Map of Europe |
The Societal Security Dilemma
*
Copenhagen Peace Research Institute
As other chapters in this book will have pointed out, one recent approach to the relationship between security and identity (particularly in the context of Europe) has been that of "societal security". In the 1993 book Identity, Migration and the New Security Agenda in Europe, 1 Ole Waever, Barry Buzan, Morten Kelstrup, and Pierre Lemaitre 2 argued that societal insecurities; insecurities over ethnic, national and religious identities, have become more and more important in relation to those over state sovereignty in contemporary Europe. The purpose of this chapter is to stay very much within this approach in beginning to construct the concept of the "societal security dilemma." 3
Within the field of International Relations (IR) neo-Realism has, for the most part, maintained a definite distinction between processes at the inter-state level and processes at the intra-state level. Moreover, given these two distinct levels, neo-Realists have argued that what really matters in IR is relations between states; that in the whole scheme of things intra-state relations have only a peripheral role to play on the IR stage. In view of this, it has therefore been perhaps somewhat surprising to learn that more recently (particularly since events in central and eastern Europe in 1989) neo-Realists have begun to pay ever greater attention to relations between ethnic and national groups; in particular, to the occurrence of ethnic conflict. (Mearsheimer: 1990; 1992; Snyder: 1993).
In this respect, one such approach has been that taken by Barry Posen. In his 1993 article "The Security Dilemma and Ethnic Conflict," Posen tried to operationalise the concept of the security dilemma out of its usual inter-state context in order to explain the outbreak of ethnic conflict in the former Yugoslavia in 1991. However, in doing so Posen has come in for much criticism, in particular from Josef Lapid and Friedrich Kratochwil. They have accused Posen of what they call "theoretical appropriation" (Lapid & Kratochwil: 1994; 1996); put simply, this entails looking at the intra-state level, but through neo-Realist glasses fitted with inter-state lenses. So how should we theorise about the occurrence of ethnic conflict? Should we perhaps look through a new pair of non-neo-Realist glasses? This chapter will argue that in certain cases (such as the former Yugoslavia) it is not so much that our inter-state lenses do not allow us to see many important insecurities clearly enough, rather that they simply do not allow us to see some other, perhaps more important, insecurities. Therefore, in beginning to construct the concept of the societal security dilemma, we won"t so much be looking through new, non-neo-Realist glasses, as simply changing the inter-state lenses in our old, neo-Realist pair to allow us to see other insecurities better. In other words, we won't so much be throwing out our inter-state concept (the security dilemma), as just making some theoretical changes to make it perhaps fit the intra-state level better (the societal security dilemma).
As its name suggests, the concept of the societal security dilemma should, on the face of it, be constructed from two other concepts: firstly, the security dilemma; and secondly, societal security.
In simple terms, a security dilemma exists when the actions of one state, in trying to increase its own security, causes a reaction in a second state, which, in the end, decreases its (the first state's) own security (Herz: 1951; Jervis: 1976; Posen, 1993). Moreover, although the first state"s actions may well have been for defensive purposes only (benign intent), the second state may still have interpreted them as offensive (malign intent). As such, the first state may have 'unintentionally' or 'inadvertently' decreased the security of the second. In this way, John Herz comments that: "It is one of the tragic implications of the security dilemma that mutual fear of what initially may never existed may subsequently bring about exactly that which is feared the most" Herz: 1966, 24) Similarly, Herbert Butterfield also saw the security dilemma as a "tragedy": 4
The greatest war in history could be produced without the intervention of any great criminals who might be out to do deliberate harm to the world. It could be produced between two powers both of which were desperately anxious to avoid conflicts of any sort (Butterfield: 1951, 19).The concept of the security dilemma can essentially be broken down into two stages: firstly, "interpretation"; and secondly, "response". In the context of the first stage, Nick Wheeler and Ken Booth argue that a security dilemma occurs when:
the military preparations of one state create an unresolvable uncertainty in the mind of another as to whether those preparations are for defensive purposes only (to enhance its security), or whether they are for offensive purposes [(to threaten another's security)] (Wheeler & Booth, 30). 5States may often try to increase their security by building-up their arms. However, some arms can be ambiguous: certain arms which can be used for defensive purposes can also be used for offensive purposes too. 6 Therefore, in some instances states may have difficulty in determining unambiguously whether another's military preparations constitute defensive or offensive actions (Jervis, 64). 7 Here, the second stage of the security dilemma also comes into play; response. Put simply, given the fact that I am uncertain about your military preparations, how should I react?
On the one hand, neo-Realists claim that, out of safety, anarchy and self-help forces states to assume a worst-case scenario. Indeed, Barry Posen argues that even though states may be well aware of the consequences of their actions (to decrease the security of others) the "nature of their situation compels them to take the steps that they do" (Posen, 28). In this way, neo-Realists argue that the security dilemma is structurally-driven. On the other hand, however, Constructivists claim that the security dilemma is not structurally driven, rather it is unit-driven by states themselves. Alexander Wendt argues that there is nothing inherent in anarchy for states to assume a worst-case scenario. He comments that [w]e do not begin our relationship ... in a security dilemma; security dilemmas are not given by anarchy ..." (Wendt: 1992, 407). Instead, it is our past and present relations with other states that determine whether or not we assume a worst-case scenario; for instance, given a similar situation what did the other do last time? Even so, neo-Realists argue, worst-case scenarios may not only be based on what the other may do now, but also on what he might do in times to come. 8 Therefore, to be sure about our future security we may still have to build-up our own arms.
As such, the precise role that anarchy plays in the security dilemma is still a contested one. Alan Collins comments that:
For some anarchy is not a cause of the security dilemma but rather a necessary condition. That is, because anarchy promotes self-help behaviour and leaves states uncertain of others' intent it creates propitious conditions for the security dilemma, but is itself not a cause .... [But] whether a cause or a necessary condition what is clear is that its role is important (Collins: 1996, 4).Worst-case assumptions between states can lead to a process of action and reaction which may often manifest itself as arms races; that is, the more that you increase your arms, the more I have to increase mine if I am to maintain the same level of security, and even, in the end, war. This is often referred to as the "spiral process".
The main problem, however, with the concept of the security dilemma is one of being able to distinguish between benign and malign intent on the part of states. Only by doing so can we determine if the first state really did "inadvertently" decrease the security of the second; and hence, only by doing so can we can really call the security dilemma a "tragedy". Collins argues that: "The essence of the security dilemma is the tragedy that unknown to either participant their incompatibility, while appearing real, is actually illusory" (Collins, 5). But, like the participants themselves, ultimately it may also be too difficult to us to distinguish between real and illusory incompatibility. 9
Although very much concerned with relations between states, Barry Posen has tried to operationalise the concept of the security dilemma out of its usual inter-state context to explain the outbreak of ethnic conflict in the former Yugoslavia in 1991. In doing so, Posen's approach is based largely on what he calls "an emerging anarchy" (1993, 27); that is, when multi-ethnic states begin to collapse ethnic and national groups must provide their own security as no effective central authority will be in place to provide it for them: It becomes a self-help situation much like that between states in the international system.
However, Posen"s approach has come in for much criticism (Joeneimmi: 1996; Lapid & Kratochwil: 1996). These have largely been concerned with Posen "imposing a realist, state centered theory" at the inter-ethnic (intra-state) level (Joeneimmi, 13-14). Josef Lapid and Friedrich Kratochwil, for example, argue that Posen"s approach "strongly suggests that all “units” in “anarchy” (be they individuals ... city-states, nations, or states) can be expected to behave in accordance with the very same neo-Realist logic (113). In other words, given an "emergent anarchy", "all" units at the intra-state level will act in much the same way as states do in the international system. Thus, they also argue that Posen largely ignores "unit level characteristics" as a cause of conflict, commenting that: "Posen lets the neo-Realist structure do most of the explaining by relying ... on a worst-case scenario" (ibid, 114).
However, in some ways these criticisms are perhaps somewhat unfair. Firstly, Posen appears to argue that, rather than "all" units, "state-like" entities in anarchy will simply act in much the same way as states. For example, in the case of the former Yugoslavia he is concerned with Serbia, Croatia, 10 the Krajina Serbs, and so on (and not with "individuals" or "city-states"). And secondly, Posen also appears to argue that, rather than anarchy in itself causes worst-case scenarios, structure simply creates the "necessary conditions" for security dilemmas. For example, he comments that: "A group suddenly compelled to provide its own protection must ask the following questions about any neighbouring group: is it a threat? How much of a threat?" and so on (1993, 27). Indeed, Posen, rather than ignoring "unit level characteristics", further comments that "history" (not anarchy) will be the main method used "to assess the offensive implications of another" (ibid, 30). 11
However, where Posen may indeed be criticised is that his pair of inter-state glasses has perhaps not allowed him to see everything that he should. While Posen may not have been wrong to assume that state-like entities (like Serbia, Croatia, and so on), although not recognised as sovereign units, will have similar security concerns to those of states (for example, territorial integrity, political autonomy), he does appear, however, to have largely ignored some other, important security concerns that such groups may well have. By contrast, Waever et al's societal security approach has not: They have changed the lenses in the neo-Realist pair of glasses, and by doing so have thus allowing us to see other insecurities more clearly (see, for example, Lapid & Kratochwil, 117-120).
The term societal security was first used by Barry Buzan in the book People, States and Fear (Buzan: 1991). Societal security was just one of the five sectors in his five-dimensional approach to security theory, along with military, political, economic, and environmnetal security. Here, however, all of Buzan's dimensions, including the societal one, were still sectors of state security: Society, for example, was just one sector where the state could be threatened.
In Identity, however, Ole Waever argues that Buzan's previous five-dimensional approach had become 'untenable' as a present context for societal security (Waever: 1993, 25). As a result, he proposed a reconceptualisation of Buzan's previous theory; not of five sectors of state security, but of a duality of state and societal security. Societal security is still kept as a sector of state security, but now it is also a referent object of security in its own right. Whereas state security is concerned about threats to its sovereignty (if the state loses its sovereignty it will not survive as a state), societal security is concerned about threats to a society's identity (if a society loses its identity it will not survive as a society) (ibid, 25). Therefore, although the state is still a referent object for the military, political, economic, societal, and environmental sectors, 'society' is also a referent object for the societal sector.
One reason for this move is that there are many societies which either do not have, or do not fit (either geographically or politically, or both), existing states. As Waever et al point out:
Societal security is important in itself because communities (that do not have a state) are also significant political realities, and their reactions to threats against their identity will be politically significant (ibid, ?).According to Buzan, societies are 'fundamentally about identity' (Buzan: 1993a, 6). Similarly, Waever argues that:
The key to society is that set of ideas and practices that identify individuals as members of a social group. Society is about identity, about the self-conception of communities and individuals identifying themselves as members of a community (Waever: 1993, 24).In other words, societies are constituted by a sense of social identity, where '[at] its most basic, social identity is what enables the word 'we' was just one sector where the state cto be used (Waever, 17).
'Society', however, can be distinguished from just 'social group'. Put simply, not all kinds of social group constitute what is meant here by society. Social groups must be significant enough to be a referent object of security in their own right; they must be able, like the state, to operate as units in the international system. This means that societal security is concerned with the security of society as a whole, rather than the security of groups 'in' society. As Waever argues:
Security operates neither at the level of the global and universal, nor at the individual level; it is about the interraction of collectivities (Waever: 1994, 7).
And, he goes on to say:
What is important in a security context is when it is possible to mobilize a significant collectivity. Security action is always taken on behalf of, and with reference to a collectivity. The referent object is that which you can point to and say: 'it has to survive, therefore it is necessary to ' (ibid, 8)Here, societies are therefore social groups which are able to act alongside the state as significant political units in the international system. In contemporary Europe 'they are mainly national' (Buzan, Waever & de Wilde: 1995, 41), but elsewhere they may be religious, for example. Given this, 'the main units of analysis for societal security are ... politically significant [ethnic], national and religious entities' (Waever: 1993, 23). Importantly, where religious and ethno-national identities reinforce each; for example, Catholic Croats and Orthodox Serbs, very strong identities are usually produced.
According to Waever, in the international system societal security concerns:
the ability of a society to persist under changing conditions and possible and actual threats. More specifically, it is about the sustainability, within acceptable conditions for evolution, of traditional patterns of language, culture, association, and religious and national identity and custom (ibid).
It is difficult to objectively define when there is a threat to societal security. As Buzan points out:
[W]hat is perceived as a threat, and what can be objectively assessed as threatening, may be quite different. Real threats may not be accurately seen. Perceived threats may not be real, and yet still have real effects (Buzan: 1993, 43).
Threats to societal security can occur when societies perceive that that its 'we' identity is being put in danger. Those means which can threaten a society's identity may range from the suppression of its expression to the interference with its ability to reproduce itself. According to Buzan, this may include 'forbidding the use of language, names and dress, through closure of places of education and worship, to the deportation or killing of members of the community' (ibid, 43). And that threats to the reproduction of a society can occur through the 'sustained application of repressive measures against the expression of the identity. If the institutions that reproduce language and culture are forbidden to operate, then identity cannot be transmitted effectively from one generation to the next, (ibid, 43). 12 Therefore, some threats to societal security may be military ones (killing members of the group, conquering historic territory). But, some threats may also be non-military ones (denying language rights, freedom of worship).
Therefore, if we look through our pair of neo-Realist glasses, but with our societal security lenses in them, we can then see that while many societies ("state-like" entities) may still be concerned with insecurities over, for instance, territorial integrity and political autonomy, the most important insecurities, however, may well be over the maintainance of their identity. Given this, perhaps it would be better to thus try to operationalise the concept of the security dilemma at the intra-state level in terms of societies (Waever et al) rather than "state-like" entities (Posen); a "societal security dilemma". Indeed, in Identity, Buzan comments that:
To the extent that tensions over migration, identity and territory occur between societies, we might by analogy with international politics talk about a societal security dilemma. This would imply that societies can experience processes in which perceptions of 'the others' develop into mutually reinforcing 'enemy pictures' leading to the same kind of negative dialectics as with the security dilemma between states (Buzan: 1993, 46).This, unfortunately though, is just about all that Buzan says about the societal security dilemma. Indeed, later in Identity, along with Ole Waever, Morten Kelstrup, and Pierre Lemaitre, he points out that the concept still needs to be given some further thought:
If we think of societies as units, do we therefore have to think of societal security dilemmas between them? Such an investigation would require further investigation into the interplay of identities (Waever et al: 1993, 190).The task in hand is therefore to begin to construct the concept of the societal security dilemma; to, in effect, take on where Buzan has left off.
By analogy with the (state) security dilemma, a societal security dilemma might exist when the actions of one society, in trying to increase its societal security (strengthening its own identity), causes a reaction in a second society, which in the end, decreases its (the first society"s) own societal security (weakens its own identity). 13 This suggests that, as between states, some difficutly in being able to distinguish between "defensive" and "offensive" preparations might also exists between societies: The difference here being that we are focusing on threats to identity rather than sovereignty, and, in some cases, on "something" rather than arms to defend against these threats. In other words, if arms are used by states to defend their sovereignty, what do societies use to defend their identity? And, if arms can sometimes be ambiguous because of both their defensive and offensive attributes, what ambiguous attributes might this "something" have?
Firstly, like states, societies can also defend their identity by increasing their arms. If one society"s sense of identity very much depends on territorial integrity (the defense of the "homeland"), then arms can be used to defend this. The problem we are left with here, however, is that the processes of the resultant societal security dilemma would closely resemble those of the (state) security dilemma. In both cases, they would exist largely as a result of the ambiguity of some military preparations. Thus, in this instance, the societal security dilemma would not be too far removed from the (state) security dilemma which Posen has tried to operationalise. In this way, our thinking behind this would neither be new nor particularly surprising.
However, the use of arms to protect societal identity suggests that some kind of conflict may already exist between two societies. In this instance, arms may be used to defend societal identity as a result of other, non-military insecurities. Indeed, Lapid and Kratochwil comment that: "In short, without an explicit theoretical treatment of the bases of group differentiation, which, in turn, generate the anarchical environment, structural arguments do not explain conflict, they merely redescribe it" (1996, 115). Or, to put it another way: Firstly, non-military insecurities may produce societal security dilemmas. Secondly, non-military societal security dilemmas 14 may produce conflict. And thirdly, existing (non-military) conflicts may lead to the collapse of multi-ethnic states, which, in turn, may also produce the same kind of military security dilemmas as with Posen"s approach. 15 This, however, would indeed represent a fundamental break in our thinking about the concept of the security dilemma.
Given this, what we therefore need think about is a situation, or situations, where military capabilities are not, in the first instance, the means by which societies try to defend their identity.
In Identity, Waever et al suggests that:
For threatened societies, one obvious line of defensive response is to strengthen societal identity. This can be done by using cultural means to reinforce societal cohesion and distinctiveness, and to ensure that society reproduces itself effectively (Waever et al: 1993, 191).Elsewehere, Waever argues that culture can be defended "with culture", and that: "If one's identity seems threatened the answer is a strengthening of existing identities" (Waever: 1995, 68). The idea of defending culture with culture is embodied in "cultural nationalism". John Hutchinson, for example, defines cultural nationalism as a "movement of moral regeneration which seeks to re-unite various aspects of the nation" (Hutchinson: 1994, 123). And, he goes on to say:
The aim of cultural nationalism is the moral regeneration of the political community, or, in other words, the re-creation of their national civilization (ibid, 124).Cultural nationalists are concerned with enhancing their culture; or, to put it another way, with strengthening their identity. Cultural nationalism may often produce an intense feeling of self-identification: It will often emphasise the various commonalities such as language, religion, ideology, and history, and will downplay other local or religious ties which may detract from its unity (Alter: 1994, 12). This process of self-identifcation may often take place because societal identity is perceived to have been weakened. As Waever points out: "hurt pride and humiliation can often contribute significantly to the rise of nationalism," (Waever: 1994, 20). He goes on to say:
It [nationalism] offers a particularly attractive mode in times of crises and depression since the link to a glorious past donates immediate relief, pride and [a] shield against shame (ibid, 21).And, that "the development of nationalism is accelerated by an idea about the existence deep in the national soul of a golden future" (ibid). Therefore, attempts to defend a threatened identity will often strengthen it.
But, as Peter Alter notices, nationalism may also have another effect:
[Societal] groups tend to define their national identity and national consciousness in negative terms, through distinction from or comparison with neighbours. Encounters with the "alien"; other forms of language, religion, customs, political systems, make people aware of their close ties, shared values and common ground that render communication with their own kind so much easier than with outsiders (Alter, 12).Societal identity is usually defined along several dimensions; for example, language, religion, and history. However, at any given point in time only one, or a few, of these dimensions will be the "dominant identity values"; that is, those values for which the majority of a society identify as most characterising "us". And, it is often by comparing oneself with the "Other" that these dominant identity values will be produced. For example, the Eastern Orthodox religion has been a dominant identity value for Serbian identity as it has distinguished the Serbs from both its Catholic and Muslim neighbours. Who we are can often mean who we are not: Some societies may need "Others" to remind themselves of their own true identity. This is what Michael Ignatieff calls a nation's "defining Other" (Ignatieff: 1993). 16
Strengthening one's own identity may therefore also involve weakening another's. As Waever argues:
[This] logic of identity means that some other often enters as part of the self-identification... [As] one's identity depends on this other, the other ends up in the dual role of being necessary for my identity, and the one who fully prevents me from being fully myself (Waever: 1994, 19).Given this incomplete (weak) identity, one might then blame the other who prevents me from being fully myself; the corollary of this being, if I could only get rid of the other I could finally be me (ibid). 17 Indeed, Katherine Verdery argues that in the case of central and eastern Europe in particular, nationalism is invariably a language of victimisation; that in times of crises it is easy to explain one's own suffering in national terms. And, she goes on to suggest, given this, it is others that "come to represent the loss of a feeling of wholeness" (Verdery: 1993, 196-201).
Cultural nationalism may therefore be ambiguous, in as much as it can be used as a means to both strengthen one"s own identity and also weaken another's. Given this, some societies may have difficulty in determining unambiguously the intentions of another's cultural nationalism. As a result (depending on the nature of the groups' past and present relations), thet may assume a worst-case scenario. This could then lead to a "spiral of nationalisms" whereby each group's actions have the effect of weakening the other's identity. In this way, societal security dilemmas exist when both groups believe that their identity has to be dominant over the other. 18
However, given that another society may also be a part of the self-identification process, my strong sense of identity will invariably depend on your weak sense of identity. The problem with this is that in terms of the concept of the (state) security dilemma there can be security for both; it"s just that the two sides can't see this. However, with this concept of the societal security dilemma there can't.
One example of this could be in terms of language rights: one society, where homogeneity within the state where it is the dominant (majority) group is seen as a requirement for its societal security, may attempt to deprive its minority group (a second society) of its language rights by trying to close the second society's own language schools. 19 This, invariably, will make the second society more determined to maintain them (threatening the existence of its language), which, in turn, will then make the first society even more determined to close them (threatening the homogeneity of the state). And so this spiral process may continue. A case in point here might be that of relations between Serbs and Croats in the Croatia.
In April 1990, Franjo Tudjman became Croatia's president. Soon after violence broke out between Croats and Croatia's Serbian minority (the Krajina Serbs 20 ) over demands by the Krajina Serbs for cultural autonomy in Croatia (Djilas, 1995, 93). The demands of the Krajina Serbs, who made up 12% of Croatia"s population (Gow: 1991, 293) had essentially been a reaction to Tudjman"s nationalist programme of "Croatia for the Croats", or "Croatisation".
Relations between Serbs and Croats had always been a source of conflict in the former Yugoslavia. On the one hand, following the creation of Yugoslavia on 1 December 1918, the Croats felt deceived that the new state was not a shared, but a centralised, Serb-dominated one. After all, the capital Belgrade was in Serbia; the royal family, the Karadzordzevic, was Serbian; the orthodox church was favoured; and the Serbs dominated the bureaucracies, the military, and the police (Wiberg: 1993, 98). On the other hand, however, the Serbs felt that they were the natural and deserving hegemons of the south Slav peoples, and without their great sacrifcies the creation of Yugoslavia would not have been possible.
In 1941, Nazi Germany established the "Independent State of Croatia" (NDH) under the leadership of the Croatian fascist, Ante Pavelic. The main aim of the Croatian fascists (Ustasha) was to create an ethnically pure "Greater Croatia". The NDH, however, was home to a large Serbian minority;1.9 million out of a total population of 6.3 million (Malcolm: 1994, 176). In doing so, during the Yugoslav Civil War (1941-45) the Ustasha killed around 300,000-400,000 Serbs in trying to create their Greater Croatia, while Serbian nationalists (Chetnicks) also killed around 200,000 Croats (Almond: 1994, 137) in trying to create a "Greater Serbia".
Tudjman"s "Croatisation" was both a symbolic and pragmatic programme. For example, many of the symbols of Croatia"s independent past were restored; the checkerboard flag (the Sahovnica) and coat of arms. Croatian in the Latin alphabet was designated as the official language and Catholicism as the official religion, 21 and dual road signs were torn down, even in Serb majority areas. The people were obliged to swear an oath of allegiance to Croatia. Many ethnic Serbs were removed from the bureaucracies and the police and replaced by Croats (Hockenos: 1991, 14-15; Wiberg, 105).
"Croatisation" was essentially a programme of homogenisation, and therefore constituted a serious threat to societal security of the Krajina Serbs. In trying to defend their identity, the Krajina Serbs called for cultural autonomy in Croatia. However, the more they demanded this, the more the Croats were determined to deny them. And thus, we have our spiral process.
One reason for this might be that the Serbs entered as "Others" in the Croats' process of self-identification. Michael Ignatieff argues that the relationship between Serbs and Croats can be seen as a case of what Freud called the "narcissism of minor difference"; that is, the less the real difference between two groups, the more it is bound to loom in their imaginations (Ignatieff, 14). He goes on to say:
Its corollary must be that enemies need each other to remind themselves of who they really are. A Croat, thus, is someone who is not a Serb. A Serb is someone who is not a Croat. Without hatred of the other, there would be no clearly defined national self to worship and adore (ibid).Similarly, Warren Zimmerman comments that: "The larger truth may be that it is the similarities, rather than the differences, among Balkan peoples that makes them vulnerable to nationalist manipulation" (1996, 12). Indeed, although Serbs and Croats appear to be remarkably similar in many ways, many Croats, for example, began to talk about differences between the two groups, not only in religious terms, but also in terms of how the Croats were "civilized Europeans" while the Serbs were more "uncivilized, Balkan-types".
When another group is part of the self-identification process homogenisation therefore becomes the likely outcome: how can we truly be Croats while there are Serbs here to deny us this? As Slavoj Zizek notes:
Yugoslavia today is a case ... in which .... [e]very nationality has built its own mythology narrating how other nations deprive it of the vital part of enjoyment the possession of which would allow it to live fully (quoted in Waever: 1994, 1).In another way, however, where homogeneity does not become a requirement for societal security, the concept of the societal security dilemma appears to work better; there can be security for both sides, if they were only able to see it. A case in point here might be that of relations between Romanians and Hungarians in Romania.
Following demonstrations in February 1990 over independent Hungarian language schools, in March 1990 violence broke out between Romanians and Romania's Hungarian minority (the Transylvanian Hungarians) 22 in many parts of Transylvania. Some Romanians tore down Hungarian language signs and vandalised Hungarian protestant churches, declared: "Out with the Huns. We are prepared to die defending Transylvania; Bozgor (Hungarian), don't forget this is not your homeland" (Kurti: 1993, 5-7)
Transylvania had long been a source of conflict between Romanians and Hungarians: although originally a part of the ancient Hungarian kingdom, the region has had a majority Romanian population since the 14th century, and enjoyed a great deal of autonomy from medieval Hungary. Following the 1848 Hungarian War of Independence, where ethnic Romanians in Transylvania had fought against Hungary, in 1867 the region became part of Hungary in the Dual Monarchy with Austria. The Hungarians pursued repressive and assimilatory policies until, following the First World War, the Treaty of Trianon gave Transylvania to Romania in 1920. In 1940, the Vienna Accord returned northern Transylvania to Hungary, only for it to be given back to Romania by the Treaty of Paris in 1947 (Dick: 1995, 20-21). Today, approximately 72% of the region"s population is Romania, while Hungarians make up approximately 24% (1.7-2 million).
Many Romanians believed that the Transylvanian Hungarians should be denied cultural autonomy. This was mainly because they believed that cultural autonomy, which included independent Hungarian language schools, would be a first step along the road towards Transylvania's secession from Romania, and later to its re-unification with Hungary itself. This would not only threaten the territorial integrity of the Romanian state, but would also threaten the identity of the region's Romanian population (becoming a minority in a foreign state once again).
For many Hungarians, however, secession does not appear to have been their intention. As C J Dick argues:
They [Transylvania's Hungarians] are perfectly prepared for the most part, to remain under Romanian rule, provided they are not treated as second class citizens. All they want is guarantees of their collective rights: education in their own language to university level, the right to use their language in administrative matters and the courts, proportionate representation at all levels of administration and no economic or other discrimination (Dick, 20).In this way, the societal security of the Translyvanian Hungarians (independent language schools) was seen by many Romanians as something that would threaten both their own state and societal security (territorial integrity). Therefore, the more the Transylvanian Hungarians called for their own language schools, the more the Romanians were determined to deny them this.
Actions on the part of many Romanians might be explained not only by Transylvania's history, but also by the general economic and political conditions which existed in Romania at this time. Given that over the past 100 years or so Transylvania had swapped hands between Hungary and Romania several times, it is perhaps easy to see why some Romanians feared that Transylvania might secede. Also, the demonstrations had coincided with Hungarian celebrations commemorating the 1848 War of Independence. But, such fears were exacerbated largely because of Romania"s poor economic and political situation. As Laszlo Kurti points out:
Such manifestations took place at a time when the Romanian economy had reached an all time low: shelved were again emptied [sic], and the black market was once more accepted as a permanent way of life for many, turning citizens against one another in an incessant, futile struggle to make ends meet (Kurti, 5).Similarly, Katherine Verdery argues that given Romania"s economic and political chaos at the time:
The contrast between the anarchy of Romania's political scene and the discipline ... of the political party of the Hungarians, makes it easy for Romanians to believe in a Hungarian plot to recover Transylvania with another mutilation of Romania, as happened in 1940 (Verdery, 196)
The purpose of this chapter has been to begin to construct the concept of the societal security dilemma. In doing so, it has changed the lenses in Barry Posen's pair of neo-Realist glasses from state security to societal security ones. This has thus allowed us to see other important insecurities; insecurities over the maintenance of ethnic, national, and religious identities. Moreover, in some ways it has fundamentally changed our thinking about the concept of the security dilemma. In this respect, our main move here has perhaps been to suggest that thinking about societal security dilemmas can open the door somewhat to also thinking about non-military security dilemmas.
Thinking about societal security dilemmas may also provide us with a better understanding as to the causes of some ethnic conflicts. In this respect, non-military societal security dilemmas could produce the kind of conflict between societies that leads to the collapse of multi-ethnic states, which, in turn, can also produce the necessary conditions for military societal security dilemmas and further (military) conflict. In other words, while in some cases societal security dilemmas may operate solely within the societal sector of security, in others they may come to operate in both the societal and military sectors of security (and usually in the political and economic sectors as well). Moreover, this may also reflect a change firstly in the aim of the conflict from, for example, from demands for cultural autonomy within a state to those for secession from a state (see also chapter VII, Linda Bishai, "Secession and Security: The Politics of Ethno-Cultural Identity"); and secondly, in the nature of the conflict from, for example, from localised ethnic violence to widespread ethnic conflict.
Although there are still many theoretical problems to be overcome, this chapter should simply be considered as a starting point in thinking about the concept of the societal security dilemma. The challenge ahead is therefore both one of further developing the kind of thinking we have started here, and of further trying to operationalise it in cases of ethnic conflict.
Alter, P. (1994) Nationalism, London, Edward Arnold
Butterfield, H. (1951) History and Human Relations, London, Collins
Buzan, B. (1991) People, States and Fear: An Agenda for International Security Studies in the Post-Cold War Era, London, Harvester Wheatsheaf (2nd Edition)
Buzan, B. (1993) "Societal Security, State Security, and Internationalisation", in Waever et al, Identity, Migration and the New Security Agenda in Europe, London, Pinter, 41-58
Buzan, B. (1993a) "The Changing Security Agenda in Europe", in Waever et al
Buzan, B., Waever, O. & de Wilde, J. (1995) "Economic, Political and Societal Security", Copenhagen Peace Research Institute (COPRI) Working Paper, no.10
Collins, A. (1996) "An Intrastate Security Dilemma", Draft Paper, University of Wales, Aberystwyth.
Dick, C. J. (1995) "Instabilities in Post-Communist Europe: Hungary", Conflict Studies Research Centre, January, 1-26
Djilas, A. (1995) "Fear Thy Neighbour: The Breakup of Yugoslavia", in Kupchan (ed.) Nationalism and Nationalities in the New Europe, Ithaca, Cornell University Press, 85-106
Gow, J. (1991) "Deconstructing Yugoslavia", Survival. July-August, vol. 33, no.4, 291-311
Herz, J. (1951) Political Realism and Political Idealism, Chicago, Chicago University Press
Herz, J. (1966) International Politics in the Atomic Age, New York, Columbia University Press.
Hockenos, P. (1991) "House on Fire", New Statesman and Society, 17 May, 1991, 14-15.
Hutchinson, J. (1994) "Cultural Nationalism and Moral Regeneration" in Hutchinson & Smith (eds.) Nationalism, Oxford, Oxford University Press, 122-131
Ignatieff, M. Blood and Belonging, Chatto & Windus, 1993
Jervis, R. (1976) Perception and Misperception in International Politics, New York, Princeton University Press
Jervis, R. (1978) "Cooperation Under the Security Dilemma", vol. 30, no.2, 167-214.
Job, B. (1992) "The Insecurity Dilemma: National, Regime and State Securities in the Third World", in Job (ed.) The Insecurity Dilemma: National Security of Third World States, Boulder, Lynne Rienner, 11-35
Kaufman, S. (1996) "An International Theory of Inter-Ethnic War", Review of International Studies, April, vol. 2, no.2, 149-172
Kurti, L. (1993) "A Culture of Enmity: Hungary, Romania and the Transylvanian Question", COPRI Working Paper, no.23
Lapid, J. & Kratochwil, F. (1994) "The Taming of the Shrew: Neorealist Appropriations and Theorizations of Nationalism", Paper for 35th Annual Convention, Washington, D.C., 28 March - 1 April, 1-53.
Lapid, J. & Kratochwil, F. (1996) "Revisiting the 'National': Toward an Identity Agenda in Neorealism", in Lapid & Kratochwil (ed.) The Return of Culture and Identity in IR Theory, Boulder, Lynne Rienner, 105-126
Mearsheimer, J. (1990) "Back to the Future: Instability in Europe After the Cold War", International Security, Summer, vol.15, no.1, 15-56.
Mearsheimer, J. (1992) "Disorder Restored", in G. Allison & G. F. Treverton (eds.) Rethinking Americas Security, New York, W. W. Norton, 213-237.
Malcolm, N. (1994) Bosnia: A Short History, London, MacMillan
Møller, B. (1995) "Non-Offensive Defence and the Korean Peninsular", COPRI Working Paper, no. 4.
Posen, B. (1993) "The Security Dilemma and Ethnic Conflict" Survival, Spring, vol. 35, no.1
Snyder, J. (1993) "The New Nationalism: Realist Interpretations and Beyond", in M. E. Brown (ed.) The Domestic Bases of Grand Strategy, Princeton, Princeton University Press, 79-102.
Verdery, K. (1993) "Nationalism and National Sentiment in Post-Socialist Romania", Slavic Review, Summer, vol. 52, no.2, 179-203
Waever, O. (1993) "Societal Security: The Concept" in Waever et al, 17-40
Waever, O. (1994) "Insecurity and Identity Unlimited" COPRI Working Paper, no. 14
Waever, O. (1995) "Securitization and Desecuritization" in Lipshutz (ed.) On Security, New York, Columbia University Press, 46-86
Waever, O, et al. (1993) "Societal Security and European Security", in Waever et al, 185-199
Wiberg, H. (1993) "Societal Security and the Explosion of Yugoslavia" in Waever et al, 93-109.
Zimmerman, W. (1996), "Last Chance for Bosnia?", New York Review of Books, 19 Dec. 1996, 10-13.
Note 1: Hereafter referred to as Identity. Back.
Note 2: Hereafter referrred to as Waever et al. Back.
Note 3: I would like to thank HÖkan Wiberg for his comments on the initial draft of this paper. Back.
Note 4: Although Butterfield, writing in 1951, never actually used the term "security dilemma". Back.
Note 5: Of course, states may also do so by joining alliances. Back.
Note 6: A tank, for instance, could be used either to attack a neighbour's territory or to defend one"s own territory. Back.
Note 7: Indeed, Robert Jervis argues that security dilemmas might be avoided if offensive weapons could be distinguished from defensive weapons (Jervis: 1978). However, according to Bjorn Møller, the situation is not as simple as this: "[I]t is not a matter of distinguishing between offensive and defensive weapons: All weapons are dual-purpose, i.e, [they] may be used for both defence and offence, albeit not necessarily to the same extent. the appropriate level of analysis is higher, namely that of military postures and strategies" (Müller: 1995, 4). Back.
Note 8: In other words, although we may enjoy friendly relations with each other now, how can I be sure that this will continue? especially if something were to happen in your country such as a military coup. Back.
Note 9: This seems to be particularly true in the case of more recent conflicts, with which this chapter is concerned with. It may be a long time after the conflict itself that enough is known to consider one side"s actions either benign or malign. Back.
Note 10: This refers to the period of conflict in the former Yugoslavia before Croatia had been recognised as a sovereign state. Back.
Note 11: For a more sophisticated analysis of the role of anarchy as a cause of conflict in the former Yugoslavia, see Michael Ignatieff, "Serbia and Croatia" in Ignatieff, Blood and Belonging (London: Chatto & Windus, 1993), pp. 12-41. Back.
Note 12: Threats to societal security may not only come from forbidding something, but also allowing it. Forbidding something will often be in terms of a threat to the minority group from the majority group. Allowing something, however, may be the other way around: allowing a minority group something may threaten the homogeneity of the state; for example, Estonia allowing the Russian minority an equal status in language. Back.
Note 13: Strictly speaking, societal security dilemmas should exist when one society "unintentionally" weakens the identity of the other. Although, as is the case with the (state) security dilemma, the problem exists of being able to distinguish between benign and malign intent on the part of societies. Back.
Note 14: There are, of course, several criteria upon which definitions of "military" and "non-military" security dilemmas could be based, and we should be clear about which one we intend to use here: for example, firstly, is it non-military because the ultimate aim of the conflict is non-military? After all, wars are usually about something non-military. Secondly, is it non-military because non-military insecurities are the cause of the conflict, even though the security dilemma is essentially being driven by a military fear of the other? Or thirdly, is it non-military because both non-military insecurities are the cause of the conflict, and the security dilemma is being driven by a non-military fear of the other? Here, we will be using the third one. Back.
Note 15: Indeed, in arguing that the (state) security dilemma was a cause of conflict in the former Yugoslavia, Posen, has, in a sense, missed the boat: the cause of the conflict is essentially what caused the state to collapse, and not conditions which were produced as a result. Back.
Note 16: "Defining Others" are often chosen to suit historical and political conditions at a particular time. Therefore, given changing conditions, over time a nation's "defining Other" will also change. Back.
Note 17: As Waever points out, however, because the "other" is also necessary for constituting my own sense of identity, to get rid of this other would also be to take away a part of me. Therefore, to pursue this kind of logic is ultimately self-defeating. Back.
Note 18: If one society accepts domination over its identity by another, then the spiral process will not occur. Back.
Note 19: This idea would equate very much to the process of "Russification". Back.
Note 20: The term "Krajina Serbs" is used here to describe Croatia's Serbian minority in general, although it should be pointed out that some fundamental differences did exist between the rural (Krajina) Serbs and the urban (Zagreb) Serbs. Back.
Note 21: Throughout Yugoslavia, Serbs and Croats more or less spoke the same language Serbo-Croat. The main difference was that most Croats, being Roman Catholic, used the Latin alphabet, while most Serbs, being Eastern Orthodox, used the cyrillic one. Tudjman, however, tried to create a separate Croatian language that was as distinct from this as was possible. And, in some cases Tudjman himself would even invent the words. Back.
Note 22: The term "Transylvanian Hungarians" is used here to describe Romania's Hungarian minority even though Hungarians live elsewhere in other regions of Romania. Back.
Note *: Paper prepared as draft of Chapter VI for "New Perspectives on Security and Identity in Europe", Adrian Hyde-Price & Lisbeth Aggestam (eds.) Back.
Note **: Paul Roe, Department of International Politics, University of Wales, Aberystwyth, and Guest Researcher, Copenhagen Peace Research Institute. Back.