Separatists,state subjectivity,and fundamental ontological (in)security in international relations

Introduction

The thriving literature on ontological security in International Relations (IR) emphasizes the role of identity, anxiety, and the need for a stable sense of self in world politics. ¹ It suggests that ‘states perform actions in order to underwrite their notions of “who they are”’ ² as particular states, based on their ‘auto-biographies’ ³ and relations with other states. ⁴ At the same time, if states act in certain ways because of ‘who they are’, this also indicates, if only in passing, that they are in fact states. However, no matter how useful and justified for many analytical tasks, assuming states as the starting point renders the process of their very constitution as states invisible, thereby neglecting the anxiety over state subjectivity itself and the role played by third states in making, or rejecting, state ‘persons’ in the first place.

As I argue in this article, the fundamental ontological security provided by state subjectivity becomes visible and acutely relevant where states are openly contested, as is notably the case in separatist conflicts and especially in conflicts between unrecognized de facto states and their unyielding parent states. ⁵ Particular states cannot be presupposed in these conflicts because determining which of the overlapping state projects constitutes a state is precisely what is at stake in them. Indeed, statehood in the modern international order is not simply a special status that given political entities seek to achieve, but rather an exclusive type of subjectivity that constructs a particular community and territory as a corporate person and delineated space, at the expense of all others. That is, any contenders for statehood which are not regarded as states find themselves inevitably on the territory and under the sovereignty of another state, which undermines their own claim to statehood and political independence.

This not only means that states cannot be analytically presupposed in separatist conflicts, but also that they pose a meta-security dilemma for third parties trying to resolve them in practice, whether by mediation, legal determination, or recognition. While the support for and recognition of one of the contenders fundamentally reassures it in its supposed statehood, it at the same time renders the other one fundamentally insecure. Considering ontological security in separatist conflicts therefore sheds light on a basic dilemma of some of the most entrenched identity conflicts, while also contributing to a reflexive and relational understanding of state subjectivity in Ontological Security Studies (OSS). It more particularly suggests that state subjectivity is not simply the starting point of quests for ontological security, but in some sense already their result. As such, fundamental ontological security is not provided to given state subjects but generated by constituting them, thereby reassuring not only those who identify with these particular states – at the expense of those who do not – but also third states which seek to preserve the state-based international order as their own environment.

In order to problematize state subjectivity in the context of separatist conflicts and to analyze the underlying dimension of fundamental ontological (in)security linked to state status, the article is subdivided into four main sections. In section 2, I revisit debates on state personhood in OSS and introduce fundamental ontological (in)security as a reflexive and relational approach to state subjectivity. In section 3, I explore the position of de facto states in separatist conflicts to illustrate their fundamental physical and ontological insecurity, while in section 4 I focus on the corresponding fears of challenged parent states. In section 5, I discuss the meta-security dilemma of state identification with which third parties are confronted in separatist conflicts, emphasizing the role that the recognition of states plays in constituting them as subjects of security. The conclusion discusses implications for OSS and IR generally and suggests possible inroads for future research.

State subjectivity and fundamental ontological (in)security

OSS illuminates the role which the anxiety and self-identity of a variety of subjects play in world politics, from individuals through minorities and indigenous peoples to supranational organizations such as the European Union (EU). ⁶ As with most approaches in IR, however, the main protagonists are states interacting with other states. And as demonstrated by debates in poststructuralist and constructivist IR more generally, ⁷ questions of state identity and subjectivity raise thorny issues. Indeed, OSS oscillates uncomfortably between the critical agenda of interrogating and de-naturalizing seemingly essential identities and the pragmatic agenda of empirically investigating the causes and effects of identity formation, even if this requires presupposing states as the very subjects whose ontological (in)security is in question. Yet, what is presupposed cannot at the same time be interrogated. ⁸

This tension reflects to some degree the opposition between substantialist and relational approaches. While the former stipulate the existence of entities preceding their specific internal qualities and external relations, the latter instead emphasize the emergence of entities as the very product of such relations. ⁹ If OSS has been shaped by relationism early on, it has also often made use of a functional substantialism, especially in IR and when it comes to states. This might not be immediately apparent. Indeed, the central concern with a need for existential security, which is achieved through routine practices and cognitive certainty about the social environment and one’s own role in it, ¹⁰ suggests a relational understanding of identity. Identities are not there to begin with but only take shape and transform – or resist transformation – in response to anxiety and social interactions. If identities appear stable, then this is so as a result, rather than as a presupposition, of social and political processes. ¹¹ As McSweeney has highlighted early on and as has been reiterated in OSS ever since, the identities of states are unachieved ‘projects’ shaped by domestic and international processes and as such they are neither given nor necessarily unchanging. ¹²

However, presupposing particular states as the subjects, persons, or selves whose identities become shaped by domestic processes, ¹³ international relations, ¹⁴ or both, ¹⁵ nevertheless already credits these states with a material and mental existence that is itself removed from further scrutiny. ¹⁶ Particular states, it seems, are there to begin with as the basic actors or subjects of international politics, which is indeed the ‘most common (. . .) substantialist starting point (. . .) in IR’. ¹⁷ To be sure, taking states as corporate actors is everyday business not only in IR research but also in diplomatic and legal practice. ¹⁸ As therefore pointed out by Mitzen, ‘the “state as person” (. . .) indexes real aspects of the ways in which states operate in world politics’. ¹⁹ Moreover, presupposing concrete states is not problematic per se – it is obviously suitable for a variety of analytical and practical tasks. Yet, the important blind spot is that by assuming that particular entities are states, and others are not, we cannot grasp how states emerge as subjects and spaces of (in)security in the first place, how they are continuously ‘yoked’ ²⁰ into existence by being assumed to be there, and how others are thus excluded from this status and typically find themselves enclosed within other states. ²¹ That is, the functional substantialist assumption that we always already know the concrete states whose physical and ontological security is at stake obscures precisely the implications of assuming particular states in practice.

This becomes visibly problematic where individual states are contested, overlap, and compete, as is notably the case in separatist conflicts. The most illustrative examples in this regard are conflicts between recognized states and supposed de facto states which control a stretch of their parent states’ formally recognized territory. ²² In these situations, we can neither simply assume that there is a conflict between two distinct states, nor that a state is opposed by a rebel group within its territory, since the conflict is precisely about its categorization as one or the other. Regarding states to be properly there, or not, is therefore not an appropriate starting point for the analysis.

Although cases of contested statehood pose a problem to any IR approach that starts with the traditional assumption of states as essentially there, they present particular challenges – and opportunities – to the study of state subjectivity in OSS. Indeed, if the state character of the opponents is indeterminate, we can build neither on the usual distinction between the primary selves of states and their contingent identities, ²³ nor on a clear division between physical and ontological security. ²⁴ The assumption of the contained self of states reflects the national-territorial form and protected legal status of sovereign states as corporate persons in modern international relations. ²⁵ Yet, if the territory, constituency, and authority of a state are contested, so is its ‘physical’ and ‘psychological’ personhood. Further, and as a consequence of the exclusive status of states as international persons, others cannot be states on the same territory and with the same population. This does not mean that armed groups and various other ‘non-state’ actors would have no distinct identity; but they do not take the form of a corporate person with a contiguous and exclusive territory, population, and authority, that is properties of full-fledged state subjects. In other words, state subjectivity is premised on its supremacy over other entities on its respective territory. ²⁶ The assumption of states as macro-subjects and spaces of (in)security thus also raises the stakes where claims to statehood overlap. In these cases, the contenders do not simply threaten each other’s physical security, while affirming their different identities, as Mitzen’s ontological security dilemma postulates for states caught in enduring rivalries. ²⁷ Instead, in separatist conflicts the contenders threaten their very personhood and existence as states.

However, if ‘identity is an idea in search for a person’, ²⁸ we might well account for conflicts over the constitution of states in terms of OSS. In fact, if we do not understand state subjects as the starting point of quests for ontological security, but as already their result, then taking the challenge of separatist conflicts seriously promises major insights for OSS. More specifically, my theoretical proposition is two-fold. First, apart from the ontological security obtained by expressing their particular national identity and their physical security tied to territory and population, ‘states’ desire the fundamental ontological security afforded to them by their recognition and perpetual confirmation as states. Moreover, state subjectivity not only sustains the political communities that are nominally organized as states, but also the state system as the everyday ‘international’ environment in which states – and others – routinely interact, in which they know ‘who they are’ ²⁹ and ‘how to get by’. ³⁰

The dark side of this security found in state subjectivity itself is the fundamental insecurity it entails for those who do not feel they belong within ‘their’ state, but who either do not organize themselves as states or do so but are denied international recognition. Where push comes to shove, clashing state projects are locked in a zero-sum game that goes deeper than a traditional security dilemma, ³¹ or any other conflict between states whose security would not depend on breaking apart or destroying other states.

Separatist conflicts share some features with domestic conflicts in which different parties with exclusive visions of national identity compete for power and the position of formal government, since in these conflicts too it might be that the winner takes all and the loser becomes dramatically insecure. ³² However, separatist conflicts are different as they lie precisely at the intersection of domestic and international conflicts. As such, they cannot be resolved ‘internally’, by reconciliation, power-sharing, or the like, at least not without the separatists stepping back from their demands for complete state personhood. Moreover, they in turn offer a unique solution unavailable in other domestic conflicts: the formal separation and recognition of distinct states. This would normalize the opponents’ relations as properly international relations. Yet, this option also reflects the key problem. As the fundamental ontological security of each party depends on its recognition as a state and the protection of its territorial integrity, it is incompatible with the same need of the other one, since their authority claims overlap. Fundamental ontological (in)security is thus deeply relational, not just by shaping national identities already enclosed in particular state subjects through the interactions with other states, but by constituting state subjects out of fragile state projects that emerge in opposition to other would-be states in the first place.

Second, this poses a meta-security dilemma for any third parties attempting to resolve the conflict on the premise of exclusive statehood, since their involvement must inevitably appear as partisan, if not as merely ineffective. Compromises can arrest and freeze confrontations, but they cannot properly resolve them, unless the state form itself is weakened. As I show in the next sections – by taking the respective perspectives of de facto states, parent states, and third states – the meta-security dilemma dramatically highlights the neglected but decisive role and implications of fundamental ontological (in)security in international relations.

Whose ontological (in)security? The precarious existence of de facto states

As expounded in studies of nationalism, the modern political imagination has been profoundly shaped by the nation as an abstract and romanticized notion of belonging in a world disenchanted and unbound. ³³ The underlying desire for renewed community has also been picked up in OSS to explain how the ‘search for one stable identity’ by individuals directs them to the nation as ‘this monolithic “entity” [which] becomes a stabilizing anchor in an otherwise chaotic and changing world’. ³⁴ Following the principle that the national and the political unit should be identical, ³⁵ the hallmark of national self-determination, and thus of modern collective ontological and physical security, is independent statehood. ³⁶

However, the history of national self-determination is also a history of disappointments, for not every ‘imagined community’ ³⁷ obtained the desired status of a state. The large-scale decolonization of former colonial territories in the mid-20th century, as well as the dissolutions of the Soviet and Yugoslav federations, created not only new sovereign states, but also a host of groups within these states which held national aspirations but were denied their own independence. These frustrated hopes have reverberated in armed struggles across the world, from the Kurdish movements in Turkey and Iraq through the secessionist wars of Biafra and South Sudan to the de facto states of the post-Soviet space, including Nagorno-Karabakh in Azerbaijan, Transnistria in Moldova, and Abkhazia and South Ossetia in Georgia. ³⁸ Unsurprisingly, some of the most prominent examples of entrenched identity conflicts discussed in OSS include Palestine, ³⁹ Cyprus, and the Republika Srpska in Bosnia and Herzegovina, ⁴⁰ all conflicts that are neither simply domestic nor international, but that revolve precisely around this distinction and the question of independence itself. As I argue in this section, for separatist groups with a measure of territorial control and in particular for full-fledged de facto states, the conflict over their status poses an existential threat that surpasses the usual insecurity of states in contemporary international relations. It notably also aligns their concerns for physical and ontological security in the unequivocal desire for their very recognition as states.

De facto states face a threat to their physical security from their parent states that exceeds the usual stakes of power politics between states. This is so for mainly two reasons. First, most de facto states have emerged in armed conflicts and fought for the territory they effectively control. From the Turkish Republic of Northern Cyprus to Nagorno-Karabakh, the informal borders between many de facto states and their parent states were established by ceasefire agreements that never transformed into actual peace. The militarized borderlands and deep mistrust between the two conflict parties therefore present ample opportunities for rapid escalations and renewed violence. ⁴¹ Second, with the lack of formal recognition as their ‘defining characteristic’, ⁴² de facto states do not benefit from the protection of sovereignty and territorial integrity granted to states by international law. ⁴³ Indeed, from a conventional perspective of international law, they are mere rebels on their parent states’ territory, and the formally recognized parent states can deal with them as a matter of their own domestic affairs, including, in principle, by the use of force. ⁴⁴ The exclusive state status reinforced by international law privileges the formally recognized parent states – and thus exposes de facto states to possible annihilation. The violent destructions of the Chechen Republic of Ichkeria in Russia, of the self-declared Republic of Serbian Krajina in Croatia, and of the rebel governance of the Tamil Tigers of Tamil Eelam (LTTE) in Sri Lanka vividly illustrate the existential threat posed to de facto states even after long periods of peace. ⁴⁵

A more recent case in point is the war between Azerbaijan and the Armenian-backed but unrecognized Republic of Nagorno-Karabakh that broke out in September 2020. A so-called ‘frozen conflict’, the struggle over Nagorno-Karabakh dates back to pre-Soviet times but largely reflects the unresolved issue of whether or not the territory would be part of Azerbaijan after the dissolution of the Soviet Union in 1991. The war between Armenia and Azerbaijan ended in 1994 with a ceasefire that left Nagorno-Karabakh formally on the territory of Azerbaijan but under its own – and Armenian-supported – de facto control. ⁴⁶ Yet, Azerbaijan could present its recent attempt to reconquer Nagorno-Karabakh by military force, and thereby re-establish control over its whole territory, as an internal affair.

There is thus, in stark contrast to recognized states, no international guarantee for the survival of de facto states. If ‘state death’ ⁴⁷ is said to have disappeared as a consequence of the ‘new sovereignty regime’ after World War II, ⁴⁸ this neglects the fate of de facto states since the ‘state death’ of rebel entities falls under the radar of ‘international’ relations. ⁴⁹ The threat of extinction thus raises the stakes for the physical security of de facto states to a level that is incomparable to threats posed to recognized states, for whom ‘traditional security concerns often no longer have primacy’, ⁵⁰ because their existence is virtually guaranteed. ⁵¹ That is, the competitions between, say, the US and China or between Pakistan and India, might concern their relative power, status, and wealth, but not their existence as states, or at least not short of a nuclear doomsday scenario. By contrast, from Nagorno-Karabakh to Taiwan, de facto states with powerful parent states that uphold their claim to the territory are permanently worried not merely to be attacked but to be extinguished.

The quest of would-be states for ontological security also takes a particular form. While scholars in OSS emphasize the importance of social recognition in shaping the identity of states generally, ⁵² de facto states struggle not only to have their specific national identity confirmed, but also to be recognized as states at all. For instance, during the dissolution of Yugoslavia, the break-away republics of Slovenia and Croatia, and with more difficulties Bosnia and Herzegovina, not only sought to express distinct national identities – by invoking different pasts, languages, religions, and ethnic compositions – but also called for their legal recognition as independent states. ⁵³ It is not enough to be a particular nation, one also has to have a state which is equal to that of other sovereign states in international relations.

Unsurprisingly, many de facto state leaders and their constituencies are heavily committed to displaying hallmarks of statehood, ⁵⁴ from issuing a separate currency through the establishment of diplomatic missions abroad to membership in international organizations, and be it the self-created ‘Unrepresented Nations and Peoples Organization’ (UNPO). Visa procedures, controls at border posts, and the emulation of diplomatic ceremonials are symbolically charged practices that illustrate these entities’ sense of ‘state self’ and desire for recognition. This even explains diplomatic relations with small island states at the other end of the world, as in the case of Abkhazia’s ties to Nauru and Vanuatu. ⁵⁵ Looking and behaving like a state is an expression of the projected self of de facto states, even where de facto states are not immediately threatened in their physical security, either because no acute enemy presents itself, as is largely the case for Somaliland, or because strong foreign supporters provide protection, as is the case for the Turkish Republic of Northern Cyprus, Abkhazia, and Transnistria. ⁵⁶ This points to the importance of recognition for the ontological, and not merely the physical, security of the aspiring state entities.

Thus, while the affirmation of distinct national identity through conflict, highlighted in OSS, ⁵⁷ also plays an important role in secessionist struggles, precarious de facto states further display a more fundamental anxiety over their very statehood. Where secessionists have established ‘physical’ features of statehood, such as a measure of territorial control, administrative infrastructure, and public services, their ontological insecurity is deepened by additional concerns over physical insecurity to the state’s ‘body’. Thus, in contrast to the usual separation of the two dimensions within the OSS literature, ⁵⁸ the physical and ontological insecurity of de facto states are not to be traded for one another but are inextricably tied up together. The external recognition as states promises security on both accounts, while the lack of recognition feeds into a fundamental anxiety over who one is in a world of states and whether one can survive as a state at all.

The loss of ‘limb’ and ‘soul’: Separatism and parent state anxiety

While separatists and de facto states are threatened due to their lack of international recognition, their challenged parent states are in turn threatened by the partition of their country. That is, what has been regarded as a path to self-determination by separatists from Québec, Scotland, and Catalonia to Somaliland, Kosovo, and Abkhazia, has at the same time posed a challenge to the identity as well as the physical integrity of their respective parent states, from Canada, the UK, and Spain to Somalia, Serbia, and Georgia. Even in the peaceful instances, contentious referendums illustrate that there is much at stake for the rest of the country, including for those who live in the contested territories but do not identify with the separatist cause. Where territorial control is at least temporarily lost to de facto states or permanently lost to newly recognized states, the threat becomes even more palpable, not merely to the state’s physical or ontological security, but to its very personhood.

Unilateral secession violates by definition the territorial integrity of the parent state and thus threatens its physical security. Indeed, if the ‘body’ of a state is imagined to be composed of its territory, its population, and its sovereignty, ⁵⁹ then secession implies at least a rupture and, where permanent, the loss of a limb. As discussed in the literature on partition in armed conflicts, the separation of territories usually involves the forced displacement of people and seems to engender as much violence and insecurity as it prevents by separating the conflict parties. ⁶⁰ There are, to be sure, different degrees of entanglement and thus of pain of separation. For instance, for Yugoslavia in 1991, the departure of Slovenia was rather smooth as compared to the parting of Croatia and Bosnia and Herzegovina because the latter were inhabited by sizable Serb minorities who opposed the independence of these new states or sought their own independence in turn. ⁶¹ Moreover, in some separatist conflicts, the stakes are higher than the loss of a particular part of the national territory and instead concern the very existence of the state. Yugoslavia and the Soviet Union are two prominent cases in point. The dissolution of these federations implied the disappearance of the union states themselves, whether or not their remaining core states were ultimately regarded as nominal successor states – as in the case of Russia – or not – as in the case of rump Yugoslavia and Serbia. Thus, secession and de facto state making present a more fundamental threat to the physical security of recognized states than do conventional security challenges between states that may vie for power but that rarely threaten each other’s existence. Recognized states are thus more radically contested by separatists from the – fragmenting – inside than by other states from the outside. ⁶²

The challenge of separatism to the ontological security of parent states also exceeds the usual dynamics of identity formation in international and domestic politics discussed in OSS. In international politics, established states usually seek to be regarded in a particular way that corresponds to their projected self-identity, whether this is informed by enmity, ⁶³ stigma, ⁶⁴ or domestic processes. ⁶⁵ For instance, the Cold-War rivalry between the US and the Soviet Union reassured both about their assumed self-identities. ⁶⁶ In domestic politics, different groups struggle over different versions of their shared national identity, or their inclusion or exclusion from the dominant national narrative. ⁶⁷ By contrast, separatists neither affirm the particular identity of their parent state from a distinct and delineated outside, nor do they challenge it from the common inside by suggesting a different articulation of the state’s collective identity. Instead, separatists challenge the state person itself by suggesting that there are multiple distinct persons with separate bodies and selves.

The persistence of Abkhazia and South Ossetia, Transnistria, the Turkish Republic of Northern Cyprus, and de facto independent Taiwan continuously remind their respective parent states of Georgia, Moldova, Cyprus, and the People’s Republic of China that their projected selves are contested and incomplete. For instance, from a Serbian perspective, the independence of Kosovo has been traumatic, undermining its traditional national narrative in which the region of Kosovo has played a crucial role. ⁶⁸ More radically, when the new ‘imagined states’ ⁶⁹ of Slovenia, Croatia, and Bosnia and Herzegovina emerged in the early 1990s, they dissolved the Yugoslav self that had united them in one person. ⁷⁰

The reactions of parent states to de facto state challenges tend to be correspondingly fierce. Where possible, territorial re-conquest appears to offer a lasting solution. For instance, Russia broke the de facto state in Chechnya in the second Chechen war beginning in 1999 and, with the assistance of local allies, re-integrated the province into the federation. This also sent the message that the dissolution of the Soviet Union would not be followed by that of Russia. ⁷¹ For Azerbaijan, the defeat in the 1992–1994 war against Armenia meant that it never established territorial control over Nagorno-Karabakh, which remained therefore a thorn in its flesh that stirred various border clashes over the last decades and ultimately led to the war in the fall of 2020, when Azerbaijan captured parts of the region in heavy fighting and under widespread national acclaim. ⁷² In Sri Lanka, following decades of civil war with the LTTE, the eventual defeat of the ‘rebels’ in 2009 was hailed as a liberation of Sri Lanka’s national territory. ⁷³ Yet, many other contested parent states, from Somalia to China, still endure the loss of territorial control to de facto states, perhaps permanently.

The emergence of new states, and its prevention, are not a question of force alone, however, and international legal status and diplomatic recognition in particular play a major role in making and securing states. Many states that are contested by separatists therefore go to great lengths to avert at least the international recognition of their secessionists, and even of secessionists in third countries. ⁷⁴ For instance, Spain and China did not recognize Kosovo’s declaration of independence so as to not encourage their own separatists in Catalonia and the Basque Country, and in Xinjiang, Tibet, and Taiwan, respectively. ⁷⁵ Similarly, the African Union (AU) fact-finding mission to Somaliland in 2005 warned that the recognition of the break-away republic would ‘open(. . .) a pandora’s box’ ⁷⁶ of separatist claims across the continent, and thus discouraged any third states from recognizing it. As a result, with the exception of decolonization and historical post-war settlements, the widespread recognition of new states against the will of parent states has remained very rare, even if de facto states are frequently recognized by isolated foreign patrons. ⁷⁷

In sum, separatist conflicts point to a deeper dimension of insecurity for recognized states than the physical and ontological insecurity they face in international and domestic relations because this fundamental insecurity stems from the fragility of state subjectivity itself. Separatists are a reminder that every state is ultimately a contingent construct upheld by constant confirmation. As in the case of de facto states, challenged parent states cannot easily afford to separate physical and ontological security concerns, since secession poses a threat to both. Although international convention largely protects recognized states, the loss of territory and the rebel claim to independent statehood undermine their own subjectivity, and even the smallest chance of the recognition of the separatists as a state provokes fundamental fears over the potential loss of limb and soul.

The meta-security dilemma of state identification

As competing state projects are locked in a zero-sum game, any third parties that become involved to resolve such conflicts face a meta-security dilemma. In addressing the security concerns of supposed de facto states – by acknowledging them as actors, granting them support or, in the extreme case, recognizing them as sovereign states – these entities can be reassured, but only at the cost of rendering their parent states fundamentally insecure. By contrast, denying the separatists recognition and fostering the authority of the parent state addresses the latter’s fears, but at the cost of feeding into the existential anxiety of the de facto state. Even from an impartial perspective committed to finding an equitable solution, the rival aspirations to statehood cannot be reconciled, that is, without undermining the projected sovereign statehood of one or the other. The tragic implications of the meta-security dilemma can be illustrated in efforts of political mediation and attempts to legally identify states.

When considering why conflicts over the creation of new states are in the last instance irreconcilable, it is first of all important to recall that these are extreme cases which are defined precisely by their radically opposed claims to state subjectivity itself. There are indeed various types of conflicts over status and identity within states – from competing narratives of the country’s national identity through the discrimination of minorities and indigenous peoples to civil wars and rebel governance – which do not necessarily take the form of opposing state projects. In these cases, conceivable compromise solutions include a national dialog, enhanced cultural rights, quotas and power-sharing agreements, limited autonomy, or full-fledged federalism, among others. These might serve as strategies of de-securitization by relaxing rival identities and blurring emerging notions of exclusive selves. ⁷⁸ Indeed, while Rumelili argues that any conflicts over identity can become aggravated by reconciliation attempts because the opponents find ontological security in emphasizing their difference, ⁷⁹ she also suggests that this problem could potentially be overcome by finding a common – future – self for the competing parties. ⁸⁰ Political division, in other words, does not have to lead to a conflict over statehood itself. However, the articulation of overlapping state projects and fixation on exclusive and encompassing state personhood renders conflicts fundamentally irreconcilable. In this constellation, the unitary self that is represented by the parent state constitutes an existential threat to the separatist cause, and vice versa. It is these extreme cases which reveal the meta-security dilemma posed by conflicts over exclusive state subjecthood.

A good example for the dilemma faced by third-party mediators are the status talks over the future of Kosovo at the time of the UN interim administration that had been authorized by the Security Council following the intervention by NATO in 1999. ⁸¹ Appointed at the end of 2005, UN special envoy Martti Ahtisaari was tasked to bring Serbia and Kosovo to agree on the future status of Kosovo, a mission that turned out to be impossible. While the Serbian side insisted that ‘Kosovo (. . .) will always remain part of Serbia’ and that any potential ‘compromise does not include independence’, the Kosovar delegation maintained ‘that there could be no concessions on the key question of statehood’. ⁸² By early 2007, Ahtisaari noted that the deadlock in his mediation efforts could not be overcome. Considering that ‘Kosovo’s current state of limbo cannot continue’, he concluded ‘that the only viable option for Kosovo is independence’. ⁸³ Kosovo politicians eventually took matters into their own hands by declaring independence in 2008. ⁸⁴ While Serbia and many others rejected the unilateral secession, Kosovo received support from Western states and was recognized as a new state by many others, indicating that the search for a compromise had failed. The – degree of – fundamental ontological security that Kosovo received came at the price of the fundamental insecurity that Serbia suffered due to its loss of a ‘sacred’ part of its national territory. ⁸⁵

Another case in point are the failed mediation efforts of the OSCE Minsk group on the status question of Nagorno-Karabakh. Since the ceasefire agreed between Armenia and Azerbaijan in 1994, the Minsk group, spearheaded by Russia, France, and the US, had sought to find a compromise solution for Nagorno-Karabakh’s status. Yet just as the self-styled Republic of Nagorno-Karabakh was categorically unwilling to consider its re-integration into Azerbaijan proper, the latter was adamant not to let the separatists leave. In September 2020, the conflict re-erupted at full scale, ending with a new ceasefire agreement acknowledging Azerbaijan’s territorial gains, which provided Azerbaijan with renewed assurance of its statehood while pushing Nagorno-Karabakh to the brink of extinction. As in the case of Kosovo and Serbia, the clashing state projects ultimately proved irreconcilable. Again, the point is not that there cannot be an uneasy coexistence between de facto and parent states, as is sometimes the case, ⁸⁶ but that their overlapping claims cannot be easily accommodated in an international order delineated into exclusive states.

It might seem tempting to circumvent the meta-security dilemma by determining which entities are ‘really’ states and which are not so as to distinguish existing states from mere claims to statehood. This is arguably a function of international law and many associated diplomatic and legal practices, including formal state recognition and legal analysis. ⁸⁷ Yet, the identification of states is often contested in international law as well. Many other states disagreed that Kosovo had become a state, and even the International Court of Justice was careful to avoid an answer to that question when asked for its opinion about whether or not the declaration of independence had violated international law. ⁸⁸ While Russia rejected the independence of Kosovo, it in turn recognized Abkhazia and South Ossetia, a move that was then spurned by most other states. Meanwhile, Somaliland and many other de facto states are denied international recognition, although they display the attributes of effective statehood in international law. ⁸⁹ Ultimately, neither recognition nor effectiveness simply determine states in international law, nor do any other particular criteria. ⁹⁰ Legal analyses usually combine various factors to make their case, and yet remain essentially contestable. ⁹¹ In short, recourse to international law reformulates the dilemma, but it does not solve it.

Instead of circumventing the meta-security dilemma, legal state identification is in fact deeply involved in deciding conflicts over statehood in favor of one side or the other. ⁹² This is largely implicit where formally recognized states benefit from the legal protection of their sovereignty and territorial integrity, while any contenders appear as rebels of one type or another. However, the fact that the advantage usually lies with recognized states should not lead to the conclusion that they are actually states, while separatists are not, for this is what is at stake in these conflicts. Occasionally, separatists do become recognized as states.

An illustrative example for the rare legal advice in favor of the recognition of rebels as new states are the opinions of the arbitration commission counseling the European Community (EC) during the Yugoslav crisis in the early 1990s. The commission issued a series of highly influential legal opinions that evaluated claims of statehood in Yugoslavia. ⁹³ While the EC had initially sought to broker peace within the Yugoslav federation, the arbitration commission assessed in its first opinion of 20 November 1991 that Yugoslavia was ‘in the process of dissolution’. ⁹⁴ The commission further recommended the recognition of several constituent republics, while rejecting demands by Serbian separatists within Croatia and Bosnia and Herzegovina for their own independence or accession to Serbia. ⁹⁵ Absent any middle ground, the statehood of the federation was lifted to endorse the rebel republics, whereas these new states were in turn protected against rebels on their own soil. These rearrangements thus not only facilitated the disappearance of Yugoslavia but also prepared the ground for the eventual defeat of the forces of the Republika Srpska in Bosnia and Herzegovina and for the extinction of the separatist Republic of Serbian Krajina in Croatia. ⁹⁶

In all these cases, political mediators and legal arbitrators faced the meta-security dilemma that reassuring either the parent state or the separatists rendered the respective other side fundamentally insecure about its territorial and national integrity, and ultimately its very state personhood. Thus, rather than finding a genuine middle ground or objectively identifying which entities qualified as states, mediators, courts, commissions, and third states seeking to facilitate a compromise or adjudicate in contested cases of state creation were compelled to eventually take sides in favor of one claim or the other.

Yet, if third parties tend to take a decision either in favor of the established state or, more rarely, the separatists, this also implies their invariable commitment to the state form as the very type of subjectivity to be preserved. Indeed, third parties typically determine which contenders are eventually regarded as states, but they do not doubt that one or the other must be a state. ⁹⁷ As Mitzen points out, states ‘can only survive as members of the states system; as such their survival motive is intrinsically relational and can only be expressed as a positioning of the self vis-a-vis other states’. ⁹⁸ By extension, states need the world to be composed of other states since this is the basis of everyday international relations and hence the environment within which states ‘get by’ and exist at all. Conventional physical and ontological security concerns of states, and all the routine interactions they involve, are premised on the state-based international order which defines concrete states as the relevant subjects of security in the first place. ⁹⁹ Although a closer look into the potential anxiety of third states would be necessary to clarify this dynamic further, it appears that determining state subjects where they are contested reassures not only the recognized contender, at the expense of its rival, but also the third states in their own fundamental subjectivity as states. As such, deciding meta-security dilemmas of state identification in favor of one or the other appears not only as a tragedy, but also as a sacrifice. The fundamental ontological security of a world of states comes at the price of the fundamental ontological insecurity of failing aspirants for statehood as well as of other political communities marginalized and neglected by the creation of exclusive state subjects and spaces stretching the earth.

Conclusion

Studies of ontological security in IR have successfully brought attention to the role of anxiety, fears, and identity in international politics, suggesting that states act in particular ways to secure a sense of who they are. However, the very state subjects whose ontological (in)security is at stake are usually taken for granted. As I have argued in this article, this misses the role that insecurities play in constituting and re-affirming states in the first place. That is, before and beyond traditional concerns over ontological and physical security, states or would-be states have a need for a more fundamental ontological security that is achieved by being constituted and recognized as states. While this dimension is largely invisible in everyday international relations between established states, it becomes evident in separatist conflicts, in which different state projects clash and overlap. Since state subjectivity is typically imagined as exclusive, rival claims to statehood in the same territory and for the same population are ultimately irreconcilable. Third parties attempting to resolve secessionist conflicts thus face a meta-security dilemma: reassuring one contender by recognizing or reaffirming its claim to statehood necessarily renders the other one fundamentally insecure.

The analysis of fundamental ontological (in)security contributes to a relational and reflexive understanding of state subjectivity in OSS and IR more generally. In particular, it suggests that there is a deeper dimension of ontological (in)security to be explored if we move from a functionally substantialist starting point of given states to an interest in how political communities are constituted as apparent state subjects which must be continuously reaffirmed. Indeed, as separatist conflicts show, the respective ‘others’ of recognized states are not only third states, their usual international counterparts, but also the rival would-be states and other entities that have lost out historically or continue to struggle for their existence but remain marginalized in the international arena. A heightened attention to the contingency of any ‘state’, recognized or not, thus allows for reflecting on fundamental (in)securities otherwise neglected and forgotten.

This also has implications for several other questions discussed in OSS. First, for the state-as-actor assumption, there is a bridge to be built between the critical argument that assuming ‘enclosed subjects’ ¹⁰⁰ such as states has problematic implications and the observation that ‘the “state as person” (. . .) indexes real aspects of the ways in which states operate in world politics’. ¹⁰¹ Indeed, one can critically investigate the implications of assuming states into or out of existence without denying the predominance of state subjectivity in ‘reality’, precisely by focusing on the (in)security it generates in practice, both for imagined communities and individuals identifying with them or not. Second, the distinction between physical and ontological (in)security appears to fade where the very personhood of states is contested, which calls for further inquiries into the specific conditions of the conventional opposition between them. Third, while the notions of fundamental ontological insecurity and the meta-security dilemma have been developed to grasp subject formation and related (in)securities at the intersection of the domestic and the international, they might also help analyze dynamics of domestic identity conflicts which only turn into secessionist conflicts because escalation pushes the quarreling parties toward separation. Finally, the persistent commitment of third states to the state form points to a more general anxiety over the preservation of the state-based international order itself, which could be further explored where it is openly challenged, as for instance by the so-called Islamic State in Syrian and Iraq. ¹⁰²

Ultimately, fundamental ontological (in)security calls for a focus on the dynamics, effects, and implications of constituting particular subjects of security. More specifically, the contingency of states revealed in secessionist conflicts points to the force of assuming both the very form of state subjectivity and concrete individual states. Indeed, state subjects are never complete or whole but retain a trace of what they have excluded. They are powerful projects of collective identity and exclusive security precisely because they are contestable and essentially unfinished. ¹⁰³ It is in this sense that apparent state subjects are not simply the starting point of quests for ontological security, but already their result.