Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Priority Themes for Research on Collective Violence Mark R. Beissinger University of Wisconsin Sta this N. Kalyvas University of Chicago bile it is true that all acts of collective violence share several things in common (they all involve the intentional exercise of physical force by two or more persons with the aim of inflicting injury on other persons or causing damage to property), the range of acts that fall under the rubric of collective violence is quite capacious. Collective vio- lence assumes such disparate forms as terrorist activity, brawls, pogroms, riots, communal violence, vigilante violence, forced expulsions, insurrec- tions, civil wars, revolutions, and genocides; at times, it has been defined to include state-sponsored terror and interstate warfare as well. Signifi- cant and largely separate literatures exist on all these subjects, and several generations of research on these topics has produced agreement within the scholarly community that no single explanation for collective violence is possible. Moreover, the "collectives" involved in the perpetration of collective violence vary significantly small networks or organizations of individu- als, less formally organized crowds and gatherings, loosely connected paramilitaries, armies with varying degrees of discipline, and govern- ment bureaucracies. Some acts are perpetrated by members of dominant groups or the state against members of subordinate groups; others by members of subordinate groups against members of dominant groups or the state; still others involve two groups of perpetrators seeking to inflict damage on each other. In addition to the variety of forms and actors involved, there are problems of explanation that emerge out of the varied scope and severity of collective violence. Collective violence can consti- tute a relatively brief event embedded within a larger chain of social 40
PRIORITY THEMES FOR RESEARCH ON COLLECTIVE VIOLENCE 41 interaction (such as typically occurs in a pogrom or terrorist activity) or a sustained series of violent acts that concatenate over a protracted period of time (as in civil war or genocide). Moreover, the severity of the violence involved in these acts differs significantly from one case to the next, ex- tending from minor property damage to thousands and even millions of lives. All of these issues beg serious attention and qualification from the researcher. To complicate matters further, in the study of interethnic collective violence more specifically, collective violence has tended to remain underspecified as an object of explanation; it has often been folded into the larger and more vague category "interethnic conflict" and thereby left undefined in its specific forms and in distinguishing it from other po- tential outcomes (that is, nonviolent modes of conflict, diffuse social vio- lence, and the like). Situations have often been coded dichotomously (con- flict/no conflict), and many scholars have viewed collective violence "as a degree of conflict, rather than as a form of conflict" (Brubaker and Laitin 1998: 425~. This has unfortunately served to cut the study of interethnic violence off from the study of other social processes that remain closely connected to it. Until the last decade the study of collective violence and the study of ethnic conflict remained largely unconnected; generic theories of collec- tive violence tended to ignore ethnicity or to subsume it within larger analytical categories of collective violence, assuming that there is nothing about the relationship between ethnicity and collective violence that would distinguish it from collective violence over other types of issues. At the same time, scholars focusing on ethnic conflict have tended to assume that interethnic violence flowed logically out of the intensity of cultural allegiance, and that these emotional attachments constituted the single, cohesive set of motivations for those perpetrating acts of violence. Evidence of other motivations (personal rivalries, revenge, power, and in some cases, respectability) has been too easily overlooked. What, then, can we identify as promising in terms of future research directions for collaboration, given the theoretical Babel, empirical com- plexity, and analytical muddle that we face in studying collective violence in multiethnic societies? We see a series of questions and themes that merit abiding or greater attention in the future and would therefore make promising topics for potential collaboration. THE ONSET, DURATION, AND TERMINATION OF SUSTAINED VIOLENT CONFLICT The theoretical and empirical research on civil war onset, duration, and termination is presently booming. Several groups of scholars have
42 CONFLICT AND RECONSTRUCTION IN MULTIETHNIC SOCIETIES produced datasets covering the universe of cases of civil wars; these stud- ies typically attempt to identify the correlates of civil war in a way that parallels the large sample study of interstate wars. Macro-level cross-national studies have so far produced a number of sometimes divergent results. There is one interesting and important re- sult upon which they converge popular grievances (coded in a variety of ways) appear to be a bad predictor of the onset of civil wars. Because most descriptive work on civil war onset focuses on cases where civil wars did erupt, they tend to endow grievances (repression, among others) with a causal force that disappears when the selection bias is corrected for. Inso- far as nationalism is part of the popular grievances, this finding under- mines much conventional wisdom about the causes of civil war; the same is true for ethnic antipathy and other culturally derived hypotheses. In- stead, opportunity costs of violence seem to play a more important role; the presence of difficult or rough terrain increases the likelihood of a rebellion, as does the presence of lootable resources, that is, high value, easily transportable goods. Diffusion and contagion effects have also been identified as potential culprits, but the available evidence so far fails to support this hypothesis. Last, regime type and ethnic fragmentation have been included in a number of studies, for which the results are curvilin- ear: Democracies and autocracies are more likely to avoid civil wars, as are highly fragmented and nonfragmented polities. A different line of macro-inquiry associates ethnic networks with civil wars because they can solve collective action problems and facilitate mobilization, such networks make civil wars more likely. However, im- portant methodological problems undermine the empirical testing of this hypothesis. Furthermore, with one notable exception, we lack work that systematically traces and identifies differences and similarities between ethnic and nonethnic civil wars. The central problem of macro-level studies is the general lack of microfoundations. Some studies do identify a number of micro-level mechanisms. Three promising lines of inquiry in this respect may be (1) economic theories of criminal behavior (pointing to informational asymmetries and other communication failures that lead to civil wars); (2) the ethnic security dilemma that may facilitate the mobilization of individuals by politicians; and (3) constraints to policing and the use of indiscriminate reprisals by incumbents that amplify small rebellions into full-fledged insurgencies. Recent large sample studies also address the duration of civil wars, and their termination and recurrence, starting from the empirical obser- vation that negotiated settlements tend to be less common in civil as opposed to interstate wars. This appears to be the case not necessarily because of indivisible stakes, irreconcilable differences, or high cost toler-
PRIORITY THEMES FOR RESEARCH ON COLLECTIVE VIOLENCE 43 ances but because of the impossibility of the warring parties making cred- ible commitments. Last, some studies have begun to look at third-party intervention and postwar settlements, particularly peacekeeping, peace- making, and partition (and find that they do not guarantee peace). The most promising line of inquiry in this area lies in the specification of better micro-mechanisms, as well as process-tracing case studies that take into account the large sample findings. In this respect, collaborative small sample studies that are carefully designed to take into account and test large sample findings hold particular promise. Collaborative research can also contribute to the design and implementation of systematic stud- ies that rely on subnational units of analysis, such as villages, municipali- ties, counties, and so forth; such studies combine a large number of obser- vations with stronger controls and allow for careful tracing of causal mechanisms. 1 ) . PARTICIPATION IN INTERMITTENT FORMS OF MOBILIZED INTERETHNIC VIOLENCE Much interethnic collective violence is intermittent and irregular in character, bursting out in intense but short episodes that may or may not lead to more sustained violent conflict. Indeed, in most societies at most times, strong taboos and institutional constraints of varying degrees of effectiveness exist against participation by individuals in acts of violence and even more so, in acts of collective violence. Yet, in most pogroms, ethnic riots, or other instances of intermittent acts of mobilized collective interethnic violence, violence enjoys substantial public support within specific micro-contexts and involves significant participation by societal members in these acts (though usually direct participation is confined to a minority of community members). Thus, an important part of the expla- nation of intermittent, mobilized forms of interethnic collective violence (such as pogroms, riots, and communal violence) is an understanding of how norms of interethnic peace and institutional constraints are set aside, and what cultural, social, or political circumstances might cause ordinary individuals to commit what in normal times would be, for most people, unthinkable acts. Here, research during the last decade has taken a number of different directions. Rational choice scholars have modeled how expectations of reasonable behavior by others can suddenly collapse, tipping toward spi- rals of mistrust and transforming violence into a seemingly rational out- come. Others have focused on the role of the media in shaping the percep- tions of individuals in ways conducive to mobilizing them into violent action against others, contending that the propagandistic manipulation of populations by media oriented toward ethnically segmented markets is
44 CONFLICT AND RECONSTRUCTION IN MULTIETHNIC SOCIETIES critical in setting off violence. Recent research has also linked ethnic riots with the absence of associational interaction across ethnic lines (social capital) and with the structure of electoral incentives of elites in the con- text of democratic regimes. Still others have focused on how norms of interethnic peace are policed and maintained, concentrating on the within- group enforcement mechanisms that sustain norms and keep spoilers of interethnic peace marginalized or on the ways in which government agents themselves have signaled acquiescence to or support of the use of violence, thereby encouraging an atmosphere of license and abuse. Fi- nally, others have focused attention on narratives of violence both the ways in which these are constructed and their contributory role to a larger discourse of metaconflict that itself becomes contributory to further vio- lence. Collaborative research could contribute to our further understanding of these phenomena in a number of respects. First, even within existing strands of research there has been a serious information problem that collaborative research could help address. Most of what we know about acts of mobilized interethnic violence comes from the testimonies of vic- tims or witnesses, not from the perpetrators whose behavior we seek to explain (for an important exception, see Tishkov, 1995~. This is problem- atic for trying to understand how individuals come to be drawn into acts of violence, for it can only address the issue of motivation indirectly, through the eyes of others. There is a particular need to bring to bear new types of individual-level evidence drawn from the testimony of perpetra- tors, such as police depositions or court testimony, triangulating these with other information drawn from victims or eyewitnesses. This would allow us to have a better understanding of the disparate motivations underlying participation in mobilized interethnic violence, and how these disparate motives become focused and mobilized within the context of violent action itself. Second, like acts of nonviolent collective action, most acts of violent collective action depend upon networks of friends, coworkers, and ac- quaintances for mobilizing individuals. In contrast to other forms of col- lective action, however, we know relatively little about the networks that underlie pogroms and riots and the process by which participants are identified and mobilized. Collaborative research could make a major con- tribution to our understanding of interethnic collective violence by focus- ing its attention on these networks and how they compare with other types of collective action networks. Third, a fruitful focus for potential collaborative research are those formal and informal mechanisms of ethnic in-group policing that can restrain violent entrepreneurs within a situation of rising tension and that uphold norms of interethnic peace. Most work has tended to concentrate
PRIORITY THEMES FOR RESEARCH ON COLLECTIVE VIOLENCE 45 on those cases in which violence has occurred rather than engaging the counterfactuals of violence, that is, where violence might have occurred but did not. But an examination of the mechanisms that have prevented potential violence from materializing in instances in which the threat of violence was real would have significance not only from a theoretical viewpoint but also from a practical one. Finally, by breaking down intermittent waves of collective violence into specific events, large sample event studies can help us understand the dynamics by which violence spreads (or fails to spread) temporally and across geographic locations. Here, the challenge has been less the measuring of events themselves than of relevant and meaningful inde- pendent variables that might be useful as explanatory variables. Access to better quality information concerning the local contexts in which violence occurs or fails to occur could produce more fine-grained and accurate quantitative models of the spread or containment of waves of intermittent mobilized violence. Media and demographic sources are simply insuffi- cient here, and access to police records would seem to be of critical value, both by furthering our systematic knowledge of the factors shaping the patterning of violent collective action and by providing qualitative in- sights into what occurs specifically within the event to turn the unimagin- ables of violence into the thinkable. TRANSNATIONAL DIMENSIONS OF COLLECTIVE VIOLENCE IN MULTIETHNIC SOCIETIES An increasingly important dimension of collective violence in the United States and the Russian Federation is the transnationalization of violence. In a number of cases, violence-producing networks cross state boundaries, connecting with violent entrepreneurs in third-party states, or at times even receiving support and inspiration from third-party states. Ethnic diasporas that cross state boundaries increasingly play a role in the production and prevention of interethnic violence. Some indeed argue that the issues raised by the treatment of ethnic diasporas abroad have grown central to the dynamics of interethnic violence in the late twentieth and early twenty-first centuries. In certain regions of the world (the Great Lakes region of Africa, for instance) violence has spread across state boundaries with relative rapidity. In the post-Soviet region this has been less true (despite the expectations of some to the contrary), though rea- sons have yet to be fully explored. Moreover, international organizations have made the prevention and settlement of interethnic violence, the sanctity of boundaries, and human rights major directions of activity and criteria by which to judge the pro- vision of aid or membership. Many transnational nongovernmental orga-
46 CONFLICT AND RECONSTRUCTION IN MULTIETHNIC SOCIETIES nizations (NGOs) also work in the area of violence prevention and moni- toring. Thus, in many respects, collective violence is no longer a phenom- enon of a single political order; it has grown increasingly transnationalized in character. Research on collective violence, however, has been slow to recognize this fact. While some work has focused on the role of inter- national organizations in regulating (or in some cases, condoning) inter- ethnic violence, relatively few works have focused specifically on the transnational spread of violence or on transnational networks of violent entrepreneurs. Any of these themes would make excellent topics for col- laborative research. RITUALIZED INTERETHNIC VIOLENCE Anthropologists and historians have stressed the discursive, sym- bolic, and ritualistic aspects of violence. There are two ways to approach ritualized interethnic violence: The first way is to "thickly" describe it; the second way is to attempt to define and measure it, compare it to non- ritualized violence, and explain how it is linked to the larger phenomenon of collective violence. Following the latter course, it is possible to intro- duce a distinction between instrumental and expressive violence. Expres- sive violence is not intended to be a means of accomplishing any other goal beyond the satisfaction of emotions associated with its use. Although expressive violence is often used to mean identity or sectarian violence- that is, violence directed against persons on the basis of who they are and not what they did or are expected to do this association is incorrect: Sectarian violence can be, and often is, instrumental. A source of confusion in distinguishing between expressive and in- strumental motivations is the following: Arguments referring to the ex- pressive and symbolic aspects of violence claim, implicitly more often than explicitly, that they are explaining the motivations of collective actors when in fact they only describe the way in which particular individuals participate in violence. However, the fact that violence may follow a par- ticular ritual does not necessarily imply that it is expressive. There is little doubt that individual motivations behind acts of violence can be and often are expressive. Sociological research on criminal violence has also pointed to expressive motivations. However, it is difficult, perhaps im- possible, to uncover with an acceptable level of accuracy individual mo- tives behind violent acts. Deducing motive from behavior is fraught with methodological problems. Motives are typically subject to (either strate- gic or unself-conscious) reinterpretation and ex post rationalization by the subjects themselves, and they can be highly fluid and mixed. More- over, the little available empirical evidence suggests that expressive moti-
PRIORITY THEMES FOR RESEARCH ON COLLECTIVE VIOLENCE 47 vations may be less widespread than we think. Insofar as expressive mo- tivations of violence reflect extreme personality features, such as sadism, the available empirical evidence suggests that people involved in the pro- duction of political violence, such as terrorists or leaders of political gangs, lack those features. Last, acts of violence can be performed in highly symbolic ways, yet still be instrumental. In short, it is problematic to explain social phenomena by reference to their manifestation among the states of individual consciousness rather than the social facts preceding them. A solution to this problem is to clearly define the level of analysis (individual or collective). Fruitful collaborative research on the ritualized dimensions of collec- tive violence could potentially focus on a number of themes: collective violence as a type of performance, with attention specifically to the rela- tionship between performers and audience in various contexts; the role of collective violence in the creation and maintenance of ethnic identity boundaries; and how rituals of collective violence are created and repro- duced. POLITICAL LEADERS, ENTREPRENEURS, AND SPECIALISTS IN VIOLENCE The first step is to specify the level of analysis. Elite and political leaders are overaggregate concepts that may include everything from the presi- dent of a country down to local actors. National elites are central in theories of civil war that point to the security dilemma. The key idea is that political elites that are losing power have an incentive to shift political competition from the dimension where they are losing (class, regional, or other) to the ethnic dimension. This argument is incomplete, insofar as it assumes that individuals (a) find this shift credible and (b) engage in violence. Thus, the necessity of disaggre- gation. We generally know little about political actors such as militias, para- military groups, insurgent organizations, and the like. In many cases these people have had connections to criminal groups, raising issues concern- ing the motivations of those directing violence. We need systematic an- swers to such questions as who are violent entrepreneurs, what are their goals, what kind of resources do they have, what are their relations vis-a- vis the underlying population. This is an extremely promising opening for theoretically informed and well-designed ground-level studies that could greatly benefit from collaboration.
48 CONFLICT AND RECONSTRUCTION IN MULTIETHNIC SOCIETIES REFERENCES Arendt, H. 1970. On Violence. New York: Harcourt Brace and Company. Bayart, J., S. Ellis, and B. Hibou. 1999. The Criminalization of the State in Africa. Oxford: James Currey. Beissinger, M. 2001. Nationalist Mobilization and the Collapse of the Soviet State. Cam- bridge: Cambridge University Press. Brass, P. R. 1997. Theft of an Idol: Text and Context in the Representation of Collective Violence. Princeton: Princeton University Press. Brubaker, R., and D. D. Laitin. 1998. Ethnic and Nationalist Violence. Annual Review of Sociology 24: 423452. Cohen, Y., B. R. Brown, and A. F. K. Organski. 1982. The Paradoxical Nature of State Mak- ing: The Violent Creation of Order. American Political Science Review 75: 901-910. Davis, N. Z. 1973. The Rites of Violence: Religious Riot in Sixteenth-Century France. Past & Present 59 (May): 51-91. Della Porta, D., ed. 1992. Social Movements and Violence: Participation in Underground Organizations. Greenwich, CT: JAI Press. Fearon, J. D., and D. D. Laitin. 1996. Explaining Interethnic Cooperation. American Political Science Review 90 (December): 715-735. Graham, H. D., and T. R. Gurr, eds. 1969. The History of Violence in America. New York: Bantam. Gross, J. T. 2001. Neighbors. The Destruction of the Jewish Community in Jedwabne, Po- land. Princeton: Princeton University Press. Hardin, R. 1995. One for All: The Logic of Group Conflict. Princeton: Princeton University Press. Horowitz, D. L. 2001. The Deadly Ethnic Riot. Berkeley: University of California Press. Kalyvas, S. N. 2000. The Logic of Violence in Civil War: Theory and Preliminary Results. Estudio/Working Paper 2000/151. Madrid: Instituto Juan March de Estudios e Investigaciones. Lake, D. A., and D. Rothchild, eds. 1998. The International Spread of Ethnic Conflict: Fear, Diffusion, and Escalation. Princeton: Princeton University Press. Tambiah, S. J. 1996. Leveling Crowds: Ethnonationalist Conflicts and Collective Violence in South Asia. Berkeley: University of California Press. Tilly, C. 1995. Popular Contention in Great Britain, 1758-1834. Cambridge: Harvard Univer- sity Press. Tishkov, V. A. 1995. "Don't Kill Me, I'm a Kyrgyz!": An Anthropological Analysis of Vio- lence in the Osh Ethnic Conflict. Journal of Peace Research 32 (2~: 133-149. Volkov, V. 1999. Violent Entrepreneurship in Post-Communist Russia. Europe-Asia Studies 51: 741-754.