What drives ethnic conflicts forward?

Which school of though better explains the causes of ethnic violence?

Introduction

The nature of ethnic violence has been always generating debates among scholars coming from the fields of comparative politics and international relations. The question discussed in this post will contrast several types of arguments, the rationalist argument that sees the ethnic conflict as a rational result of group and individual behavior (Steinberg 1981; Glazer & Moynihan 1975) with the premordialist argument that observe the ethnic violence as an inevitable clash based on innate antagonisms of contending groups (Smith 1986; Kaplan 1993), the constructivist, and the symbolist politics. I will try to project which of the theory provides a better answer to the questions: whether conflicts among different ethnicities are natural phenomena or and what are the factors that trigger the escalation of ethnic violence?

Rational choice theory perceives ethnic conflict to be a result of the rational pursuit of private or group interests. The conflict between different contending groups is not fought along cultural lines but is rather waged for achieving of such rational interests as wealth, power, and security. The rationalist logic in understanding ethnic conflict comes from their instrumental understanding of ethnicity. In this context, Glazer & Moynihan (1975), and Steinberg (1981) claim that ethnicity is “… a tool used by individuals, groups or elites to obtain some larger, typically material end” (cited in Lake and Rothchild 1998, p. 5). In the rationalists’ understanding, ethnicity is just a powerful political identity that is not different from other political identities. In this context, rational theorists do not identify any variations between ethnic conflicts and other forms of violence, interest-based contentions, which are constructed on ideological differences. As Kaufmann (2006) puts it, for rationalists “ethnicity is primarily a label used for political advantage” (p. 49). 

The main problem with the rationalist argument is that ethnicity should not be treated as a regular form of political identification. While the political identity can change over time the ethnic identity remains always immutable. So, there is no clear definition given to the concept of identity from the rationalist perspective. This conceptual ambiguity introduced by the rationalists requires scholars to be more specific in defining the difference between ideological and ethnic violence (Biddle cited in Kalyvas and Kocher 2007).

Biddle specifies that unlike politically based conflicts, the ethnic wars are fought for the resistance of regional groups rather than for national principles or for the redistribution of roles between social classes. The political based conflicts are fought for maintenance of a specific ideology while the latter for survival of ethnic groups. In the same order, Kaufmann (2006) claims that the difference between the two is that “ethnic cleavages produce ‘ethnic’ or ‘intercommunity’ civil wars, whereas ideological cleavages result in ‘revolutionary’ or ‘ideological’ civil wars” (cited in Kalyvas and Kocher, 2007, p. 186). So, the ideological cleavages lead to rivalries between the state elite and the rebels for the loyalty of their constituents, while ethnic divisions result in struggles between clearly-defined and “mutually exclusive groups” (p. 186).

Kalyvas and Kocher (2007) critique is that ethnic and ideological divisions are artificial. By looking in a comparative perspective on Iraqi and Vietnam wars, the scholars draw the conclusion that it would be misleading to make a clear-cut differentiation between ideological and ethnic wars. Both elements, according to them, are present in the same wars and source one from another. Moreover, they notice that civil war violence should not be necessarily perceived uniquely as a reflection of existing cleavages between contending groups but also as a generator of new cleavages in itself. In this sense, with the escalation of a conflict situation the opposing groups can intensify the already existing cleavages that consequently lead to the rise of violence (p. 204).

A distinctive rationalist model is the security dilemma model. According to Posen 1993; Fearon 1995; Snyder & Jervis 1999, the conflict situation between ethnic groups is a result of the existing insecurity and uncertainty among them. Consequently, the dilemma in ethnic conflicts is that the collective fear against a potential conflict leads groups to violence. Thus, by increasing the level of personal security, the groups decrease the level of general security, fact that leads to violent outcomes. According to Lake and Rothchild (1998), and Kaufman (2006), ethnic wars are caused primarily by information failures and commitment problems. These two factors impede contending groups from reaching a commonly agreed, negotiated bargain. By private information the scholars mean “anything known by one group but not the other, including how intense their preferences are about specific policy objectives, how cohesive the group would be if challenged, or how military leaders would use their forces should fighting break out” (Lake & Rothchild, 1998, p. 12). Therefore, the result of uncertainty about each others’ intentions increases the level of insecurity, which consequently pushes ethnic groups into violent conflicts.

One more problem stemming from information failure is that groups resort to bluffing strategies to be able to augment the level of personal security, which consequently, leads to an increase in the general level of insecurity. According to Lake and Rothchild (1998), with the escalation of violence, the cost of information dramatically increases also. In this sense, the conflicting groups choose to enhance their offensive and defensive capacities instead of spending financial resources on finding information about political claims and real intentions of rivals. Kaufman (2006) adds, in the same context, that conflicting parties are oftentimes unable to calculate the potential conflict outcomes and in this way, the potential loser may not be able to avoid the conflict before it escalates.

The second problem identified both by Lake and Rothchild, and Kaufman is the problem of credible commitments. According to them, the groups would prefer to bear the high costs of war today in order to avoid a potential defeat later. Consequently, the mutually advantageous agreement is impossible because both parties expect the other to defect from the agreement commitments, and consequently to use the situation for military benefits. Lijphart (1967) explores potential solutions for achieving credible commitments, and namely: first, institutionalizing political power-sharing arrangements between ethnic groups (electoral rules or group vetoes); second, placing important economic assets in ethnic minority’s oversight; third, maintaining an ethnic balance in the army and police forces (cited in Lake & Rotchild, 1998, p. 14).

Another factor that increases the chances of a security dilemma is the first-strike advantage (Melander, 2009). Each conflicting ethnic group knows, that by striking first, it could reach a net advantage over the situation. However as literature shows, this military action is not common for all communal conflicts. After carrying out a statistical analysis of ethnic conflicts, Melander (2009) draws the conclusion that the first-strike strategy is mostly used in communities where the number of competing groups is approximately equal. And vice-versa, in communities where the opposing groups are highly disproportionate, and one clearly dominates the other, the first-strike risk is extremely small (p. 96).

Another derivation of the rationalist model, which tries to explain the ethnic violence, is the predatory elites approach. According to Figueiredo and Weingast (1999), the security dilemma does not sufficiently explain how political elites manage to convince the masses to prefer violence to peace (p. 50). In their model, predatory elites are the main engines behind ethnic wars and genocides. Bloody conflagrations, in this sense, occur as a consequence of politics of the ruling elites which recourse in this sense to manipulation of masses (cited in Kaufman, 2006, p. 46-47). In this context, the theorists mention that there are three types of conflicts that could be triggered by elites. First, the ideological conflicts, which are characterized by ideological struggles defined in economic or class terms, (the case of Communist revolutions in Russia or China). In the same category scholars place the religious crusades. Second, the criminal assaults on state sovereignty, which challenge the state authority over a certain territory by individual or cartel powers (regional warlords or drug leaders in Colombia). The last is the raw struggle for power of local leaders.

The struggles for power makes political leaders use all possible political cards. The use of ethnic politics as an instrument to access power is hence briefly portrayed in the cases of Bosnia, Azerbaijan, Kenya and India. The Kaufman (2006) critique regarding this model is that is illogic for leaders to venture in ethnic politics that would lead to violence among ethnic groups because in the vast majority of the cases, they lose political power as an effect of the application of this strategy.

Mueller’s (2000) conclusion after analyzing the Rwandan and Serbian wars, is that ethnic wars in fact do not exist. He claims that the ethnic conflict is somewhat similar to the non-ethnic conflict since it is carried out by small “bands of opportunistic marauders recruited by political leaders” (p. 42). Moreover, that in former Yugoslavia the violence escalated not because of nationalist claims from the opposing parties but because well-armed groups of thugs and looters managed to take under control certain regions as an effect of the breakdown of central authority (p. 47). According to Mueller, the composition of these paramilitary groups was 80% criminals and 20% radical nationalists. According to his calculations, even in the case of the Rwandan genocide the number of perpetuators could not exceed fifty thousand people, which constituted less than two percent of the whole Hutu population.

In conclusion, Mueller (2000) says that ethnicity as such, does not play a role in the escalation of violent conflicts but it is a feature around which perpetrators and politicians construct their politics or reasoning behind inhuman behavior. The solution that Mueller (2000) proposes is to increase the international policing in conflict regions. The presence of well-armed international military forces will discourage the criminal mobs to get involved into military actions. Yet, even though Mueller (2000) reasoning sounds well-argued, the scholar fails to answer some key questions. First, assuming that the ethnic wars in Serbia and Rwanda were led by non-ethnically bound military gangs searching for war loots only, why did they degrade to genocidal practices then which are economically non-profitable? Moreover, why did they ally with nationalistic leaders if their interests in the war significantly differed?

Analyzing the connection between the predatory elite behavior and security dilemma, Kaufman (2006) identifies that it is rarely possible to find a pure form of security dilemma where the leaders are not also led by predatory motives. Consequently, Kaufman declares “chauvinist leaders always claim to be driven by security motives, but what makes them chauvinists is that they deem their group’s security as requiring dominance over rival groups - which is, naturally, threatening to the others” (p. 54). The opposite is also valid, which means that if there are two parties are competing for authority over the same territory, we deal with a security dilemma. In this sense, none of the parties feel safe except when their status demands are fulfilled. Yet, these two cannot be achieved at once (p. 55).

To conclude on the rationalist approaches, it is worth mentioning that ethnic identity is regarded as an instrument for the achievement of the rational ends. The leaders use it as a tool to justify group or personal interests. That is why acceding to this school is seen mainly as a tool of mass manipulation and motivation of the perpetuated violence against other groups.  

Contrasting rationalist argumentation of the ethnic violence primordialist school treats ethnic conflict as an inevitable result of cohabitation of different ethnic groups on the same territory. According to this understanding, the ethnic identity is a characteristic received from birth, which defines the state of belonging of an individual to a specific ethnic group. In this context, the ethnicity is the main agent, which makes the individual identify himself with a specific community in opposition to others. Consequently, as Van den Berghe (1981) mentions, the conflict is rooted in ethnic division of societies and will always remain a part of their history (cited in Lake & Rothchild, 1998, p. 7). In addition, Moynihan (1993) reinforces the argument by stating that "nation states no longer seem inclined to go to war with one another, but ethnic groups fight all the time" (cited in Fearon & Laitin, 1996, p. 716). In a similar context, Horowitz (1985), Rabushka and Shepsle (1972) notice that heterogeneous societies are more likely to experience intense domestic conflicts. A related argument is extended by Gellner (1983) who emphasizes “the nationalist principle is violated when there are multiple ‘potential nations’ within state boundaries” (cited in Fearon et al., 2007, p. 2).  The conclusion in this sense would be that a territorial political unity could be formed only in the cases when the region can become ethnically homogeneous, which could be achieved by killing, expelling, or assimilates all non encompassing ethnic groups.

Bhavnani and Miodownik (2008) try to bring empirical evidence to the Primordialist argument. Running a statistical analysis of the relationship between ethnic salience and the probability of violent conflict in a number of African states, they conclude that the stronger the feeling of ethnicity of members across rival groups, the higher the probability of conflict at the lowest level of ethnic polarization. They deduce that political domination by minority groups increases the probability of violent conflict to escalate. Thus, a larger disproportion of political representation leads to higher likelihood of political conflict between ethnic groups.

The main shortcoming of primordialist school is that it fails to explain the variation in the magnitude and localization of ethnic violence. By conducting an empirical study on the post-Soviet space, Laitin  (2001) did not identify a correlation between ethnic differences and ethnic wars. He did not find any relevant evidence to explain why ethno-cultural diversity in some post-Soviet states led to violence while in others it did not. Despite this argument Laitin does not negate the salience of the ethnic component on the escalation of violent conflicts claiming that ethnic diversity influences only marginally the probability of ethnic wars while.

Checking the same hypothesis on the other data set Gurr (2000) attempts to identify the tendencies in the escalation and de-escalation of violence in the period 1999-2000. For this purpose the scholar analyzes fifty-nine conflicts, and concludes that twenty-three of them de-escalated while only seven escalated (p. 54). This fact talks about a fluctuation of the level of violence in ethnic conflicts, which proves the non-sustainability of the primordialist approach according to which the number of violent conflicts should equal the number of cohabitating ethnic groups at an approximately same level.

In exchange, Gurr (2000) puts forward the concept of “accommodating behavior”, which assumes that ethnic groups can easily institute a system of accommodating policies, which would neutralize the chances for conflict situations among them. According to Gurr, the accommodating policy could be easily enforced if three main rules are followed. First is to promote democratic institutions and practices, which would enforce equal representation of all cohabitating ethnic groups and hence institute a set of norms restraining unilateral repressions. The second is the involvement of international actors, and the implementation of international human rights norms, that will make the region more secure and will align it to the existing global standards. Finally, an international global elite should be created, which will set up a global governance system by integrating state policies into a system of global politics (p. 59). Moreover, the author mentions that “preventive diplomacy which could be popular - not only because early engagement cold be cheaper than belated crisis management, but also because it is the preferred instrument of the new regime” (p. 58). According to him, the only risk for instituting a system of accommodating behavior would be the emergence of hegemonic elites who will tend to use the state for the promotion of their own interests. Additionally, the author mentions that anticipation and resolution of ethnic conflicts could be easily achieved by granting large territorial autonomy to minorities. Consequently, the risk of dividing the state could be anticipated by instituting a power sharing mechanism in which minority groups would be involved (p. 52).

Consequently, having shown that the primordialist approach does not provide complete answers by assuming that the ethnic conflict is inevitable because it results from competition of ethnic groups, we still do not know what is the outcome of state policy if a majority or a minority ethnic group leads a country. In this sense, Fearon et al. (2007) investigates whether there is a greater risk of the escalation of an ethnic conflict if an ethnic minority holds full political authority. Their initial assumption is that the exclusion of ethnic groups from power would lead to the agitation and ethnic politics.  Thus statistical analysis proves that there is a very low degree of chance that in the countries with presidents belonging to a minority ethnic group would start civil war.

Constructivist explanation brings an added value to the debate on causes of ethnic violence. For Anderson (1983), Dominguez (1989), Young (1993) and Brubaker (1995) ethnicity is neither immutable nor completely open. For them, ethnicity is constructed from an intertwined network of social interactions. Therefore, ethnicity is seen as a social construction. That is why ethnicity cannot be treated as a cause of violent conflicts by itself as it is already a product of social construction. This theoretical framework claims that the majority of ethnic groups prefer to achieve their interests and demands in a peaceful manner, by using the institutionalized political channels (p. 6). For constructivists the conflict emerges from pathological social systems, which cannot be controlled and guided neither by individuals nor groups (Lake & Rothchild, 1998, p. 6). Kalyvas (2008) mentiones in this regard that for constructivists the conflict has a self-propelling power in itself – it tends to strengthens ethnic identity which leads as a consequence to intensification of the violence.

Making a comparison between rationalist and constructivist argumentation about the ethnic conflicts Kalivas (2008) finds that they both are in favor of differentiating between ethnic and non-ethnic wars. The difference is however that the rationalist school does not practically differ between the two while the constructivist school wants to divide them to show the differentiation pattern. For instance, when combatants in ethnic wars decide to change the side they are fighting for is very rare, however this happens oftentimes in ideological conflicts. Kalyvas mentions in this case that both political and ethnic identities have similar chances to be abandoned in wars. For instance, when their army was defied the Republicans left the country. Similarly with the thousands of Caucasian Russians, nationalist Chinese, and anticommunist Vietnamese, they left their countries because they could not align to the new political identity (p. 1047). The conclusion Kalyvas comes to in his analysis of ethnic and non-ethnic wars is that these may not necessarily differ as much one from another as it is suggested in the scholarly literature. He claims hence that the existing variation originates less from the nature of identities but rather from the strategies of opposing groups in ethnic wars (p. 1048). That is why the causes o ethnic violence should not bi singled out or differentiated from the ideological ones.

In addition, Kalyvas (2008) proves that the constructivist theory does not resist the empirical testing of individual behavior during conflicting situations. Accordingly, by constructivist assumption the individuals should back the ethnic groups they belong to. Nevertheless, the scholar proves that a significant number of ethnic defections can be observed in wars. The scholar also remarks that these conditions are not specific only to ethnic wars but also to non-ethnic violent actions, fact that reinforces his stance that there is no clear division between ethnic and non-ethnic wars (p. 1050). As he puts it, the main reasons of ethnic defection are the habit of individuals to follow the leader of the clan, material incentives and fear of reckoning and revenge (p. 1052). Hence, the scholar highlights the salience of ethnic defection mainly because it obliterates the elements that make ethnic identity essential for collective action (p. 1055). Accordingly, Kalyvas mentions that ethnic civil wars where ethnic defection takes place degenerate into competitions for the loyalty of constituents and therefore are similar to non-ethnic civil wars. To conclude, constructivism for Kalyvas is a comprehensive theory but it does not fully explain ethnic-based behavior.

From the perspective of symbolic politics theory, which is a branch of the social-psychological theory, the main causes of extreme ethnic violence are rooted in group myths. According to Kaufman (2006), myths justify hostility among ethnic groups and create the state of fear of group extinction. In the same time, for symbolic politics the myths are the main triggers of chauvinist mobilization (p. 47). Thus, ethnic identity is neither rational in its nature nor a socially construct manipulated by elites. It is a uniform pattern of belief of a community about “the other”. Consequently, according to Smith (1986), every ethnic or national group has its own “myth-symbol complex” (cited in Kaufman, 2006, p. 50). Thus, the myth is the community coagulator, which identifies what elements should be incorporated in the common culture and which not. Consequently, the myth is the defining element of the group history.

Thus, being constructed on the findings of the neurologist science, symbolic politics theory shows that emotions rather than rational calculations are the main stimulators for individuals to act. Emotions make people choose the set of possible priorities. Thus being in line with thit the rationalist school argument Kaufman (2006) affirms that “fear, for example, causes people to prioritize security over other values such as wealth” (p. 51). Consequently, he notes “the more a group’s myth-symbol complex focuses the group hostility on a particular adversary, the greater the probability of a violent clash with that adversary” (p. 51). According to him the escalation of ethnic conflicts is a direct result of negative mythological arrangements, which exists in that society. The myth justifies hostility versus the outer group and based on the fear of extinction encourages members of that group to strike first. The given prescription is to change the intergroup symbolic politics. The transformation of mythological principles that underpin social beliefs would reduce the chances for predatory leaders and chauvinist politicians to use the myth as a vehicle for the promotion of their interests (p. 47).

Even though Kaufman (2006) attempted to show that his theory is more successful than the rational choice theory in explaining cases of extreme violence in ethnic conflicts, this theory did not attract many supporters. The main critics of symbolist politics theory claim that myths and symbols cannot explain the variation in outcomes of ethnic politics (Fearon & Laitin, 2000, p. 864). They also emphasize the impossibility to treat political symbols as independent self-satisfactory elements as they did not exist prior or independently of human communities. In this context, it is claimed that political symbols cannot be treated solely as explanatory factors for chauvinist politics. They should also be analyzed as consequences of this type of politics. If political symbols are constants and underpin the political identity foundation, why then ethnic conflicts are not permanent also? Why do they escalate only at certain points in time and vary from one region to another? The symbolist theory does not explain the dynamics of civil wars and their occurrences. 

Which school better explains the causes of ethnic violence?

My argument is that rational school has a powerful tool to explain the causes and escalation of ethnic conflicts. However its explanation power is confined to those cases, which could be, characterize as being non-ethnic conflicts. That is why I would insist for a multi-scholarly approach for explanation of ethnic conflicts

Assuming that there is no scholarly consensus among the schools on the causes of ethnic conflicts I will try to identify which of the aforementioned better explain and cover a larger range of causes of ethnic conflicts. For this purpose I will mainly rely on the empirical analysis of Sahadevan (2003) five major causes of ethnic conflicts identified in Central Asia.

The first cause identified by Sahadevan (2003) is best explained by the constructivist theory. This is the concern of ethnic groups to lose their identity. This situation usually emerges in the case of small bordering minority groups, which were massively involved in decolonization movements, and consequently nurtured their aspiration for political independence. This type of concern engages ethnic groups usually sharing a distinct remote territory in ethnic politics fact usually non-tolerated by the numerically dominant majority. The second and the third causes are the fear of assimilation and marginalization. This cause could be answered through both perspective of both structuralist and rationalist schools. In the case of Jat Sikh, for instance, Sahadevan (2003) mentions the fact that the community was facing an internal decline accompanied by external threat sourcing from assimilative attitudes of the Hindu who treated Sikhism as a “sword arm of Hinduism” (p. 407). The author qualifies, in this sense, colonization as a state-sponsored ethnic oriented political program, which aims to neutralize the power and political positions of existing ethnic groups and to convert them in minorities within their own territory (p. 407). Hence the rationalist argument would stand for the fear of elites to lose their advantageous social status thus will recourse to the ethnic politics to oppose the assimilative and colonization practices. The structuralist argument in this case would stand for the rigidity of the ethnic values against the heterogeneous assimilating ones. The cause of relative deprivation could be also explained by the rationalist logic however it finds a better explanation power in the theories of social movements. Consequently, Sahadevan mentions that ethnic relations in Asia have always been a state-controlled affair, thus the denial of equality status for some minorities would directly affect their survival. The last cause identified by Sahadevan (2003) is directly related to the fight for power among existing ethnic groups. This is again a both rationally and culturally drieven argument. The scholar agues in this sense is that the majoritarian hegemonic principle of governance instituted in all analyzed states may lead to the total exclusion of ethnic minorities from political life.