What causes ethnic conflicts? 3 What causes ethnic conflicts? The province of Bougainville in Papua New Guinea comprises two large islands—Buka and Bougainville—as well as a number of smaller groups of islands, and hosts a population of a little more than 150,000 people. It is one of the country's most remote provinces, being located some 600 miles from Port Moresby, the capital of Papua New Guinea. Although Bougainvilleans can be broadly characterized as Melanesians, unlike the rest of Papua New Guinea's citizens they bear an unmistakable mark of distinction from them in that the colour of their skin is much darker. Easily insulted and discriminated against because of this, Bougainvilleans share this trait with the inhabitants of the Solomon Islands, a country barely 15 miles to the south of Bougainville. Although this suggests a degree of distinctiveness that sets Bougainvilleans apart from the rest of Papua New Guineans, it is not the only reason for the ethnic conflict that escalated into violence in the late 1980s and took a decade to resolve. The communities that make up Papua New Guinea today have only a weak, if any, sense of shared national identity. Made up of some 700 linguistic groups within a population of only about five million people, Papua New Guinea is highly fragmented. In addition to distinct linguistic identities, regional identification also plays a significant role, often cutting across language groups, and vice versa. The reason for this amazing diversity lies in the colonial imposition of territorial structures on disparate communities that had existed separately from each other for centuries. Just as in Africa, the arbitrary nature of the resulting boundaries had its source in the strength and reach of the colonial powers rather than in any broader consideration of the wishes of the indigenous population. Thus, the eastern half of the island of New Guinea was J vided between Germany and Britain in 1885. In the twentieth i jiltury, Australia gradually took over as the administrative power. 1 'le UK transferred its half of the island in 1902, and the Australians tuieupied the remaining portion during the First World War. When l'jpua New Guinea became a trust territory, Australia continued to administer both areas jointly until 1975. Already during the period of the trusteeship, Bougainville's Lconomic, social, and political development was neglected. There was little investment on the part of the Australian government, who even 'outsourced' the establishment of schools to missionaries. Unsurprisingly, Bougainvilleans became increasingly frustrated .ii id calls for the independence of their islands began to emerge in [he early 1960s. Although the demand for independence was not universally and equally shared among all Bougainvilleans—some were happy to settle for autonomy within Papua New Guinea— from the mid-1960s onwards, one particular development acted as a catalyst to unite them into a strong ethno-nationalist movement. This was the arrival of an Australian mining company in 1964, which, after a successful exploration, established the world's largest open-pit copper and gold mine. Shares were split 80:20 between the mining company and the government of Papua New Guinea; the owners of the land on which the mine was built were compensated with only small royalties. Added to the environmental damage from the mine's operation, traditional Bougainvillean cultural and social life was also disrupted by settlers from mainland Papua New Guinea who were brought in as labourers. Further consolidated by the murder of two Bougainvillean civil servants, the ethno-nationalist movement among Bougainvilleans became focused on secession as the only solution to their plight. When attempts by moderate Bougainvillean leaders to establish a far-reaching autonomy regime within Papua New Guinea failed before the country's 59 Ethnic Conflict What causes ethnic conflicts? independence in 1975, Bougainville unilaterally declared its own independence on 1 September 1975, two weeks before the official date set for Papua New Guinea to become an independent state. Neither the government in Port Moresby nor any other government around the world recognized the Bougainvillean declaration of independence. The central government subsequently dissolved the provincial government in Bougainville and began to withhold Bougainville's share of the mining profits. By June 1976 the situation escalated—Bougainvilleans attacked and destroyed government offices on the island; the central government responded by sending in riot police. Further escalation into all-out civil war, however, could be prevented by direct negotiations betweeii Bougainvillean leaders and the central government, leading to the restoration of the provincial government. More than 10 years passed, without any major incident, but also without any substan tial improvements in the situation for the Bougainvilleans-environmental degradation accelerated, tensions between indigenous people and settlers continued, no adequate compensation was paid to Bougainvilleans, and no real investment was made into the islands' development. A new generation of Bougainvilleans was less and less willing to put up with this situation, and became more aggressive in its demands and means. When their representations did not lead anywhere either, they decided to attack the mining company and its employees directly in 1988. The crisis escalated further when, in early 1989, tit-for-tat killings between Bougainvilleans and settlers spiralled out of control and into several days of riots. As the central government brought in troops, Bougainvillean opponents of the mine organized themselves into the so-called Bougainville Revolutionary Army, and a 10-year civil war ensued, killing thousands of civilians. When the conflict in Bougainville had been all but settled in the late 1990s, historical and cultural differences between the ethnic groups in the neighbouring Solomon Islands contributed to the rapid and violent escalation of the ethnic conflict there. The former British protectorate, an island group in the south-west Pacific, north east of Australia, has a population of less than half a million and is, like its northern neighbour, Papua New Guinea, linguistically very diverse. The main island is Guadalcanal with an ethnically rela-lively homogeneous rural population of some 60,000 and about tln additional 50,000 people of diverse ethnic backgrounds living in the country's capital, Honiara, also located on Guadalcanal. The more populous island, with over 120,000 residents, is Malaita. Friction between the different groups began to arise during and after 'he Second World War, when Malaitans proved more flexible in making use of the economic opportunities offered by an American military presence and the subsequent British development of the country's capital. However, as long as the Solomon Islands remained a British protectorate, tensions between the indigenous groups paled in comparison to the misgivings that all Solomon Islanders had about the British administration of the islands. Independence in 1978 gradually began to change this and tensions started building up, primarily between Malaitans, the economically and politic-,illy dominant majority, and rural inhabitants of Guadalcanal who resented the Malaitans' success and accused them of discrimination .mil exploitation. Leaders of indigenous groups on Guadalcanal complained, not unjustly, about illegally appropriated land, excessive migration of Malaitans to Guadalcanal, the impunity with which Malaitans were able to commit crimes, including murder, against the indigenous population of Guadalcanal, and the unfair distribution of the economic wealth generated on Guadalcanal. The inability and unwillingness of successive Malaitan-dominated governments to address these issues led to an escalation of tensions into inter-ethnic violence in the late 1990s. Loosely organized armed gangs of indigenous residents of Guadalcanal started a campaign of terror against settlers from other islands of the country, killing and maiming dozens of them, and driving some 20,000 from the land on which they had lived for a generation or more, forcing them to flee to the capital Honiara. The tactics used by these armed gangs are typical for many similar conflicts elsewhere: burning down homes, killing livestock, looting, and widespread intimidation. An additional feature of the escalating conflict in the Solomon 60 61 Ethnic Conflict Islands was the impact that the conflict in neighbouring Bougainville had. On the one hand, refugees from the violence there sought and found refuge on Guadalcanal, thus preparing the ground foi Bougainville militants to use this part of the Solomon Islands as a safe haven. In exchange, they provided training to local militants and helped them to recover weapons left after the Second World War as well as to acquire new ones in raids on police stations. On the other hand, cross-border incursions by the defence forces ot Papua New Guinea prompted the government of the Solomon Islands to buy sophisticated military technology, including high-powered firearms, to improve its border security. Yet these guns and other military hardware were later used by Malaitan regular anil irregular forces in raids on civilians and militants in Guadalcanal. In addition, the course of the conflict in Bougainville, especially when seen through the prism of militants from the northern neighbour, provided ample 'proof for militants in Guadalcanal that central governments could be successfully challenged, and that violence was an effective and legitimate means to do so. Thus, militants from Guadalcanal organized themselves into the Guadalcanal Revolutionary Army—an obvious tribute to the Bougainville Revolutionary Army—and the Isatabu Freedom Fighters. Both groups later renamed themselves the Guadalcanal Liberation Front and Isatabu Freedom Movement and began to coordinal e their activities more closely. The Guadalcanal militants entered the stage in 1998 as loosely organized cells without a unified command structure, but soon controlled most parts of rural Guadalcanal. Most of these groups abided by the terms of a New Zealaru. brokered peace accord of October 2000, except for the Guadalcanal Liberation Front who continued fighting until August 2003 and were forced to surrender only after the terror regime that they established in the southern part of the island turned the indigenous villagers whom they claimed to represent and former allies: from among the Isatabu Freedom Fighters against them. The insurgents faced the government's paramilitary police and a quasi-elite squad, the Rapid Response Unit, later known as th( Special Task and Rescue division, as well as groups of so-called What causes ethnic conflicts? ' special Constables, Similar to many other conflicts of this kind, ! R-i^ular security forces of the government and pro-government Malaitan militias cooperated to such an extent that they had 1 iccome indistinguishable in their aims and means by the time they uMkially joined forces in the Malaita Eagle Force, which was soon I to become notorious for its human rights violations. Emerging rela-I lively late in the conflict in early 2000, the Malaita Eagle Force drew ivi ruits and support from Malaitan settlers on Guadalcanal, many i if whom had been displaced by Guadalcanal militants. They also enjoyed backing from former and active Malaitan members of the government's security services and from officers who were unhappy with government policy in the conflict. Succeeding with a coup against their own government in June 2000, they launched Operation Eagle Storm aimed at eliminating all Guadalcanal insurgent groups. Fighting continued for three more years, killing several hundred people, displacing thousands more, and devastating the country's economy and public services. The combination of creed, greed, and grievances, exacerbated by poor leadership, particularistic interests, and spill-overs from conflicts in neighbouring countries, that led to conflict and violent escalation in Bougainville and the Solomon Islands is not unique. \ Discrimination, the persistent violation of human rights, and deliberate economic and social neglect, on the one hand, and the j ability of leaders of the affected communities to capitalize on their resulting grievances, forge and appeal to their ethnic distinctiveness, and compensate some or many with actual and promised material rewards, on the other, are essential in understanding the causes of ethnic conflicts. In Nigeria, for example, ethnic conflict is allegedly about the control of the country's vast oil reserves. In other parts of Africa, diamonds are a major source and prize of ethnic civil war. Kosovo has neither oil nor diamonds. The Albanian population there smuggled for their right to self-determination, to be free from Serbian oppression, to have their own schools and universities in which students can study in Albanian, in short not to feel like second-class citizens in their ancestral homeland in which they 62 63 Ethnic Conflict What causes ethnic conflicts? outnumbered their oppressors by a ratio of nine to one by the late 1990s. In the Israeli-Palestinian conflict, the main conflict Issut is statehood—the demand of Palestinians for their own state and the demand of Israelis to have their existing state secure from a constant threat of annihilation by fundamentalist extremists. In Northern Ireland, Catholics and Protestants are almost equal in number; the province is part of one of the most highly developed industrial countries in the world and part of the country thaL 'invented' democracy. Compared with many poverty-stricken and war-ravaged societies, the conflict in Northern Ireland should havr been resolved long ago without costing the lives of over 3,000 people over a 30-year period. Yet the visions that each community has about what a just solution of this conflict would look like are diametrically opposed and virtually incompatible—the same stretch of territory cannot, at the same time, be part of the UK and part of the Republic of Ireland. This very brief survey reveals that, although the stakes in what is commonly referred to as ethnic conflict are extremely diverse, in each of these and other cases it is organized ethnic groups thai confront each other—minorities and majorities—with and without the backing of state institutions and with or without external support. What we therefore need to do is try to bring order to this 'chaos' of diverse claims and interests, examine to what extent ethnic conflicts actually are about ethnicity, and explore to what extent ethnicity is merely a convenient common denominator to organize a conflict group in the struggle over resources, land, or power. In other words, we need to determine what causes ethnic conflicts. In the preceding chapter we explored the nature of ethnicity and nationalism and found that, although they are facts of life almost everywhere, the mere fact of people having an ethnic identity and the presence of nationalist ideologies do not necessarily and inevitably lead to conflict, let alone civil war, ethnic cleansing, and genocide. Moreover, closer examination of many conflicts labelled as 'ethnic' leads some to argue that ethnicity is sometimes no more than a convenient mechanism to organize and mobilize people into homogeneous conflict groups willing to fight each other for resources that are at best indirectly linked to their ethnic identity. In the last months of 2000, a debate was waged in the Northern Ireland Assembly and in the Assembly Commission on whether Foster lilies should be treated on a par with poppies. Easter lilies have been historically one of the most important symbols of the Republican and Nationalist movements. They are closely connected to the 1916 Easter Rising, which stood at the beginning of a chain of events that eventually lead to the partition of Ireland, and have subsequently become a symbol of commemoration for those who have died in the struggle for a free and united Ireland, including IRA terrorists. The poppy, on the other hand, sold by the Royal liritish Legion in memory of the soldiers who fought for, and died in defending, the UK in war time is of particular significance for members of the Unionist and Loyalist communities because it i-mphasises the sacrifices made by them contributing to the British war efforts and is for them a reminder of the strong links between Northern Ireland and Great Britain. Yet, both communities not i >nly value their own symbols very highly as intricate parts of their identities, but also take exception and offence to the other's symbol because they see it as a reminder of either oppression or of terrorist violence, respectively, inflicted upon them. Thus, proposals for outright parity between the two communities' symbols—initially manifested in a suggestion to allow the sale of Easter lilies alongside poppies in the Assembly buildings—were rejected outright by Unionists. When they were watered down to a compromise solution, to the effect that Easter lilies should become part of the flower arrangements during the Easter Recess of the Northern Ireland Assembly and approved in the Assembly Commission, some Unionists viewed the matter as so urgent that they invoked a special provision under the 1998 Agreement, according to which the Assembly had to be recalled just before Easter to discuss the matter. The irony of holding this debate on the third anniversary of the Agreement was certainly not lost on the former Lord Mayor of Belfast, Alban Maginness of the Nationalist Social Democrat and Labour Party (SDLP), who wondered 'what sort of lunacy has 64 65 Ethnic Conflict E | i What causes ethnic conflicts? descended upon the Assembly that it has to be urgently reconvened over a bowl of lilies'. Of course, many conflicts are about much more than sin h s\ iti-bolic issues—they are about the physical security of memlvis nt a particular group and their right to live in peace and freedom \iieu : 1 groups in border areas and ethnic groups that straddle mteni.t i \ tional boundaries). Political factors that need to be present for Kli- i y"1 nic conflict to erupt include discriminatory political institut ii ins ! i (which pursue policies that deliberately disadvantage inemln rs j of particular ethnic groups), exclusionary nationalist ideologii •, i (that question, for example, the equality of all citizens, deny i m- 1 zenship to members of particular ethnic groups, or advocate fo [ cultural assimilation of minorities), and contentious inter-grcuji I and intra-group politics at mass and elite levels (in other words political sphere that is mostly carved up between ethnic pi i ■ I it-s | that compete with each other on ethnic platforms). Economic ajul social factors play an equally important role in setting the scene tc >i potentially violent ethnic conflicts: economic problems, . saw a change in their share in power and influence but bee i nt increasingly subjected to culturally discriminatory practices. muIi as those embodied in the Sinhala Only Act of 1956, according l(i which Sinhala replaced English as the official language for government and business, and the introduction in the early 1970s n| communal quotas for university entrance that also discriiiim.-iiwl against Tamils. The new constitution in 1972, changing the conn try's name from Ceylon to Sri Lanka, rejected the Tamil demand for internal self-determination and granted Buddhism the foremost place in the life of the state. Over time, political and cultural d's-crimination of Tamils also significantly reduced their education.il and employment opportunities. With even the dimmest hopes of an eventual federalization of Sri Lanka disappearing, th> r.miil community became increasingly radicalized, but remained nonviolent until a hard-line government in Colombo resorted to Lhe use of force to suppress peaceful protests in the Tamil-dominatLJ Jaffna peninsula. This transformed tensions between the Sinhale-n' majority and Tamil separatists into a violent ethnic conflict in tl -.i-mid-1980s in which thousands have died. As a consequence of the civil war that is still ongoing despite negotiations, facilitated by Norway, between the Sri Lankan government and the Tamil Tigers, and of the continuing and increasing 'Sinhalization' of Sri Lanka, the economic situation on the island has also deteriorated for both communities, albeit disproportionately affecting the Tamils. This brief account of the situation in Sri Lanka highlights the importance of another dimension of the account of the causes of ethnic conflict. Apart from underlying conditions, such as political, economic, and cultural discrimination, a second set of causes is required to 'enable' the outbreak of violent ethnic conflicts. labelled 'proximate causes', these are factors that increase the likelihood of conflict in a situation in which all or some of the underlying 'ingredients' are present.5 In this category of proximate causes, a distinction can be made between internal and external tjctors that operate at the mass level and the elite level. Internal elite-level causes are bad leaders and external ones are bad neighbours; internal mass-level causes are serious domestic problems and external ones are bad neighbourhoods. In the Sri Lankan case, the influence of bad leaders is the most significant, whilst external factors are almost completely absent. The stance that the Tamil Tigers nave taken on the legitimacy of using violence against the central government and on policing their own community, and the refusal, for two decades, to accept negotiations unless the government recognizes the right of the Tamils to self-determination and implicitly to establish their own independent state on part of the island, sat uncomfortably with a central government which for a longtime had refused to recognize Tamil claims, repeatedly reneged on promises and commitments, and often used excessive force, including against civilians. Although there were also some internal mass-level factors that contributed, if not to the outbreak of the conflict, at least to its prolongation (such as the revival of radical Buddhism among Sinhalese and the desperation of young Tamils in a situation in which living standards decreased constantly and positive prospects for the future, including any employment opportunities, were almost completely absent), it is fair to say that the choices that the leaders in both communities made were decisive factors in creating a situation in which, even after 20 years of fighting, there are few prospects for sustainable peace. This view of the causes of ethnic conflict and the distinction between underlying and proximate causes also enable an explanation of why, despite similar basic conditions, not every situation of ethnic tensions leads to full-scale civil war. Take the case of the former Yugoslavia. Why did the Serb-dominated and controlled Yugoslav army not put up a big fight over Slovenia? One explanation is that the absence of a significant Serb minority in the republic greatly diminished Serbian appetite for a fight, whereas in 70 71 Ethnic Conflict Croatia, Bosnia and I Itrzegovina, and Kosovo not only were tin u-Serb minorities that wanted the continued protection of Belgrade, arid had militarily mobilized for their own 'self-defence' but, unhkr Slovenia, these areas were also important for the projw r .>| a Greater Serbia and Serbian public opinion could more easily |v swayed to support the war effort there. Undoubtedly, the Mill im-v.i regime in Serbia proper was not the only one to blame. A sin ]ln\\ virulent form of ethnonationalism emerged in Croatia in tin- Ijtr. 1980s/eariy 1990s. With a nationalist government under Ham,. Tudjman that openly embraced the symbols of the wai time Ustasha regime, which had engaged in mass killings of Serb i i\ il-ians, local Serbs had every reason to fear discrimination and vunv at the hands of the new rulers, yet their own radicalization miiiv contributed to the escalating spiral of violence. Bosnia and Herzegovina suffered from very similar bad leade.i and bad neighbour syndromes. Wedged between Croatia and Sei In.' and large ethnic Croat and Serb minorities in the emerging country, Bosnian Muslims at times fought a two-front war against both j Greater Croatia and a Greater Serbia policy, endorsed and actively supported by foes inside and outside the country. Little wonder that the third ethnonationalist movement in the region, that for a Greater Albania, or, as some would not wrongly argue, for a G n ilci Kosovo, was soon to clash with the rival ideology for a Gu.ilir Serbia. Yet, what is interesting about the case of Kosovo is thai independence was, for almost two decades, pursued by non-vr iU nt means alone, and only from about 1996 onwards did the conIIid between Serbs and ethnic Albanians escalate into violence, precisely because the Milosevic regime saw an opportunity in mani ■ pulating the Kosovo issue such that it could strengthen its grip on power in Belgrade. Had it not been for the decisive military intervention led by NATO, this strategy may well have worked.6 A final point in support of an argument that places special emphasis on the role of political elites is the case of Macedonia. Situated in a bad neighbourhood comparable only to the Great Lakes Region in central Africa and the Caucasus, with several 'bad neighbours' around denying the country's choice of name (Greece), ■HHHBlih:': ^ 48h*~ "-- What causes ethnic conflicts? I ,\k ethno-cultural distinctiveness (Bulgaria), and questioning its - hinders and ultimately its statehood (Kosovo), these gloomy Hjdiirrial conditions were easily matched by internal ones including liigh unemployment, political and cultural discrimination, and an I LthFiic geography according to which a partition of the country I umld have been executed without great problems. Yet, despite 5 sji. H [comings and mistakes that led to significant, albeit brief, iiilcr-ethnic violence in 2001, reasonable and responsible leaders, 1 ..up potted by NATO and EU facilitation, prevailed and negotiated jn agreement in the summer of 2001 that saw an end to violence jm.' the beginning of a reform process; this process has been slow to u.it'e, but has nevertheless progressed and begun to address some of the underlying causes that enabled the violent escalation of the ci mflict. Yet, although leaders' choices may explain particular outcomes— prolonged war in Bosnia and Herzegovina compared with a brief spell of violence in Macedonia—one part in the explanation of curses of ethnic conflicts remains missing, and this is why some L aders choose war over peace, confrontation over accommodation. These choices, of course, remain meaningless unless leaders can mobilize sufficient numbers of followers for the one or other course of action. In order to understand the motivations of leaders and followers better, it is worthwhile considering a particular rational choice explanation of ethnic conflict that focuses on the so-called security dilemma and assumes that individuals always make rational choices, weighing the costs and benefits of their actions. David Lake and Donald Rothchild, for example, argue that 'the fundamental causes of ethnic conflict' are 'strategic interactions between and within groups'.7 Within the realm of strategic interactions between groups, 'three strategic dilemmas can cause violence to erupt: information failures, problems of credible commitment, and incentives to use force preemptively (also known as the security dilemma)'.8 As for within-group strategic interactions, 'ethnic activists and political entrepreneurs may make blatant ethnic appeals and attempt to outbid moderate politicians, thereby mobilizing members, polarizing society, and magnifying 72 73 Ethnic Conflict the intergroup dilemmas'.9 Often exacerbated by proximate fat tors such as primordial identities, economic decline, regime ih.:i'5'in a lastm8 stalemate, but sooner or later they will need to settled, and this will be possible only by way of agreement In-nyeen both states involved—a point vividly illustrated by the .'. i"i>9 South Tyrol settlement (involving Austria and Italy) and i tin- 1^98 Agreement on Northern Ireland (involving the UK and " | I he Republic of Ireland); the way to the latter was partly paved by tarlier 1985 Anglo-Irish Agreement which did not, however, dir-'I cctly involve any of the political parties and paramilitary groups in i; Noi thern Ireland itself. It is also important to bear in mind that the iik rc prospect of cooperation between host state and patron state i, not automatically greeted with great enthusiasm by their local rr>i|:cges: Unionists in Northern Ireland cried foul in 1985 (and some even continued to do so after 1998), and characterized the deal struck between the governments in London and Dublin as a sell-out. Loyalists committed more acts of violence against t .jt holies, and sporadically against British security force. So too did the IRA, empowered by an arms shipment from Libya. Arguably, both sides, among other things, feared that agreement between the two governments would permanently prevent them from achieving their political goals—a united Ireland and no influence from Dublin in Northern Irish affairs. Implicitly or explicitly, the notion of the security dilemma is thus at the core of a wide range of causal explanations of ethnic conflicts. The strategic, pre-emptive use of violence is generally thought to be more likely in conditions of emerging anarchy which heighten the uncertainty of identity groups about their future (phy-■sical or cultural) survival. Yet, these fears do not always and necessarily translate into actual violent conflict. In other words, there is no automatism that leads from an existing security dilemma directly into a violent ethnic conflict. To establish a relationship between an existing security dilemma and the violent escalation of a conflict, it is therefore once again useful to look at the decisions that individuals take in response to a specific security dilemma— masses that push their elites in the direction of a particular course of action; or leaders who choose a particular policy and the followers | who are willingly led down a specific path. 74 75 Ethnic Conflict What causes ethnic conflicts? One way to describe the essence of the security dilemma i\ h.ii an increase in one person's, group's, or state's security is at the saint time perceived as a threat, or decrease in security, by another actor. My acquisition of a baseball bat to protect my property from burg, lars decreases the physical security of the burglar to go about his 'business' and is likely to give him incentive to carry a siniilai ut more powerful weapon as well, which in turn decreases my oivn security when defending my property, and may prompt me to up the stakes by acquiring a firearm, and so on. The same dynamics, albeit at different scales, work at the level of groups and u.ite* What agents taking this course of action—as individuals, groups n states at all three levels—have in common is a strong belii-l thai ,i condition of anarchy prevails in which they cannot but relj us, themselves to ensure their security. I might not trust the police to protect me from burglars, ethnic groups may not be confident Jut state institutions are sufficiently impartial and effective in enforcing law and order, and states will possibly have little i"iiih in the power of the organized international community to defend them against predatory neighbours. As a result of this reli.intv mi oneself, not only are there incentives to escalate an arms race but also such incentives exist with regard to the pre-emptive use ol violence. What if my opponent in the future acquires a weapon so powerful and expensive that I cannot match it? Would I not be better off to attack now when I am still in a position of i\ laiiw advantage, thus thwarting any future attacks that might catch me off-guard or under-prepared? Following such logic of consequentiality is an all too f.in:!liiir pattern in the escalation of ethnic conflicts. What distinguishes an ethnic security dilemma from 'traditional' interstate security dilemmas is that it is based on a more comprehensive view of what security is. The ethnic security dilemma thus involves elements of physical security (literal survival of group members), political security (freedom from oppressive regimes that exclude group members from meaningful participation in the political process), economic and social security (freedom from economically and socially exclusive regimes that deny equal opportunities of economic and 76 ..n. >al advancement to group members), cultural security (freedom i\iiirt forced assimilation), and environmental security (freedom !~ from-environmental destruction and resource scarcity). Thus, when [m .il Serbs took up arms in Croatia and in Bosnia and Herzegovina, Ihev did it in part because they feared that their relative advantages o\.'i their opponents would diminish over time, leading to their n.iysical, political, economic and social, and cultural security as members of a particular ethnic group coming under threat. The \i-)lent suppression of peaceful demonstrations in Sri Lanka and i Kosovo in the 1980s followed a similar logic by state actors: crush-i n'.; separatist movements before they gain greater momentum and intimidating those who might join them if the state did not send j clear message. What often additionally drives state actors is the i nation of a potentially detrimental precedent: without the harsh upression of insurgents in East Timor or Aceh, in Chechnya, in k. >sovo, or in Biafra, others might follow suit and the result would In- the disintegration of the existing state. In other words, from the pi rspective of the government, the physical security of the state and its institutions is at stake. Even though many of the predictions that state actors make in attempts to justify their violent crack-downs may prove wrong with the benefit of hindsight, it is the perceptions of an existing security dilemma that determines the course of action. Intercom-munal violence in Northern Ireland in the late 1960s and early 1970s involved many incidents that in today's terminology would be called ethnic cleansing. Yet the creation of ethnically homogeneous neighbourhoods there, the forced displacement of Greek and Turkish Cypriots in 1974 on Cyprus, as well as earlier cases of expulsion and forced resettlement in the period immediately after the Second World War in east central Europe were all driven by the very same security concerns: in situations where only the group | can guarantee above all physical and cultural security, an ethnically homogeneous neighbourhood, region, or state is easier to ; defend against threats. This logic, however, forgets that everyone j living in such ethnically homogeneous areas is also more vulner-I able to attack, because violence aimed at another ethnic group can i 1 77 Ethnic Conflict What causes ethnic conflicts? be executed much more easily if they can be readily identified according to where they live. This also extends to other, non-physical aspects of security: economic and social advancement in mainstream society may be impossible for someone carrying the stigma of living in a certain neighbourhood or region that identifies him or her as belonging to a specific ethnic group; likewise, environmental security may be under threat if only one group is; likely to be affected negatively by the consequences of an irrigation project, deforestation, etc. Although the security dilemma is no doubt a very useful tool to explain the emergence of violence in ethnic conflicts, it is important to bear in mind that there are two other strategic dilemmas that can contribute to the violent escalation of ethnic conflict. Apart from the security dilemma, information failures and problems of credible commitment are important for our understanding of how and why leaders are able to mobilize their followers and radicalize them to the point where the (pre-emptive and indiscriminate) use of violence against members of a different ethnic group becomes a legitimate strategy for group survival. For leaders to succeed in mobilizing and radicalizing their followers in this way, followers need to believe in the reality of the threat to their security that leaders claim. In other words, there needs to be enough 'credible evidence' about the intentions of rival groups to execute threats for ordinary people not only to condone but also to participate actively in committing acts of violence. This evidence, however, cannot easily, or can only to a certain extent, be manufactured by leaders of one group without the 'help' of other groups. Again, the example of the wars in former Yugoslavia is instructive. Had it not been for Croat president Tudjman's embrace of the symbols of fascist Croatia and his virulent ethnonationalism, demanding independent statehood for Croatia, many local Serbs may have trusted in the ability of Yugoslav institutions to protect them, or at least in the sustainability of the multi-ethnic peace that had alter all prevailed throughout the post-1945 period, and for the most part also for centuries before the beginning of the Second World War. Thus, whilst the 'problematic group history' is reduced to a single episode in history, the fear among Serbs in Croatia and Serbia proper, combined with ambiguous signals from Croatia, enabled Milosevic to hype up fears among Serbs to such an extent that local insurgency and all-out war were seen as the best options to guarantee the group's security. Uncertainty among Serbs about the real intentions of Croatia's leaders—that is, a failure to communicate them—and vice versa meant that eventually both sides felt justified in the use of violence. In other words, for security fears to lead to violent ethnic conflict it takes leaders of both groups to 'cooperate'.12 That such escalating spirals of fear can occur in majorities and minorities also becomes obvious if we consider ethnic conflicts in the post-Soviet periphery. In Nagorno-Karabakh, Armenian fears of being subject to genocide (grounded in a traumatic episode in Armenian history towards the end of the Ottoman Empire) and Azerbaijanian fears for the territorial integrity, and thus very much for the existence of their state, clashed during the break-up of the Soviet Union.13 For both Armenians and Azeris, control over Nagorno-Karabakh became symbolic and existential for their survival. With each uncertain about the other's intentions and left 'alone' to resolve their problems after the dissolution of the Soviet Union, fears of ethnic violence turned into self-fulfilling prophecies. Credible commitment problems—the third of the strategic dilemmas—manifest themselves in a somewhat different, albeit not unrelated, manner. Take the example of Northern Ireland since 1998. In the Agreement concluded that year, the British and Irish governments, and all but one of the mainstream political parties in Northern Ireland, committed themselves to a new system of government in which local parties representing both communities would exercise a wide range of powers independently of the government in London. As a result of the bonds that one part of the population—varyingly referred to as Nationalists, Republicans, or Catholics—felt with the Republic of Ireland, arrangements were also put in place for extensive cross-border cooperation between the government in Northern Ireland and its counterpart in Dublin. This 78 79 Ethnic Conflict What causes ethnic conflicts? Agreement meant significant concessions from both communities: Unionists, always keen to preserve Northern Ireland's links with Great Britain in the UK, accepted power sharing with Nationalists and Republicans, who, in turn, gave up their unqualified demands for reunification with the Republic of Ireland and acknowledged that this could be brought about only through a referendum in Northern Ireland in which a majority of the population voted in favour of such a move. With a 30-year history of violent conflict, costing over 3,000 lives and injuring many thousands more, both communities had quite different ideas about what the Agreement actually meant—in other words what they had committed themselves to. For Unionists, sharing power with Nationalists and Republicans was the price to pay for peace—that is, for the decommissioning of all weapons by paramilitary organizations, especially the IRA. This link, famously expressed in the slogan 'no guns, no government' was, however, only vaguely and ambiguously estab^ lished in the Agreement itself, which noted that all parties would use their best efforts to bring about paramilitary decommissioning within two years of the conclusion of the Agreement. Hence, any subsequent interpretation of the Agreement by Unionists legitimizing their refusal to share power with Sinn Fein, who are considered to be closely connected to the IRA, was based on Unionists' fears that Republicans were not seriously committed to peace because of their refusal to decommission their weapons. This, in turn, was seen by Republicans as 'clear evidence' that Unionists, too, were not committed to the Agreement, and especially not to sharing power with them, and used the decommissioning issue only as a smoke screen behind which they could hide their real intentions. Locally referred to as 'the blame game', this debate has raged on for over six years now. The UK government had to step in several times and prevent the collapse of the entire process by temporarily suspending Northern Ireland's government institutions. The latest suspension occurred in October 2002 over allegations of continued IRA intelligence gathering, which prompted Unionists once again to walk out of government. However, unlike on previous occasions, no deal was possible between Unionists and Republicans that could have enabled the restoration of government in Belfast. Subsequently, elections were held in November 2003, returning hardliners in both communities as the strongest parties in the new assembly. By December 2004, despite intensive negotiations, no agreement had been reached on a new government, and both sides continued blaming each other, in public at least, for the lack of progress and doubting the sincerity of their opponent's commitment to peace. Alleged IRA involvement in the UK's largest ever bank heist in Northern Ireland, and the murder of a Catholic man in Belfast by senior IRA members, did little to assuage Unionist fears of. continued Republican paramilitary activity. Likewise, the IRA's publicly made offer to kill those who had committed the murder did not enhance the democratic credentials of the organization or portray it as a group with a firm grasp of the notion of the rule of law. Yet, a short time after this remarkable demonstration of lack of judgement, Sinn Fein President Gerry Adams called on the IRA, in a press conference on 7 April 2005, to abandon their armed struggle for good and join a peaceful and democratic campaign for a united Ireland, not, of course, without praising the IRA for holding 'the line and fac[ing] down a huge military foe, the British Crown Forces and their surrogates in the unionist death squads'.14 Although Northern Ireland is not a unique case in this respect, it is distinct from many other cases in that it stands on one end of a spectrum of cases in which peace agreements that were successfully concluded, and often received international praise, subsequently faltered when it came to their implementation and operation. In Northern Ireland there was no major resurgence of violence, whilst the collapse of the Oslo Accords between Israel and the PLO was followed by the second, and much more devastating, Intifada. At the most extreme, the breakdown of the Arusha power-sharing pact i n Rwanda between Hutu and Tutsi led to the worst act of genocide since the Holocaust. It would obviously be wrong to blame only the lack of credible commitment on the part of the conflict parties for {his, but doubts about how serious a former deadly enemy is about his commitment to sticking to a peace deal feed seamlessly into a renewed security dilemma that provides elites with a rationale, and 80 81 Ethnic Conflict masses with incentives, to resume violence for fear of being eventually worse off if they gave their opponents the benefit of the doubt. This is, unfortunately, too frequent a phenomenon to be brushed off lightly. In a study of almost 50 civil wars that began between 1940 and 1992 (albeit with only a small minority among them being ethnic conflicts), it was found that 'negotiated settlements that appear to have all the elements of success—a cease-fire agreement, specific arrangements for future governance, resolution of underlying issues—still fail if they lack the guarantees necessary to reassure groups to proceed with implementation'.15 A final important point to be made in this discussion of the causes of ethnic conflict is related to the observation that strategic dilemmas, such as those related to information failures, credible commitment, and ultimately group security, occur not only in the interactions between groups but also within them. Moderate Hutu in Rwanda were prepared to share power with Tutsi; moderate Nationalists and Unionists shared power successfully in Northern Ireland in 1974, albeit for only three months; within Israel, there are severe policy disagreements between Likud and Labour on how best to secure the future of Israel; not all Chechens favour violence and terrorism as the most suitable approach to resolve differences with Russia; in Spain's Basque country, almost every terrorist atrocity committed by ETA is followed by large public protests of moderate Basques speaking out against the use of violence; and many ordinary Tamils suffer more from the oppression that they experience at the hands of their 'own' Tamil Tiger 'liberation movement' than from counter-insurgency measures by the government in Colombo. How is this relevant as a cause of ethnic conflicts? Obviously, intra-group competition, or even rivalry, is hardly a cause of com flict itself. At best, it would be a proximate cause, that is, a factor that makes conflict more likely in the presence of other conditions. But even this is not necessarily self-evident and it takes specific individuals to act as catalysts in making ethnic conflict more likely.16 So-called ethnic activists, who individually feel a strong need of identification with members of their own ethnic group, can What causes ethnic conflicts? create significant and often self-intensifying and self-sustaining sncial pressures that lead to the polarization of societies and the formation of conflict groups based on their members' ethnicity. These pressures, obviously, work all the more effectively the more 11 sources in a given society are accessible only on the basis of group i-iembership. The more social life centres on activities of one's own kin group—the bars you go to, the civic organizations you join, the neighbourhoods you live in—the more so-called cross-cutting i leavages become eroded. What may begin in the purely social and n: ivate spheres soon spreads to the economic sphere—whom you work for, whom you hire for your business—and ultimately to the ■ iclitical—which party you j oin, which candidate you vote for. Very often, these processes are not sequential, but occur in parallel and .11 great speed. They produce and are the products of genuine fears .hat people have about their security now and in the future, and do n- it exist in a vacuum. Nor do they inevitably lead to ethnic conflict. Strongly polarized societies, such as Belgium and to a lesser V'\.tent Switzerland, have existed for long periods of time without inter-ethnic violence. Even in societies as deeply fragmented as Northern Ireland, Sri Lanka, or India, violence is not an inevitable or permanent feature of life. A second catalyst is, therefore, often necessary to exacerbate volatile situations in such a way that security dilemmas become so acute that the pre-emptive use of violence seems the only way to secure the survival of the group. This catalyst is the political entrepreneur who, although not necessarily sharing the same emotional need for ethnic identification as the ethnic activist, has no hesitation in utilizing and manipulating ethnic identity and the social splits that it can create as a powerful tool to mobilize people, and thus create the 'human resources' deemed necessary for pursuit of political power.17 Take the example of Slobodan Milosevic—he effortlessly managed the transformation from ordinary communist apparatchik to nationalist Serbian leader once the opportunity structures for the pursuit of power radically changed in the crumbling communist regime in Yugoslavia in the late 1980s. Milosevic's shift from communism to nationalism as his powerbase 82 83 Ethnic Conflict What causes ethnic conflicts? fundamentally changed the playing field not only within Yugoslavia but also in Serbia itself. The links that Milosevic created between the survival of the Serb nation and events in Croatia, Bosnia and Herzegovina, and ultimately Kosovo, made it all but impossible for moderate leaders to make their voices heard, let alone attract public support for different, less nationalist political agendas. This was particularly obvious during the build-up to the Kosovo conflict. Milosevic effectively managed to present himself as the guardian of the Serb nation and won significant majorities for himself and his party in relatively free elections. Moreover, even the so-called democratic opposition did not dare to challenge him on his Kosovo policies out of fear of being marginalized in the electoral contest dominated by hard-line nationalist rhetoric. Yugoslavia is not unique in this context. This kind of ethno-nationalist mobilization and the rise of hard-line, often uncompromising, and radical leaders are phenomena that can be witnessed in many other, not only ethnic, conflicts across the world. They occur in contexts of heightened insecurity and ongoing or impending regime change, but again not inevitably and not forever. The rise of Ian Paisley and his Democratic Unionists in Northern Ireland to become the strongest political party after the November 2003 elections, the electoral victory of Ariel Sharon in Israel in 2001, the political and military power amassed very quickly by the Shia cleric Moqtada al-Sadr, the dominance of Turkish Cypriot politics by Rauf Denktash for more than three decades, and the showdown between a more compromising prime minister and a more hard-line president in Sri Lanka and the perseverance of the latter and her party in elections in 2004 are all examples of these phenomena. This process of ethnic outbidding in which leaders of different political parties within one ethnic group attempt to present themselves as the only safeguards against their ethnic group being 'sold down the river' by their irresponsible rivals complements the efforts of ethnic activists. Together, ethnic activists and political entrepreneurs increase the homogeneity of existing ethnic groups and the polarization of society as a whole, thus making it more likely that tensions between groups will escalate into violence, that conflict will continue without hope of resolution, and that agreements will fail to live up to their promises. The security dilemma and related dilemmas focusing on credible commitment problems and information failures offer a powerful explanation of the dynamics leading up to the violent escalation of ethnic conflicts. Yet not all rational choices that individuals and groups make are related to their own physical and cultural security. A number of ethnic conflicts also suggest that economic motives—greed and the opportunity to pursue greed-driven agendas—are often equally if not more significant causes, which help us understand the occurrence of ethnic conflict. Since the late 1990s, a number of scholars have tried to expand the debate on the causes of ethnic conflicts beyond the traditional grievance paradigm. In the context of debates about so-called 'new' wars, arguments about economically motivated violence gained and, to some extent, maintain significant currency. Paul Collier and Anke Hoeffler of the World Bank, for example, argue that ethnic conflict occurs if the incentive for rebellion is sufficiently large relative to its costs.18 They maintain that the incentive for rebellion is the product of the probability of victory (the ability of the government to defeat rebels) and its consequences (the benefits of state capture or secession). Costs are seen as a combination of the opportunity cost of rebel labour (that is, loss of earnings in high-income countries makes rebellions there unlikely) and the disruption to economic activity caused by warfare (which makes both state capture and secession less attractive because it reduces the potential benefits to be reaped by the rebels). In their original 1998 study, Collier and Hoeffler found that civil war was overwhelmingly a phenomenon of low-income countries, that natural resources made things worse, unless there were plenty of them, that countries with laTger populations have higher risks of war and that these wars last longer, and that it is not ethno-linguistic fractionalization as such that is damaging to societies but that degree of fractionalization that most facilitates rebel coordination (for example, very different language, culture, physical appearance, as well as relatively compact 84 85 Ethnic Conflict What causes ethnic conflicts? settlement areas of the groups rebelling). Several years later, Collier and Hoeffler offered a more advanced explanation of how greed affects the likelihood of ethnic conflict breaking out: exceptional opportunities for the rebel movement.19 Indicators for the existence of such exceptional opportunities were seen in the availability of financing for a rebellion (extortion of natural resources, donations from diasporas, subventions from hostile governments) and in opportunities arising from atypically low costs (low per capita income, low rate of male secondary schooling, and low growth rates of the economy). Collier and Hoeffler also investigated the relevance of a range of other opportunities, such as the availability and cheapness of military equipment, military skills among the population at large, weak governments, and a terrain favourable to rebels (such as forests and mountains). Their findings again emphasize the importance of economic factors: the presence of natural resources that can be used in financing rebel movements, diaspora support and low costs in terms of foregone earnings, and the availability of conflict-specific capital (equipment and skills) were the most significant factors. The only significant non-economic factor that they found was ethnic dominance: where one ethnic group has monopolized the state and its resources whilst others remain excluded from meaningful political, economic, and social participation. Under such conditions, the risk of conflict nearly doubled. Clearly, in the light of many of the examples used in this book, a purely economic explanation of ethnic conflict does not seem very satisfactory, and so it should not come as a surprise that many scholars have criticized Collier and Hoeffler's work. Among them, Karen Ballentine and Jake Sherman counter Collier and Hoeffler's sweeping generalizations with in-depth case studies.20 They and their contributors find that economic factors are nowhere the sole factor causing outbreak of conflict: in Kosovo, Sri Lanka, and Bougainville, grievances and insecurity bred by systematic exclusion of minorities from political and economic participation had a much larger impact on the escalation towards violent ethnic conflict. In the Democratic Republic of Congo (DRC), however, conflict is clearly more economically driven, although it also has its causes in political misrule, corruption, socioeconomic deterioration, and institutional decay. As far as the impact of economic factors on the duration of ethnic conflicts is concerned, Ballentine and Sherman and their contributors accept that access to lucrative economic resources is a more important factor, but poignantly ask whether self-enrichment is an end in itself or instrumental in funding a rebellion that has its causes in serious socioeconomic or political grievances. Others have criticized Collier and Hoeffler's studies along similar lines. Michael Pugh and Neil Cooper, for example, note that Collier and Hoeffler's selection of cases excludes all anti-colonial insurgencies, in which (political) grievances were by far the most influential factors.21 Like Ballentine and Sherman, they also note that the false consciousness postulated by Collier, stating that 'rebel supporters .--■!.'- are gulled into believing the discourse that self-interested rebel leaders promote'22 may infer motives from actions, but still leaves motives in the realm of speculation, because it remains unclear whether economic agendas are primary causes of conflict—that is, whether people rebel to satisfy greed—or whether they are instrumental in order to help realize political agendas of grievance. " "T ' Importantly, this debate about the causes of ethnic conflict should not be seen in the mutually exclusive way in which it is sometimes presented by advocates of one or another line of thinking. As we see again in Chapter 4, at different levels of our analysis of the causes of ethnic conflicts different factors matter in different ways. Political leaders may be able to mobilize support for their own power games if they can convince their followers that they are in danger of annihilation. If people see concrete economic advantages in pursuing an agenda of systematically excluding members of another ethnic group from opportunities of political, economic, and social participation in a given state, they may willingly buy into a discourse of cultural survival. These are phenomena that are not only linked to areas of ethnic conflict in 'uncivilized' parts of the world; they occur in so-called advanced or post-modern societies that have supposedly moved beyond the narrow confines of 86 87 Ethnic Conflict ethnic nationalism as well. Debates about immigration, asylum fraud, or becoming a cultural minority in one's own homeland are widespread across western Europe, not only but especially at election times. The often emotional responses that people have to such politically and economically highly charged issues once again underline the importance of individuals—leaders and followers alike—and the decisions that they make in determining whether conflict will occur and endure. Yet individuals do not act in a vacuum. Leaders in particular are crucial in the process of catalysing existing tensions into open warfare, but they can do so only if circumstances permit. A difficult economic situation can be talked up into an impending collapse of a country's ability to feed and shelter its citizens, a political impasse in parliament can be presented as a serious crisis, and ambiguous pronouncements by the leaders of other ethnic groups about their intentions can be used to induce fears about the inevitability of violence, but hardly ever does conflict merely 'appear' overnight. An important implication of this build-up towards violent conflict, rather than its sudden eruption, is that although violence may be temporarily prevented or contained by focusing on leaders inside and outside the country and by ameliorating the consequences of serious domestic problems, a conflict itself cannot be resolved without addressing its underlying causes. 88