Escalation of Ethnic Conflict:
A Survey and Assessment

bull2.gif (594 bytes)  Introduction
bull2.gif (594 bytes)  Domestic Attributes of Instability
bull2.gif (594 bytes)  Interactions between Groups
bull2.gif (594 bytes)  International Interactions
bull2.gif (594 bytes)  Conclusion
bull2.gif (594 bytes)  Notes

Escalation of Ethnic Conflict:
A Survey and Assessment

David Carment
Carleton University

Patrick James
Department of Political Science
Iowa State University

(return to top)

Increasing concern in the early 1990s over the negative impact of the end of the Cold War strengthened fears that the break-up of the Soviet Union would unleash a barrage of lethal ethnic wars and state failures. The resulting instability, it was argued, would displace millions of people and create a wake of social upheaval unparalleled in this century. The idea of an impending global crisis became firmly rooted in both the belief that ethnic conflicts were becoming a major source of global insecurity and in the assumption that the international community was ill-equipped to deal with this new source of insecurity.

A cursory glance at the range of disputes, violent or otherwise, supports the impression that ethnic conflicts have been multiplying at an exponential rate. By 1993, for example, there were at least 48 existing or potentially violent conflicts in progress. These included Romania, Mauritania, Rwanda-Burundi, Senegal, Togo, Nigeria, Kenya, Papua New Guinea, Fiji, Algeria, Egypt, China, Bhutan, Brazil, Mexico, India, Kosovo, Albania, Greece, Bulgaria, East Timor, the Republic of Macedonia and Tadjikistan. By 1996, however, the total number of serious conflicts (those with 1000 battle-field fatalities or more) had decreased to pre-Cold War numbers either because of military defeat, negotiated settlements, government reform or some form of concerted third party intervention. The most notable of these successes are the Bosnian, Philippine Moro and Northern Ireland conflicts.

Nevertheless, public concern about the extreme cases - those involving a combination of widespread violence, state failure and spillover to neighbouring states - remains at the forefront in the development of ethnic conflict analysis. For every civilized divorce Czech-style, analysts note, there have been at least two dozen more armed conflicts. In extreme cases ethnic conflicts, violence and spillover appear to be inextricably intertwined and unstoppable. For example, at the height of the slaughter in Rwanda in 1994, an estimated 8,000 people were systematically being butchered per day. An estimated total of 850,000 Tutsis and moderate Hutus were killed over a four month period in 1994. In the fall of 1996 an estimated one million warrior refugees and survivors of the Rwanda genocide were caught in a "out of control" interstate ethnic crisis involving the armies of Rwanda, Zaire and paramilitary Tutsi ethnic militias. The crisis now threatens to engulf all of Central Africa and the Great Lakes Region. 0

Why are some multi-ethnic states susceptible to violent escalation, state breakdown and collapse while others are not? Current perspectives on violent ethnic conflict and its underlying causes are surveyed in an attempt to answer that question. The focus will be twofold: First, we assess the various attributes of states at risk of ethnic violence. Second, we examine the domestic and international interactions that lead to violence between ethnic groups. We are concerned in particular with states at risk of violence during internal political transformation. For example, power transitions, coups and changes in civil-military relations are generally believed to increase the propensity of a state to engage in both risky foreign and domestic policies. With respect to processes, the emphasis is on how interactions between ethnic groups influence this risky behavior (Weingast 1995; Hardin 1995; Gurr 1994). Through an assessment of attributes and dynamic processes we intend to demonstrate that collective action leading to violence depends very strongly on the course of interaction as well as general attributes. We also argue that ethnic conflict refers to the form the conflict takes, not to its causes. 1 To say that ethnic conflict arises because there are distinct ethnic groups is, at best, tautological. By themselves, ethnic differences are insufficient to guarantee political mobilisation, or intergroup conflict.

Arguments that emphasize the singularity of specific factors should be avoided. The proper analysis of ethnic conflict calls for a careful assessment of root sources in addition to proximate causes. Since, ethnic conflicts are dynamic and multifaceted phenomena. The analysis of ethnic conflict should focus on the development of models based on theories of multiple causation.

This investigation consists of five sections. In section one we assess basic structural determinants of violence. The second section examines interactions between ethnic groups as a source of violent behavior. Third, we evaluate the role of international factors. Fourth, and finally, the conclusion reviews the factors that contribute most significantly to the violent escalation of ethnic conflict.

Domestic Attributes of Instability
(return to top)

By themselves, politically defined ethnic interests are insufficient explanations for the onset of violence. Indeed most conflicts are only superficially ethnic and are stimulated by a combination of non-ethnic factors. These include a blend of structural, (the relative size and location of ethnic groups) instrumental (the role of elites who have a vested interest in advancing particularist agendas) and normative determinants (an appreciation for institutionalised forms of conflict resolution as opposed to coercive measures). Only after these characteristics have been carefully assessed and balanced can it be determined why some states manage political transitions while others do not.

In most cases it is usually the minority and not the majority that initiate political changes leading to confrontational tactics despite the best intentions of political moderates. The primary problem is that deeply divided multi-ethnic states are inherently unstable. There are compelling theoretical reasons for this state of affairs. If a less powerful group is to agree voluntarily (and democratic institutions are in essence voluntary agreements) to endure a dominant group's rules, its interests also must be assured, including safeguards that the more powerful group cannot exploit it. Indeed, it is the minority, wary of the interests of the majority, that ultimately determines the viability of any agreement between majority and minority. Unfortunately, the contractual agreements between minority and majorities in many new states is so weak that minorities cannot be convinced that their interests are best served through accommodation.

If there is a lag in the development of those political institutions that safeguard minority rights, behind social and economic change then it becomes clear why minorities have reason to be concerned. For example, a convergence of interests among Croatian, Slovenian and Serb leaders resembled a balance of power system, characterized by a shifting pattern of flexible coalitions. Eventually this balancing broke down, moderates were discredited and Yugoslavia moved from an equilibrium based on decentralized constitutional arrangements to a state in which coercion became the main instrument of control. The Yugoslavia case demonstrates how exogenous shocks, such as imperatives for rapid economic and political change, can create exploitative opportunities.

A variety of perspectives identify a constellation of factors that account for the causal relationship between state breakdown and the resurgence of ethnic conflict in regions that were once though to be immune to such problems. For example, structural arguments contend that extreme violence is likely under conditions of power parity. For example, when the size and number of groups within states are relatively equal, as in Yugoslavia, conflict becomes more likely. Diverse, plural states encounter special problems in maintaining strong institutional capacity. These states have divided political loyalties and are less likely to develop civic cultures conducive to the pursuit of peaceful policies for the reduction and management of ethnic conflict. Similarly, if a minority is large in size and territorially concentrated, as in the case of Russians in the Ukraine, it not only constitutes a greater threat at the outset but also possesses better resources for its defense. While the state may be hostile toward the minority, its capacity for coercion is reduced. Others contest that power disparity will lead to violence. This argument suggests that it is easier to mobilize an entire population when there are fewer opponents and few structural factors that limit violent strategies. For example, an ethnic minority that is small in size but geographically concentrated, like the Russians in the Baltic states, may predispose these states toward coercive policies. 2 Stability can be defined as a particular set of relationships in which all components are evolving but with the rates of change between ethnic groups and institutions slower than those in other components. In contrast, rapid social and economic mobilization leads to violence and instability. When social mobilization and political participation are high, but institutions are incomplete, the capacity for the state to manage the demands made upon it are diminished. Such regimes will lack the political capacity to carry out reforms peacefully and turn to coercive means to bring about economic and political change. 3

When the state becomes the principal instrument for advancing ethnic group interests, violence becomes more likely. In this case, ethnic identities are evoked in certain structural circumstances to advance the material and political interests of actors whose primary purposes are not ethnic. Subsequent myth-making and dredging up of past events become symbols around which ethnic groups coalesce. These symbols make inter-ethnic violence appear just, honourable and legitimate. For example, until the break up of Yugoslavia, bureaucratic political arrangements managed potential sources of ethnic friction quite effectively. Formally, Yugoslav socialism was a tight-knit system, with public opinion and ideology strictly controlled by an elite bureaucracy. The absence of confrontations and conflicts between classes prevented, in part, the growth of a liberal emphasis on individual rights and liberties. The convergence of interests among Croatian, Slovenian and Serb leaders resembled a balance of power system, characterized by a shifting pattern of flexible coalitions. Eventually this balancing broke down and Yugoslavia moved from an equilibrium based on decentralized constitutional arrangements to a state in which coercion became the main instrument of control.

According to Gurr (1994) four factors determine whether or not a minority will mobilise against state dominance. The first is the degree of economic, social and political disparities between groups. In general, severely deprived groups have a greater chance of becoming politically active. By itself, though, deprivation is not a sufficient condition for ethnic conflict to ensue. A group must also have a common purpose, leadership and organization.

A second factor is the salience of group identity. Cross-cutting identities or low cohesion among ethnic groups is thought to reduce the probability that an ethnic group will be able to act in concert. Motivations for forming ethnic groups may be material, as in mobilization for the defence of interests, or as attempts by the group to either frustrate or promote modernization, for example. Changes experienced at one level, such as dehumanization (a psychological factor), stimulate cohesiveness and eventually increase polarization between groups. Symbols are important group markers in this process of mobilization.

Third, organizational skills and regional concentration are also crucial to the development of political activity. Leadership is crucial to the rise and growth of ethnic movements. Increased scales of ethnic organization encourage ethnic mobilisation to the extent that small-scale bases of ethnic organizations are weakened in favour of large scale ethnic affiliations that provide the organizational framework and constituency for ethnic collective action.

Finally, ethnic mobilisation must elicit a response from the dominant group or state against which it is reacting. Reciprocity and interactions also are important factors to consider. Conflict between dominant groups and minorities usually involve issues of national identity; the expansion and centralization of nationalist political authority which creates a competitive arena for state controlled resources; and the recognition of ethnicity as a basis for resource competition and political access.4

Interactions Between Groups
(return to top)

The foregoing suggests that independence of government from other spheres of state activity will determine the extent to which a state succumbs to ethnic group pressures. In highly developed societies, institutions are arguably better insulated from external influence. Rebellious minorities can be co-opted into the political process and leaders are unlikely to rely on coercion to suppress such challenges. In contrast, a highly repressive regime (police state) will discourage any form of rebellion (Muller and Weede 1990).

However, when determining the likelihood of ethnic violence in transitional societies, it also necessary to assess interactions between civilian and military leaders. Ethnic nationalism, coercion and power transitions go hand-in-hand. As documented by Morgenthau (1957) and many others, the state emerges in crisis whether induced internally or externally. Tilly (1978) suggests that successful use of coercion by a state in order to suppress local, ethnically-based challenges will encourage more of the same. Hence coercion against minority ethnic groups also is a normative factor: elites become habituated to the use of violence. Violence becomes part of the elite political culture that is assimilated into the national identity.

In states with little or no experience in managing ethnic tensions and weak institutional constraints, hegemonial exchange (and its more coercive variant, the control model) usually emerges (Lustick 1979; Rothchild and Chazan 1988). Hegemonial exchange is a statist response intended to manage the overt aspects of intergroup ethnic conflict. Elites bargain for the distribution of resources and control the population through patron-client relations. Even when present, electoral politics are not that influential with regard to elite behavior.

Control approaches arise as the result of a military coup. During periods of political upheaval, soldiers will have difficulty in putting ethnic affiliations aside. The inability of newly elected governments to manage internal ethnic tensions becomes a prime reason for the armed forces to support ethnic leaders who promise to address their concerns. If the military is pulled into civilian affairs, the ethnic base of a regime will narrow progressively until one or two groups dominate the rest.

If the military comes to power it can acquire civilian trappings and seek to build up an affiliated national political party. Former army leaders then become divorced from the command structure within the military, giving rise to conflict over policies between the military and the politicians. Eventually the boundaries between the military and socio-political spheres become fragmented (Horowitz 1985).

These events occur most often in multi-ethnic societies that exhibit porous civil-military boundaries. Africa and Asia are notable in this regard. However, not all coups will lead to institutionalized military control. According to Johnson, Slater and McGowan (1984), among 45 majority-rule states in Africa, 25 experienced coup d'etat between 1960 and 1982. Since independence, military coups have occurred in nearly half of Africa's states -- 56 attempted coups and 102 reported coup attempts in these states. So why has the military intervened this often? Most interpretations focus on the military as a corporate entity in which ethnic cleavages play an important role (Nordlinger 1978; Enloe 1980; Jenkins and Kposowa 1992).

In Africa, ethnic plurality and political competition account for the bulk of military-led ethnic violence. The propensity for military interference within Kenya, Uganda and Tanzania is due to two factors: (a) differences in ethnic composition, which were less salient in Tanzania than in Uganda or Kenya; and (b) fragmentation of power and weak institutions. Uganda has the weakest institutions and the most severe ethnic tensions and consequently the greatest degree of military interference. Tanzania, in contrast, is the most ethnically homogenous and resistant to military interference, with Kenya in between. Although social forces confront each other directly in varying degrees, these societies exhibit low overall levels of politicization. In Uganda the pivotal role of the military reflects the fear of an unstable ethnic situation. In Kenya the civil service provides institutional bases of support and serves as a buffer against military intervention. Hence the interplay between ethnic composition and levels of institutionalization provides an answer as to why the military dominates some states but not others (Goldmann and Wilson 1984). In some states, coups have been attempted but failed (Gabon, Guinea, Gambia, Kenya), with others having no serious attempts at all (Botswana, Cameroon, Ivory Coast, Senegal, Swaziland).5

In post-independence Somalia, political parties formed along regional clan fissures, which only compounded the problem of national unity. Somalia always has been homogenous in religious and linguistic terms; its clan-based struggles for identity shift according to external pressures. For example, in August 1989, Somalia's leader Siad Barre responded to growing clan pressures in two ways: repression against the civilian population and development of a new constitution that emphasized UN-monitored, multi-party elections to be held in February 1991. Similar processes seem to have occurred in Asia. Horowitz (1985) showed that the colonial recruitment of martial groups to counterbalance dominant ethnic groups (Arakanese, Shan and Karens in Burma; Sikhs and Gurkhas in India) plays an important role because some ethnic groups became over represented within the military. Asia provides empirical evidence that a combination of such factors is at work. Leaders of military coups emerge from a given ethnic group and come to rely on its support within the military. The result is action that reflects a narrow range of interest defined in part by the support garnered from their ethnic group. Studies of Pakistani coups support the corporate interest theory, stressing that the most important factors in accounting for Pakistani coups are limitations on military resources, incompetence of the bureaucracy, ethnic cleavages and competition from the paramilitary (Baxter 1991; Rizvi 1991). The impact of these factors has been a crisis of legitimacy under military rule; elites have been coopted by specific ethnic groups and frustrated in attempts to expand their support. Pakistan's social heterogeneity and overdeveloped authoritative structures led to a praetorian society that emphasizes demobilization, power concentration, and patrimonialism. In its search for legitimacy, the Pakistani military resorted to Islamic appeal with a controlled electoral process (Jalal 1990).

Rahman's (1991) study of the 1977 Bangladeshi coup is predicated on similar processes, including civilian interference in promotion and recruitment. The primary factor in sustaining military power appears to be maintenance of corporate interests. Confirmation of this civil-military interaction process in India is provided by Ganguly's (1991) assessment of the military. He argues that the greatest threats to military intervention in India are communal unrest, the Khalistan issue, and treatment of the Army as a corporate unit that receives a greater portion of the federal budget every year. India's potential for a military coup is based on the army's increasing role in assisting civilian authority and institutional weakness in managing internal conflicts. There also is concern about the increasing size of India's paramilitary forces and, as a consequence, the government and the army have been pursuing a wider role in enforcing civil authority.6

All post-communist states in Eastern and Central Europe, to varying degrees, suffer from a sense of insecurity that derives directly from the weakness of governmental and political structures. Democratic institutions are fragile, economic transitions are ambiguous and societal loyalties are fragmented. These factors reduce the effectiveness and impartiality of the military; because of its relatively simple and hierarchical command structure, sensitive issues cannot be depoliticized. Politicization of the command structure causes the military to look for allies within civilian political structures in order to control rebellious minorities. This process will include forging alliances with the police and eventual intervention in civilian politics, either directly through coups or indirectly through a civilianization process. In either case, if the military comes to power, it may acquire civilian trappings and seek to build up a national political party linked to itself. Former army leaders then become divorced from the command structure within the military, giving rise to conflict over policies between the military and the politicians (Roeder 1991). Eventually the boundaries between the military sphere and the sociopolitical sphere will become fragmented (Horowitz 1985). The main function of the armed forces in many emergent states in Eastern and Central Europe will be to defend the interests of specific ethnic groups within the state. The Yugoslav case, for example, reveals the persistence of ethnic loyalties in the Yugoslavia National Army (JNA) that superseded professional loyalties.

International Interactions
(Return to Top)

Even though the number of intense interstate conflicts has declined in recent years, it would be erroneous to assume that ethnic conflicts will remain isolated domestic affairs. Only some ethnic conflicts are rooted at the domestic level. Whether it is through refugee flows or violent interstate disputes, ethnic conflicts affect entire regions not just single states. For example, nationalism and ethnic conflict are inextricably tied to disputes over territory in the form of secession and irredenta.7

Factors in these domains are highly visible and focus exclusively on the affective dimension. Ethnic affinities relate directly to the problem of sovereignty, that is, how the authoritative claims of states and feasibility of implementation evolve in response to international linkages. Motivations include (a) a shared sense of historic injustice; (b) shared identity; (c) religious affinity; (d) common ideological principles; and (e) a degree of inchoate racial-cultural affinity (Heraclides 1991; Suhrke and Noble 1977: 1- 15). Efforts to control the flow of people, culture and resources and to assert authority are significant. Quite often, authority is not defined solely in terms of territory; partial sovereignty can exist. Transnational ethnic affinities are salient in interstate relationships. Not only do elites view ethnic affinity as an opportunity to be exploited; specific groups on whom they rely for support also see these international ethnic linkages as potentially useful.

Transnational ethnic affinities exist among most groups in the international system, especially those that have undergone diaspora (e.g., Russians living in the Ukraine and the Baltic states, the Tamils of South India, the Chinese of Southeast Asia) (Taylor and Jodice 1983; Neilsson 1985; Gurr 1992). However, defining transnational affinity is difficult because of the multiple ways to establish ethnic identity (Rothschild 1981; Horowitz 1985). Race, religion, tribal (kinship) and linguistic cleavages may not coincide, so international ethnic affinity in one area (linguistic) may be at odds with another area (kinship) (Chazan 1991). Moreover, elites often will play up certain transnational identities (pan-Arabism as opposed to Islam, for example) in order to reduce the salience of others. While an ethnic group may share a common bond across a border, that does not guarantee mutual interest.

Moreover, mutual interests are strongest for groups that have high international ethnic affinities and see the 'other' group as an enemy of the supporting state. The other or 'out' group in this instance can be the state-center or an ethnic minority. If affinities do not look promising for a convergent mutual interest, then a state is less inclined to adopt an ethnically oriented foreign policy. The group or state in question is less likely to pursue ethnically-based support. Future escalation is unlikely unless these conditions change.

Closely linked with transnational ethnic affinities is the concept of ethnic cleavage (Shih 1992). Cleavage refers to the degree of divided political loyalties among the ethnic groups within a state. For example, ethnic groups that aspire to self-determination, but are willing to work through existing political institutions and procedures, would be characterized as having a lower level of cleavage than those who seek to transform the political status quo through force, external assistance, or both. For intervening states, cleavage provides an opportunity to be exploited. For the state in question, cleavage is an obvious domestic constraint. Divided loyalties occur within a state that has marginal system maintenance functions, weakened institutions and political parties based on ethnic groups. In such instances the elite of an ethnic group pursuing secessionist policies actively will seek external support and goods, as will the state-center. Low levels of cleavage between ethnic groups occur within a state that has developed strong institutions, political capacities for the management of ethnic tensions and cross-cutting cleavages that reduce the capacity for ethnic mobilization.

Two features of Yugoslavia's ethnic diversity illustrate these linkages. All of Yugoslavia's ethnic groups, including Croatian, Slovenian, Muslim and Serb, tend towards geographic concentration; and in the other states, including Croatia and Bosnia, Serbs constitute significant minorities. Although not a majority of the former Yugoslavia's population, Serbs always constituted its single largest ethnic group (Ramet 1992a). Instrumental in shaping domination is the fact that Serbian leaders controlled the Yugoslavian Army and political apparatus for at least ten years prior to the outbreak of war. The presence of transnational ethnic affinities created a security dilemma for Serbia, Bosnia and Croatia (although less so for Slovenia) and had ramifications for the subsequent formation of Serb policy toward each of the seceding states. For example, in relative terms, far fewer Serbs are found in Slovenia and Macedonia than in Croatia and Bosnia. Thus the conflict between Slovenia and Serbia primarily concerned further decentralization within the Yugoslavian political structure. Efforts to protect, retrieve or even utilize the small minority of Slovenian Serbs as a means of organizational leverage were second in importance to the larger issue of maintaining Yugoslavian integrity.8

In addition to third party involvement ethnic conflicts become internationalised through diffusion. Diffusion refers to the process by which ethnic conflict influences the behaviour of ethnic kin in neighbouring and distant host states. Contagion, demonstration effects, information flows, and material and ideological support for ethnic diaspora are types of diffusion.

Contagion refers to a situation in which events in one state change directly the ethnic balance of power in a neighbouring state. Through this path, ethnic displacement, refugee flows and spontaneous population transfers constitute a form of contagion. The spillover of displaced ethnic groups creates changes in demography thereby creating regional instability. For example, the violent outflow of Tutsis and moderate Hutus from Rwanda to Zaire and Burundi in 1994 has created a new class of militant ethnic leaders that Howard Adelman has characterised as "refugee warriors."

Second, ethnic conflicts in one country may prompt groups in another to make more extreme demands. This constitutes a demonstration effect. Groups in one state, witnessing ethnic mobilization by ethnic groups in another, may increase their own political activities. The latter recognize that internationalization of their demands can both simultaneously encourage internal mobilization and weaken the saliency and effectiveness of the state by creating international forums for substate grievances. This legitimization process is supported by the existence of supranational organizations and human rights organizations, which provide a forum for subnational ethnic claims. Consider the 25 June 1991, declarations of independence first by Croatia and then by Slovenia as a demonstration effect that emboldened both states to commit to full separation from the Yugoslav Federation.

Third, diffusion occurs through information flows and transnational media networks that condition the behaviour of transnational ethnic coalitions. Information flows directly influence the level of protest and rebellion and the level of ethnic mobilization among ethnic brethren. For example, consider the way in Bosnian Muslims were portrayed by the Serb media in order to undermine the Dayton and Vane-Owen Peace Accords. These peace plans were presented as a threat of Muslim fundamentalists in Bosnia, seeking to perpetrate genocide against Serbs, an argument that pervaded the Belgrade media.

Fourth, ethnic diasporas provide material and non-material support for politically mobilized ethnic groups. These affective links are crucial for an ethnic separatist movement to prosper and grow. Two types of linkage are notable. The first are particularist identities between groups that straddle borders. The second linkage is the impact a global diaspora has on the development of ethnic leadership pools in non-neighbouring states. For example, after the Gulf War there was realignment of international interests in South East Asia among the Muslim community. There was a shift away from Saudi Arabia to Iran, as the primary external focal point of support for many Islamic protest groups.

(Return to Top)
Based on the analysis of domestic attributes, interactions between groups, and international interactions, the following conditions are associated with the violent escalation of ethnic conflict:

1. a state is experiencing power transitions and rapid transformations in institutions that weaken its regime;

2. a single ethnic group comes to dominate policy-making procedures and institutions and inter-elite accommodation is weak or nonexistent;

3. the state provides a differential advantage to a minority group over issues related to distribution of resources;

4. political channels for the articulation of minority interests are weak or underdeveloped;

5. the military is dominated by a single ethnic group;

6. elites maintain institutions specialized in the use of coercion;

7. affective motivation and ethnic cleavage are present;

8. demonstration effects include the success of insurgencies and minority struggles in neighbouring states;

9. there is third-party support for an insurgency from allies of the ethnic group.

As anticipated, both attributes and processes appear on the list of relevant factors. Thus the escalation of ethnic conflict cannot be traced to any single factor; it is a product of pre-existing conditions that combine with actions by groups and states to produce violence.

To assess the strength of each of these variables in explaining outcomes an analytical approach would require the development of fairly complex models. Such models are oriented towards assessing both the likelihood that specific events are likely to occur and in the identification of specific factors and their interactions that feed into creating such events. For analysts trying to make sense of the range and phases of ethnic conflict, a good beginning is the creation of profiles of specific conflicts and the types of events under scrutiny. Not only are there significant differences in type there are also differences in levels of aggregation as well. For example, there are low- intensity clashes between groups within states such as in Sri Lanka, there are ethnically-based intrastate wars such as the Ethiopia-Somalia conflict and there are spillover conflicts involving refugee flows into neighbouring states as occurred during the Rwanda crisis. On occasion there is overlap, interactions and multiplier effects so that all three (or more) levels are manifest in one conflict. such as in the former Yugoslavia.

Benchmark profiles of each type of conflict could provide the empirical basis for testing the validity of more broadly defined theoretical frameworks. Profiles also provide an important base point for determining how new cases should be classified. Ultimately any measurement of success in risk assessment depends heavily on the ability of case-study and country specific information to be combined with statistically-based forecasting techniques.

Another way to place these results in context is to draw parallels with crisis escalation in general. Are the conditions associated with ethnic conflict consistent with the factors that facilitate the violent escalation of international crises? To answer that question, it is appropriate to review the findings of the most comprehensive study available, namely, the research conducted by the International Crisis Behavior (ICB) Project on 70 years of data (1918-1988). Brecher (1993: 154-166) found very strong support for the following factors in connection with the escalation of crises into serious clashes or full-scale war: the triggering action, response from the target, geographic distance, nature of the pair of regimes involved, number of actors involved, and conflict setting.

Obvious parallels emerge between findings from the ICB Project and the list of factors identified in this chapter. Brecher (1993: 164) discovered that violence occurred much more frequently within dyads composed of authoritarian regimes. This result is consistent with quite a few items from the above-noted list; ethnic conflict is more likely to become violent when a single ethnic group emerges as dominant, controls the military, disregards minority rights, and elites become coercive. Brecher's (1993: 164) discovery that crisis escalation to violence becomes more likely with geographic proximity parallels conditions seven and eight from the current list, which focus on affective motivation, ethnic cleavage, demonstration effects, and minority struggles in neighbouring states. Item nine parallels two of Brecher's (1993: 164) findings: Crises are more likely to escalate to violence when (a) three or more actors are involved and (b) they take place within a protracted conflict. Third-party involvement in an ethnic conflict is likely to stimulate interest among more actors (and further complications) and the presence of an intermediary may signal that the events are unfolding within some larger process, such as a protracted conflict.

Finally, Brecher (1993: 164) discovered that crises with violent triggering events, and target responses of equal or higher severity, had a much greater chance of resulting in serious clashes or war. This finding serves as an additional reminder of the importance of limiting and preferably preventing violence in ethnic conflicts. It also makes clear that the level of violence in an ethnic conflict will depend on a group's beliefs about its environment and the willingness of opponents to suffer escalation costs.


0. Ethnic conflict is a dynamic process in which at least five stages of escalation can clearly be identified. These include a latent stage in which differences between ethnic groups are made salient but there is no overt conflict; an onset phase whereby a "trigger" creates the conditions for violence; a peak point leading to large scale confrontation between groups leading to repression or domination; a de-escalation phase, including perhaps some form of third-party involvement and finally a termination phase resulting in the resolution or transformation of the conflict. Ethnic conflicts can last months, years or decades The most salient are protracted conflicts, fluctuating in intensity over the course of several decades and involving entire communities. These include the Arab-Israel conflict, the Kashmir dispute, the conflict over Cyprus and Sri Lanka. The key issue for analysts is to learn how to detect ripe moments, turning points and conflict escalation in order for policymakers to stem, prevent, mitigate and channel, conflictual behaviour before it becomes violent. The implication is that identifying patterns of interaction can be used to understand how and when the conditions of escalation and de-escalation occur. If one party can be sought to reduce its conflictual behaviour it is possible that the other side will reciprocate. International efforts can be employed as leverage in order to induce a "hurting stalemate" or to influence the behaviour of one of the participants.

1. From an analytical perspective, the most recent evidence indicates that ethnic violence is in modest decline. According to the Minorities at Risk Project between 1945 and 1980 there was gradual increase in ethnic rebellion and non-violent protest. Levels of rebellion reached their peak point by the early 1990s. Projections indicate that between 1996 and the year 2000 ethnic strife will decline even further. In addition, the bulk of the violence today occurs within states not between them. For example, since 1980 there have been few interstate wars and even fewer interstate wars that are primarily ethnic in character - Armenia and Azerbaijan, Iran and Iraq and Yugoslavia are important exceptions. However between 1945 and 1988 there were in excess of 200 interstate ethnic crises involving secession, irredenta or some combination of both. These include the Arab-Israeli Wars, Cyprus, Kashmir and the wars between Ethiopia and Somalia.

2. Another structural indicator related to demography is whether or not the economy is efficient or redistributive. Ethnic groups seeking resources have two options. First, they can pursue national policies that increase aggregate social wealth through methods that are labelled efficient (i.e., attempt to improve the condition of almost all of the groups within a society). Second, an economic system dominated by a single ethnic group can be expected to produce group-specific benefits or "rents" that typically distort the economy through redistribution (i.e., seek to improve the conditions of one group in society at the expense of another).


4. It follows that ethnic minority rebellion is most likely against a moderately repressive regime. Other kinds of political action are more attractive when the government is not punitive, while extremely brutal authorities will discourage all forms of resistance. In those states where ethnic leaders depend on the support of more than one ethnic group or focus on cross-cutting identities, ethnic mobilisation may be less likely (because there is an attempt to improve the condition of almost all of the groups within a society). In contrast, an economic system dominated by a single ethnic group is said to be redistributive may be more susceptible to minority grievances.

5. Some states have managed to avoid coups for other than economic reasons, such as French-speaking African states relying upon military assistance agreements with France (and even defense pacts with France -- Cameroon, Gabon, The Ivory Coast and Senegal). In a similar way, Kenya maintained British training facilities that may have inhibited a coup during the formative years of independence.

6. According to Ganguly (1991) there is no imminent prospect of a coup in India, but concomitant attempts to reduce excessive reliance on control should be expected.

7. However, despite various appeals to ethnic sensibilities, violent Yugoslav-type scenarios will continue to be uncommon. Of greater concern are Rwanda and Bosnia-type state collapses which require specific international tasks in restructuring a state's institutions, civilianization responsibilities and credible third party commitments that guarantee minority rights in the absence of internal political order.

8. These factors may explain, at least in part, why Serbia relinquished control of Slovenia (and possibly Macedonia, where Serbs are also a small minority) so quickly, compared to the conflicts in Croatia and Bosnia. Perceived benefits to Serb leaders, if any, did not match the costs of attempting to retain Slovenia. This does not entirely explain the use of force, but it may help in accounting for the brevity of the Slovenian war. Like Croatia, Slovenia had its own defense forces and strong ties to the West. The outside connections became evident when Germany sponsored Slovenia's transition to independence. Potential confrontation with the West, coupled with fewer domestic benefits to the Serb leadership, led to a brief ten-day war (a relatively peaceful political transition, considering that the only other successful secession to that date was Bangladesh). Force could be used as long as the international community viewed the conflict as a civil war, that is, an internal affair of the still-existing Yugoslavian regime (which it did).

Return to top of page

Close Window

npsiaban.gif (11946 bytes)