for the nascent Israeli state is best illustrated by the fact that Israel launched a secret operation of transferring 75,000 date palms from Iraq into Israel in 1955 (Weiss 2010, 206).³ Another iconic example of projection is the ‘natural monument’ at Mount Rushmore, South Dakota. By featuring the carved faces of four US presidents, Washington, Jefferson, Lincoln and Roosevelt, the monument projects four phases in the first 150 years of US history: birth, expansion, preservation and development. Historian Herbert Samuel Schell depicts the mountain-sculpture ‘as a symbol of greatness and durability which embodies the dreams, ambitions, and accomplishments of the American people’ (Schell 1975, 378).

The second form of linking material environments to state identity narrative is introjection and involves absorption of the material environment into the project of the self (Leach 2006, 78). Introjection entails the appropriation and incorporation of physical objects into collective self-identity narratives. Perhaps the most widely diffused practise of introjection is to simply delineate a space and ascribe it a special status as a place where important imaginary or real nation-forging events happened. Most if not all states have ‘ethnoscapes’ (Smith 1999) or ‘core territories’ (White 2000, 41) of paramount importance for national identity. When imbued with religious symbolism these landscapes acquire sacred status that further strengthens emotional attachment to them. For example, Kosovo was introjected into Serbia’s national identity narrative as the location of a mythic battle in 1389, thus facilitating a sense of the trans-historic continuity of the Serbian nation (Subotić 2016). Another example is the introjection of Jerusalem and the Holy Land into the Zionist identity narrative (Sand 2010; 2012).

Introjection can also be achieved through a narrative that depicts national identity as a product of particular natural landscapes. Thus, for example, discourses on national identity started to emerge in the late twentieth century portraying the Alps as the landscape that transformed poly-ethnic Switzerland into a homogeneous whole (Kaufmann and Zimmer 1998). A similar way of introjecting the natural environment into collective self-identity is by portraying particular landscapes as reflections of the national character. For example, in England it is ‘the South’, tame and civilized, which has been constructed as a reflection of true Englishness as opposed to the rugged periphery in ‘the North’ (Shields 2013, 231). In contrast to this, in Canada it is widely held that it is the tough North that expresses the national spirit, while in Scotland, Switzerland and Austria a similar quality is ascribed to the countries’ harsh mountains (Palmer 1998).

Built environments can also be introjected into collective identity narratives and thus turned into ontic spaces. One way of introjecting built environments into state identity narratives is through discourse on national heritage. While restorative practices date back to the dawn of human civilization, the nationalization of heritage emerged in Europe with the advent of the nation-state (Swenson 2013). In Germany, for instance, the idea of a national heritage in Germany was born out of a longing for national unity in the aftermath of the Revolutionary Wars (Nipperdey 1976 in Swenson 2013, 14).

Conversely, the dispossession of heritage can have adverse effects on the ability of states to maintain their national identity narratives. For example, Kosovo is considered to be the cradle of Serbian statehood and medieval monasteries and churches are seen as the material evidence of this (Defrese 2009) As such, they are the material anchor of Serbia’s claim to trans-historic continuity and a source of its ontological security. The tendency of the authorities in Priština to appropriate medieval Orthodox Christian heritage as ‘Kosovar’ has been perceived in Serbia, to use the words of Serbia’s former foreign minister, as a ‘systemic campaign of identity theft’ (B92 2011).

Another way of introjecting built environments into national identity scripts is through archaeology. The emotional power of archaeology, as Silberman has pointed out, is that it links ‘the present to a particular golden age’ (Silberman 1995, 295). Through archaeological excavations, stories about time immemorial become ‘tangible’ and therefore appear more real. This helps nation-states maintain their biographical continuity and fend off existential anxieties. For example, archaeology in Israel has been used as one of the key instruments for establishing continuity between the biblical past and the contemporary Jewish state (Gori 2013, 216). By focusing primarily on the biblical period, it has been clearly used as an instrument of legitimation of the Jewish state. Moreover, by obscuring pre-Jewish and Arab cultural heritage it has also been an instrument of de-legitimation of Palestinian political claims. In the words of Benjamin Netanyahu, ‘The Western Wall Tunnels and the Herodian and early Jewish remains’, excavated after Jerusalem was captured by Israel in the Six-Day War of 1967, ‘became the bedrock of Israel national existence’ (Silberman 2001, 500).

By firmly anchoring the self into the material world, states are inoculated from existential anxieties brought about by the prospect of rapid and thorough change. However, regardless of discursive efforts, no ontic space can fully protect state identities from unavoidable tides of change. Collective identities are in a continuous state of social negotiation and flux. Material environments, built and natural alike, undergo change too. As a consequence, ontic spaces brought into being through discursive linkage between the self and the material environment are inherently unstable and in need of continuous monitoring, maintenance, repair or reinvention.

Within this process, states bear agency but operate with identity discourses over which they never achieve full control. State representatives will strive to achieve a coherent, consistent and healthy sense of their state’s self. However, the vision of the self that they are trying to produce and maintain is never entirely under their control. State identity often faces domestic or international contestation, especially in times of crisis or transition. Individuals need a coherent sense of the state self because they are ‘emotionally connected to nation-state’ (Steele 2008, 16). To use Vamik Volkan’s term, collective identities are large ‘canvas tents’ that emotionally bind individuals into a group. In times of collective anxiety, ‘the members become preoccupied with repairing and mending the tears in the canvas of the large-group tent’ (Volkan 1997, 28).

Societal discourses are influenced by state actions, and they often subscribe to narratives advanced by state elites. However, if the state is unable to maintain a coherent narrative of the self, society will try to repair the story in order to protect ‘ideational stability’ (Marlow 2002). In the absence of state action to repair the fractured sense of the self, given the plurality of voices in every society, this societal ontological self-help can result in further contestation, fragmentation and even more anxiety.

In order to fend off these anxieties, state representatives resort to defensive measures such as adopting new practices, inventing new narratives (Steele 2008, 48) or employing avoidance mechanisms (Lupovici 2012, 818). In addition to these defensive mechanisms well known in IR literature, I argue that states may also resort to reappropriation of their material environments. They do this in order to fend off anxieties stemming from ideational destabilization but also to retain their agency, both domestically and internationally.

To sum up, I argue that by understanding the role of material environments and how they are turned into ontic spaces we are better equipped to understand ontological security processes in world politics in all their complexity. Most importantly, these insights should help us make more sense of particular empirical puzzles, to which I turn in the next section.

‘The centre of the state’: the General Staff HQ in Belgrade

In this section I use the case of the General Staff HQ in Belgrade to illustrate the theoretical claims developed above. As I demonstrate below, socialist Yugoslavia turned this building into an ontic space by projecting its identity narrative onto this landmark building. In 1999, NATO bombed the building in a war that marked the final phase of Yugoslav disintegration. Since then, the building has lingered in a state of dereliction, as the two options—to reconstruct it and to remove it—have equally threatened Serbia’s identity. The state had initially avoided dealing with the building because it was unable to make up its mind which story of the self to tell. Nevertheless, society spontaneously started introjecting the ruin into a collective identity narrative as a monument to NATO aggression. In order to repair the fractured sense of the self, in recent years the state has incorporated the ruin into the project of the self and reappropriated this ontic space.

Projection of Yugoslav identity into the General Staff HQ

For the past 200 years, Belgrade has been the capital of Serbia, and for most of the twentieth century it was also the capital of Yugoslavia. Its administrative centre is located at the intersection of Nemanjina and Kneza Miloša Streets, where both the central government and its key ministries are seated. This site was strategically chosen by the Serbian ruler Prince Miloš Obrenović in the early nineteenth century for its proximity to the rivers Sava and Danube, but also because it was sufficiently distant from the reach of Turkish cannons and provided easy access to Prince’s Palace in Topčider (Kovačević 2001). In the mid nineteenth century, the Artillery School building was also built on this site, opposite the seat of the government, only to be destroyed in the Second World War. In terms of its historical and symbolic importance, this cityscape is second to none, and, in the words of architect Bojan Kovačević, it represents ‘the centre of the state of Serbia’.⁴

In 1954, the Yugoslav military invited nine prominent national architects to participate in a competition to design the new military headquarters to be located in this capitol complex of Yugoslavia. The winner was Nikola Dobrović, whose project of a modernist V-shaped building encapsulated best how the Yugoslav state and its military wanted to be represented both domestically and abroad.⁵ This particular design was selected because it was starkly different both from the surrounding neo-classical pre-war buildings (symbolizing the bourgeois ancien régime) and from the military buildings popular in the Soviet Union at the time (Weiss 2000, 162; Davenport 2015). This was particularly important for a state that politically broke away from Moscow in 1948, fashioned a unique style of self-managing socialism and even signed a military alliance with two NATO member states, Greece and Turkey, in 1953.

The project proposed by Dobrović struck another important chord. The V shape of the building was an artistic projection of the Sutjeska river gorge where one of the major battles of World War II was fought in 1943 (Bogunović 2000). In the battle, communist guerrillas led by Marshall Josip Broz Tito (later Yugoslav president), vastly outnumbered by the opposing Axis troops, put up fierce resistance in spite of suffering heavy losses. After the war, the Battle of Sutjeska became a central part of Yugoslav war mythology. It was memorialized in the most expensive movie ever made in Yugoslavia, featuring Richard Burton as Tito (1973), while the Tjentište War Memorial became one of the key World War II memorial sites (Perica 2004, 97). According to Dobrović, who fought as a partisan as well, the Generals Staff HQ was meant to represent features of a defiant and brave nation. In his own words, ‘the builder broke off a piece of the mountains in which the fiercest and the most decisive struggle for the fate of the peoples of Yugoslavia was led, and he moved them to the centre of the capital’ (Kulić 2009b, 264).

The building was completed in 1963 and for more than three decades not only housed the military HQ but also represented the new identity of socialist Yugoslavia. It served as a screen onto which the state projected a landscape—the Sutjeska canyon—in which a defiant new multi-ethnic nation was forged during the formative battle against the Nazi occupiers. The state also projected onto the building the peculiar international identity of socialist Yugoslavia, positioned between the capitalist West and the Soviet-led Eastern bloc. For almost three decades, the building played a role of ontic space (one among many) that helped the Yugoslav state maintain its sense of biographical continuity.

Socialist Yugoslavia, however, broke apart in 1991, and the General Staff building continued to be the military HQ of the ‘rump Yugoslavia’ (Serbia and Montenegro) led by Serbian president Slobodan Milošević. The Yugoslav Army, based in this building, was actively implicated in a series of armed conflicts in Slovenia, Croatia, Bosnia and Kosovo. The communist and multi-ethnic symbolism of the building, however, was increasingly at odds with the rising nationalism and historic revisionism of the 1990s. After the fall of communism and implosion of Yugoslavia, as Kulić observes, ‘the building and its architect were too important to be dismissed, but associations to the previous system were no longer welcome’ (Kulić 2009a, 10).

Introjection of the ruined General Staff HQ into Serbia’s identity

In 1999, the fate of the nation once again became inextricably connected to the fate of this landmark building. From 24 March to 10 June 1999, NATO conducted the Allied Force air campaign against the regime of Slobodan Milošević in order to protect Kosovo Albanian civilians from the onslaught by Yugoslav and Serbian security forces.⁶ In addition to tactical targets, such as military facilities, forces and vehicles, NATO also bombed numerous strategic targets, such as command and control facilities, military and police headquarters, government ministries, petroleum refineries, etc. (Robertson 2015, 13–14). These strategic targets, some of which were highly controversial, included the General Staff building, portrayed by the Western media as ‘the heart of President Slobodan Milosevic military machine’ (BBC 1999).

The building was bombed twice, first on 30 April and again on 8 May 1999. On both occasions the building was empty and the only casualties were firefighters and journalists who rushed to the scene between the two air strikes. According to estimates made by Serbia’s Ministry of Defence, structure A (on the left side of Figure 1) was destroyed beyond repair. Structure B (on the right side of Figure 1) was slightly damaged and could be fully reconstructed, except for the entrance, which had to be cleared away.⁷ The bombing of this historic cityscape was a particularly traumatic moment for the city and for the nation. In the words of Bojan Kovačević, ‘the symbol of Yugoslav and Serbian statehood and the army was attacked, the city of Belgrade and its physical and mental image were seriously wounded’ (Kovačević 1999).

Figure 1. The ruined General Staff building. Source: Author.

After 11 weeks of NATO air strikes, Yugoslav forces withdrew from Kosovo in June 1999, leaving the province to be supervised by the international community. Serbia, at the time still under the authoritarian rule of Slobodan Milošević, embarked immediately on the reconstruction of the country (G17 1999). The reconstruction continued at an even faster pace after the fall of Milošević and with the onset of Serbia’s democratic transition in October 2000. By 2015, most buildings destroyed in the NATO intervention had been successfully reconstructed. For example, 57 out of 58 bridges throughout Serbia have been repaired or rebuilt (Vlajić 2014). In Belgrade, too, most buildings destroyed during the intervention were either repaired or reconstructed, including the iconic Avala TV Tower, the Ušće Tower, Dragiša Mišović Hospital and Hotel Yugoslavia. However, a few buildings in Belgrade that were hit in 1999 have remained completely or partially unrepaired. The General Staff HQ is one of them. For the past 17 years, the building has been in a state of dereliction.⁸

From Coventry, Warsaw and Berlin to Hiroshima, Nagasaki and New Orleans, the resilience and recovery of destroyed cityscapes has been a nearly universal phenomenon (Vale and Campanella 2005, 3). Throughout history, public authorities have rarely decided to leave the ruins of war intact, as reconstruction is part of the healing process. ‘Architectural scars, in the form of rubble, demolished cities or symbolic cities’, as Brent Steele has pointed out, ‘will most likely be reconstructed, if not into memorials, then into other buildings’ (Steele 2013, 143). Sometimes, in the aftermath of destruction, proposals do emerge to leave the ruins of war intact as grim reminders of history. However, the war ruins are usually left to ‘linger in a state of limbo’ only when they are not considered valuable enough to merit investment (DeSilvey and Edensor 2013, 473). How can the continued presence of the ruined General Staff HQ in Belgrade, a site so central both politicaly and historially, be accounted for?

According to conventional wisdom in Serbia, the General Staff HQ remains in ruins due to a combination of economic and bureaucratic reasons. The building has indeed been listed as a cultural monument (since 2005) requiring full reconstruction, which has deterred potential investors.⁹ This can hardly explain why the government hasn’t, after such a long time, either reconstructed the ruin or removed it from the list of cultural monuments and sold the site to private investors. Since the democratic transition that started in October 2000, excessive executive power, weak democratic institutions and strong charismatic leaders have plagued Serbia’s politics (Subotić forthcoming). There is little doubt that bureaucratic obstacles would have been removed had the fate of the General Staff building only been subjected to an economic cost–benefit analysis. It is a top-notch property worth up to €100 million which has been decaying for 17 years in clear sight of the prime minister’s office (Mučibabić and Vukasović 2013).

The continued dereliction of this war ruin seems to be a symptom of something deeper that remains out of reach of the bureaucratic–economic narrative. I argue that the state has left the ruin intact because it has been introjected into the state identity narrative. As a repository of Serbia’s collective memories of defiance in the face of foreign aggression, the ruined General Staff serves as an important source of ontological security for the Serbian state. This process of turning the ruin into an ontic space by introjecting it into the state identity narrative, however, hasn’t solely been the result of state agency. To understand how this process unfolded, a discussion about domestic debates on what to do with the ruined building is in order.

Ever since the bombing in 1999, the public debate about the General Staff HQ has revolved around three positions: restoration, removal and preservation. The first option, to restore the building to its former shape, has been advocated by Serbian architects, urban planners and World War II veterans. They have based this proposal on two rationales. First, the site of the building, they argue, is too historically important to be sold to investors. For example, on one occasion the veterans and retired Yugoslav military officials demanded of the government that the building be preserved as a monument and a ‘reminder of heroism of more than 7,000 fighters who lost their lives in the Battle of Sutjeska and 305,000 fighters who died for the victory and freedom in the National Liberation War’ (Danas 2013). The second rationale for restoration is the preservation of the building’s unique architectural design, universally acclaimed among experts as a masterpiece of Yugoslav modernism (Kovačević 2001).

The proposals to restore the original look of the General Staff HQ have raised emotionally disturbing questions about self-identity and collective memory. Historical revisionism has been part and parcel of the nation-building project in Serbia (and across the post-Yugoslav space) since the early 1990s. At its core, the process has entailed a negative reinterpretation of the partisan struggle during World War II as well as a general repudiation of the Yugoslav socialist project. Consequently, the prospect of restoring the General Staff HQ, which is a quintessential symbol of the partisan struggle and communist Yugoslavia, has threatened to reinstate an embodiment of an abjected Yugoslav and communist other. It is true that post-communist interpretations of the General Staff building, as Vladimir Kulić remarks, have tried to emphasize ‘ideologically neutral aspects of the building’ while ‘ignoring or openly denying others’, such as the Sutjeska gorge analogy for example. The purging of the communist past from the building, however, has not been entirely successful (Kulić 2009a, 10). Against such a backdrop, it becomes clear why the option of restoring this symbol of partisan resistance and Yugoslav socialism has deeply disrupted Serbia’s narrative of the self. Consequently, the proposals to restore the building have so far had little effect.¹⁰

Others in the government, including officials in the Ministry of Defence, have proposed to remove the ruin, for three specific reasons. First, as they argue, the ruin is unstable and on the verge of collapse, especially structure A, currently protected by scaffolding. Second, removal of the ruin is believed to be economically more cost efficient than restoration, while the restored edifice with its communist symbolism and reduced functionality is less attractive to potential investors. Finally, the advocates of removal have also strongly argued against restoring the building and putting it back to use as a military HQ. According to a former defence minister of Serbia: ‘Nowhere in the world is the military located in the city centre … the building was overly ambitious even for communist Yugoslavia, let alone for Serbia’.¹¹

From the point of view of those who advocate removal, the only obstacle to selling the building is its status as a cultural monument, requiring the current and future owners to keep the original exterior design (Mučibabić and Vukasović 2013). Under such conditions, they reckon, no investor will seriously consider buying the building. Advocates of removal have therefore proposed the deletion of the building from the list of cultural monuments. This would allow future investors to tear down the old building and erect a completely new one, potentially twice as big and commercially much more profitable.

The proposals to remove the ruin have impinged on Serbia’s self-identity, probably even more so than the proposals to restore it. In fact, the calls for the removal of the General Staff HQ have been interpreted by many as an attempt to erase what has become the site that narrates Serbia’s ‘truth’ about NATO bombing being an illegitimate act of aggression. For example, when US real estate developer and later presidential candidate Donald Trump expressed an interest in building a hotel on the site of the General Staff HQ, a member of the Serbian Parliament, Zaharije Trnavčević, fiercely opposed this idea on grounds that the building represents ‘a monument to an unprecedented crime’ (B92 2014). Responding to accusations that the government was planning to remove traces of NATO misdeeds, Serbia’s Prime Minister Aleksandar Vučić stated that ‘it goes without saying that we should build a monument to all the victims of NATO aggression, to all the soldiers who were killed, but I don’t see any point in this [building] looking the way it does’ (B92 2013b).

Consequently, all proposals to remove the ruin have been unsuccessful. For example, Serbia’s Ministry of Defence, the de facto occupant of the building (the owner is the Serbian government), has been advocating the sale of the site to an investor who would, in return, construct the HQ of the Ministry of Defence and General Staff elsewhere, preferably outside the city centre. The Serbian Property Directorate took a similar position and initiated the cancelling of the building’s status as a cultural monument in 2006, but to no avail. In 2013, the Ministry of Construction and Urban Development again initiated the removal of the building from the list of cultural monuments and suggested its clearance because of safety issues (B92 2013a). Just as before, however, this initiative failed, as it met with strong opposition inside government institutions, as well as among experts and the general public.

The third position in the debate on what to do with the General Staff HQ has been to preserve the ruin and leave it intact as a de facto monument. Interestingly enough, government representatives have never openly advocated this option. For example, a former defence minister claims that ‘nobody seriously considered that option because it would be completely inconsistent to have such a monument and at the same time be a NATO partner’.¹² Indeed, all Serbian governments since 2000 have pursued, more or less enthusiastically, a pro-Western foreign policy. To officially turn the ruined General Staff HQ into a monument would definitely not square well with the proclaimed goals of EU membership and NATO partnership.¹³ This would undermine Serbia’s future self-projection, something that is as essential for grounding the self as memories of the past (D’Argembeau et al. 2012).

All three options that have been suggested in the debate have clearly impinged on the ontological security of the Serbian state. This has created a situation that Amir Lupovici calls ‘ontological dissonance’. This is a situation where not only are two different identities of a collective agent threatened, but also a ‘clash stems from the perception that the measures required to placate the ontological insecurity of each of the threatened identities are themselves in conflict’ (Lupovici 2012, 810). As the Serbian state has been unable to change one of the threatened identities in such a short time span, its representatives have first opted for avoidance, which has allowed them ‘to separate the threatened self from the source of the threat and secure the boundaries of the self’ (Lupovici 2012, 818).

In spite of state inaction, the ruin has nevertheless been craving an unambigous meaning. Its destruction simultaneously purified it from its unwanted communist aura all the while fully preserving, even enhancing, the potential of its symbolism of national defiance and victimhood. In order to repair the fractured identity script, society spontaneously started to imbue the site with meaning. Over the years, the ruined General Staff HQ has thus spontaneously grown into an inseparable part of the Belgrade cityscape and a de facto national monument of defiance and victimhood (Bobić 2012).

The spontaneous monumentalization of the ruined site started with everyday practices. Due to its central location, many, possibly hundreds of thousands of, people pass by the General Staff ruin every day. Anyone arriving at the main Belgrade train or bus stations will also come across the building on their way up Nemanjina Street. Foreign tourists visiting Belgrade have inevitably stumbled upon the buildings in Kneza Miloša Street on their way from and to the airport. In the first few years after the 1999 air campaign most tourist guides to Belgrade included a section on the must-see ruin from NATO bombing in Kneza Miloša Street, while sightseeing tours included a visit to the ruined military HQ.¹⁴

In the absence of state action, the ruin was imbued with meaning by society itself. It was the citizens, as Davenport has pointed out, who ‘found in the persistence of the ruin a symbol of the resistant qualities which it once embodied, with the former allies of the First and Second World Wars now seen as the aggressors in the form of the perpetrators of the NATO bombing’ (Davenport 2015, 181). The term ‘citizens’ here refers not only to Belgradians but also to visitors and representatives of all major national institutions that are headquartered in the near vicinity of the site. Moreover, given the fact that the debate about the ruin often featured in the nationwide media, the General Staff HQ has spontaneously become a national lieu de mémoire.

As years went by and the everyday introjection of the site consolidated, elites started to tune in to the process. The first proposals to keep the status quo and turn the building into an official monument of NATO intervention, which is what the building came to mean at the level of everyday practice, started to appear in public discourse in 2014. For example, the Patriarch of the Serbian Orthodox Church, Irinej, argued in May 2014,

If someone would ask me, I would never repair the ruins in the centre of Belgrade … Taking into consideration that the Republic of Serbia still doesn’t have a unique museum or monument dedicated to the victims of NATO aggression, we believe that the buildings on the current location should be reconstructed to serve that purpose. (Irinej 2014)

From then on, the idea of leaving the ruin as it is began to quickly spread across the public sphere. For example, Zaharije Trnavčević, a member of the National Assembly, proposed that the building should be preserved in ruined form ‘as a memorial, a memento of the unlawful bombing of Belgrade and Serbia … we have a duty to preserve the memory of the illegal and unpunished behaviour of America and great Western powers that make up the core of NATO’ (Trnavčević 2013). Soon after, right-wing organizations started to advocate a similar idea. In a protest staged in front of the building, a member of one of these groups argued that, like the ruin of Pavlov’s House, kept to preserve the memory of the Battle of Stalingrad, the preservation of the ‘Military Headquarters is not a question of economics or aesthetics, but of history and ethics. These are not ruins but scars. This is not a heap of stones, but a living man who speaks and warns: never to forget, never to repeat!’ (Srpski kod 2014).¹⁵

As the introjection of the site solidified at the societal level, state agency was also drawn into the process. By 2015, the status of the building as a spontaneous monument of NATO aggression (and Serbia’s innocence) was indirectly endorsed by the state. On the occasion of the 16th anniversary of the beginning of NATO intervention on March 24, the Government of Serbia organized a ceremony in front of the ruined building. For that purpose, the site was transformed into a theatrical scene illuminated with a lightshow that spotlighted impact craters on the building in order to expose the wounds and scars of foreign aggression.¹⁶ The ceremony was broadcast live on national TV and attended by top-level Serbian civilian and military leaders and several thousand citizens holding lit candles.

The show began with a replay of the characteristic soundscapes evoking NATO intervention: air raid sirens, bombs thundering and buildings crumbling. The programme continued with the rendering of the Serbian national anthem by a young girl, subtly evoking the death of Milica Rakić, the three-year-old killed by NATO bombs while sitting on a potty, symbolizing all the innocent victims killed by NATO. The subsequent act was devoted to the memory of the bombing victims, represented on stage by a policeman, a soldier and a girl holding a teddy bear. The culmination of the ceremony was the address by Prime Minister Vučić, in which he reiterated a long list of NATO crimes against Serbia and vowed, ‘We remember, and everybody else should also remember, we Serbs have a long memory and will never forget. For each of the 78 days of bombing, each of the victims will be remembered’ (B92 2015).

With this, the introjection of the ruin into the dominant national discourse about the heroic, defiant, victimized and righteous self was complete. The state thus reappropriated the ontic space in order to fend off societal anxieties about ‘forgetting the past’ and ‘losing the self’. However, this reappropriation hasn’t removed the underlying tension between the vision of the self as a victim of NATO, on the one hand, and a Western state, on the other. The new ontic space constructed on the site of the ruined building mitigated present anxieties about breaking the link with the past but hampered the ability of Serbia to sustain its future self-projection.

Conclusion

Across social studies, ontological security has been understood as a need of human beings to have confidence in the constancy of their social and material environments of action. Within IR, a vast majority of studies have focused on social environment, i.e. states’ relationships with their significant others. Regardless of their stance within central debates in ontological security theory on the sources and agents of ontological security, IR scholars have so far mostly neglected the role of material environments such as familiar natural landscapes or symbolically important cityscapes. In order to fill this gap, I have developed ideas on how to think about the role of natural or built spaces with respect to the ontological security of states.

In particular, I have proposed that material environments serve as an important source of ontological security not only for individuals but also for states. In the face of the transience of social relationships, states need an additional anchor for their collective identity narratives. By mooring their identity to material environments, states secure their sense of biographical continuity and fend off anxieties stemming from the prospect of a divided and fractured self. However, the material environment doesn’t play this role in and of itself. In order to assume this role, material environments needs to be discursively linked to projects of the self and this can be accomplished either through introjection or through projection. Once incorporated into the narrative of the self, material environments become ontic spaces. While state representatives hold some agency in the process, they certainly don’t operate in a vacuum but rather operate within identity discourses over which they never achieve full control.

I have illustrated my theoretical claims in a case study of the General Staff building in Belgrade destroyed by NATO in 1999. During the Cold War, the Yugoslav state projected its identity onto the building. After the implosion of Yugoslavia and the downfall of the Milošević regime, the Serbian state avoided making any decisions about the ruined building for such a long time because all the options proposed in the public debate impinged on one of the state identity narratives. In the meantime, the ruined site has been gradually introjected into the national identity discourse through everyday practices and spontaneously turned into a monument of resistance to foreign aggression. Although the initial impulse to introject the ruined building came from within society, which was moved by the unchallenged force of pre-existing discourse about defiance and victimhood, the Serbian state was eventually sucked into the process as well.

The ruined building, however, continues to expose tensions between various state identity narratives of Serbia. As a result, the state keeps repeating its promises to do something about the building while the anxiety about all three proposed options, coupled with epiphenomenal bureaucratic or economic obstacles, continues to paralyse any practical steps in that direction. If and when the state finally decides what to do with the building it will have to openly deal with questions about identity that normally need to be bracketed out if existential anxiety is to be kept at bay.

The theoretical ideas developed in this article shed a new light on the psychological and emotional roots of the seemingly irrational attachment of nation-states to particular natural or urban landscapes that are often highly contested. These insights also analytically articulate a previously unexplored theoretical vantage point for the study of protracted conflicts over ontic spaces which leave little room for negotiation and compromise in particular. Furthermore, the enlargement of ontological security theory to also include material environments of state action promises new avenues for cross-disciplinary exchange with fields previously little explored in IR such as architecture, geography, archaeology and heritage studies.

Future studies could further expand our understanding of the role of the material environment in ontological security processes in world politics in a number of different ways. First, while in this article I have empirically looked at the built environment, prospective research could expand the analysis to also include natural landscapes. Here, a great potential lies for cross-fertilization with an existing body of sociological research interested in the interplay between landscape and collective identity. A second possibly rewarding avenue for further research is to examine more closely the process of linking and de-linking of material environments with collective identities. Particularly interesting would be to investigate the role of different state and non-state actors in the process and strategies they deploy. Finally, this paper has been biased in favour of human agency in ontological (in)security, and the way that human beings make the non-human environment, such as war ruins, meaningful. It would not be off the mark to ask the reverse and investigate how human beings and their sense of continuity in events are ordered by non-human agency, including urban or natural landscapes. By decentring the human in favour of non-human agency, ontological security scholarship could make a contribution to the ongoing ‘non-human turn’ in the humanities.

Disclosure statement

No potential conflict of interest was reported by the author.

Funding

This article has been written as part of my work on the project ‘Political Identity of Serbia in Regional and Global Context’, No. 179076 at the University of Belgrade, Faculty of Political Sciences, financed by the Ministry of Education, Science and Technological Development of the Republic of Serbia.

Notes on contributor

Filip Ejdus is a Marie Curie scholar at the School of Sociology, Politics and International Studies, University of Bristol and assistant professor at the Faculty of Political Sciences, University of Belgrade.

Notes

I thank Jelena Subotić, Iver Neumann, Brent Steele and three anonymous reviewers for their helpful comments. An earlier draft of this paper was presented at the International Studies Association Annual Conference, New Orleans, 2015.

¹ The unit of analysis debate is sometimes confused with a closely related but distinct agency/structure debate which revolves around the question of the source of ontological (in)security (Zarakol 2010).

² To my best knowledge, the only exception to this is the work of Catarina Kinnvall, who briefly discusses the role of home for ontological security in IR (2004, 747).

³ Seven plants mentioned in the Bible are palm dates, wheat, barley, grapes, figs, pomegranates and olives.

⁴ Personal interview, 21 January 2015.

⁵ Although there are technically two structures—A and B—they conceptually make a single building and will therefore in this article be referred to in the singular.

⁶ In 1999 Serbia was one of the two republics (with Montenegro) within the Federal Republic of Yugoslavia (FRY). In Kosovo, Serbia’s police forces were fighting together with Yugoslav military forces against the Kosovo Liberation Army. The FRY ceased to exist in 2006 when Montenegro decided to leave the federation in a referendum, thus making Serbia an independent state.

⁷ Personal interview with a representative of the Ministry of Defence, 13 January 2015.

⁸ Only the tower, which belongs to structure B, has been reoccupied by the military.

⁹ The designation of a ‘cultural monument’, whatever the final fate of the building and whoever its future owner might be, guarantees that its exterior façade has to be preserved in its original form (Republic of Serbia 2005).

¹⁰ The only success of this group was an initiative launched by Serbian architects and urban planners to put the building on the list of cultural properties under the designation ‘cultural monument’ in 2005.

¹¹ Personal interview, 19 January 2015.

¹² Personal interview, 16 January 2015.

¹³ Serbia became a candidate for EU membership in 2012 and began accession negotiations in 2014. It joined NATO’s Partnership for Peace programme in December 2006 but declared military neutrality one year later in a parliamentary resolution. As the reason for declaring military neutrality, the resolution cites ‘the overall role of NATO’ from the 1999 bombing to the building of an independent Kosovo (Republic of Serbia 2007).

¹⁴ In the words of the Vice President of the Association of Tourist Guides of Serbia, ‘all panoramic tours of Belgrade include a visit to the ruined Military Headquarters’. Personal interview, 20 January 2015.

¹⁵ The Pavlov House was named after Red Army Sergeant Yakov Pavlov, who commanded a platoon that used the building as a fortification during the Battle of Stalingrad in 1942.

¹⁶ The full video of the ceremony can be seen here: https://www.youtube.com/watch?v=VjhX9UN1GzU