Ethnic Conflict as a Global Social Problem

Thomas D Hall & Lester M Jones. Handbook of Social Problems: A Comparative International Perspective. Editor: George Ritzer. Sage Publication. 2004.

Ethnic conflict is a complex topic, fraught with a great deal of baggage and hidden meanings, including two very thorny issues: (1) what do we mean by “conflict”; and (2) what does “ethnicity” mean. Many debates and arguments about ethnic conflict find their roots in assumed meanings for these terms and the baggage they carry with them. Sometimes the debates resolve once all parties work with the same or similar definitions. The discussions about “ethnicity” are further muddied by the historical context or contexts to which they refer. Not only have the terms varied widely in social science use (Omi and Winant 1994), but also the actual phenomena have changed over time (Hall 1984, 1998).

A great deal of recent social science writing has been directed at unpacking or deconstructing the theoretical assumptions embedded in various definitions, and even “facts.” While often useful, such discussions can rapidly devolve into solipsistic musings that have little or no use in actual analysis. Yet, to ignore these issues is to invite misunderstanding by writers and readers of each other, and more significantly by social scientists of the very phenomena they are trying to comprehend. One way of escaping this morass is to include in discussions of definitions the theoretical baggage that accompanies them. This is often a very recursive process; once some definition is advanced, further investigations—theoretical and empirical—may call into question some of the assumptions upon which it was originally based, and indicate that some refinement is required. Still, definitions are tools to aid analysis, not ends in themselves. But, like all tools, they must be used cautiously, carefully, and with attention to the purpose for which they are employed.

I begin with some basic definitions that I refine in the course of this essay. I attempt to be clear about my assumptions so that readers who operate from different definitions and perspectives can still find some use in this discussion, even if they decide to transform some of it to fit their own understandings of the basic terms.

“Conflict” is an immense category for human disputes that range from relatively mild disagreements, such as the meaning of words, interpretation of events, and so on, to extremely violent attempts to eliminate another person or group of persons. Conflict can also range from blatant and overt, to very subtle and hidden. In general, lower levels of conflict are an improvement of more intense and more violent forms. Not all conflict is “bad.” On occasion it can be useful to all parties, more typically to various parties differentially.

“Ethnicity” is far more complex. A basic definition might be a group of humans who share significant elements of culture and who reproduce themselves socially and biologically. Typically, they share a sense of identity, which is typically, but not always, recognized by other ethnic groups. Thus, ethnicity is socially constructed in at least two ways. First, it is reproduced through socialization processes, wherein new members—again typically through birth, but at times through various forms of capture or adoption—learn how to act as members of the group. Second, it is socially constructed by self-recognition and recognition by others. In such social construction processes, the ethnic identity may as often be a consequence of other actions, including conflicts, as it is an end in itself. In the language of statistical analysis, ethnicity can be either a dependent or independent variable. In fact, most often it is both in complex recursive social processes that shape and reshape its content and meaning.

The rub, or difficulty, in all this is the rooting of the identity in biological reproduction, which is often assumed to be “natural,” “innate,” or even “immutable.” This generates at least two problems. First, the problem is that “race” is often assumed to be quintessentially biological, but is in fact often if not always socially constructed (Cavalli-Sforza and Cavalli-Sforza 1995; Gould 1996; Montagu 1997; Omi and Winant 1994; Smedley 1999). This is the familiar debate between “primordialist” or “essentialist” and “socially constructed” approaches to ethnicity and race. In situations where there is a widespread assumption that ethnicity is biologically rooted, typically in some aspect of phenotypical manifestation, such as skin tone, eye folds, tooth shape, hair, and so on, ethnicity has become racialized. Thus, “race” is both a special case of “ethnicity” and a problematic one: how and why did this or that feature or constellation of features come to be seen as “racial,” and why not others?

The second problem is that ethnicity is typically constructed historically. Where the historical depth is sufficient (what is sufficient is itself problematic), it can often appear “natural,” “primordial,” “essential,” or “racial.” In those times and places where humans did not move far from their home territories, there is in fact an association of phenotypical features, culture, and identity. This association, however, is not “natural” or “normal,” but an artifact of relative geographical stability, which itself is historically rooted. However, ethnic groups in fact change and transmute through time, fragmenting, amalgamating, and exchanging members. Where these processes are slow, they again may appear to be racial or biological because their social and historical construction is not readily perceptible to a casual observer. Two processes are especially revealing. First is the exchange or migration of individuals or families across boundaries, even while the boundaries remain stable (Barth 1969; Haaland 1969). Indeed, Frederick Barth (1969) goes so far as to claim ethnicity is demarcated by its boundaries, not the “cultural stuff” it contains. Second is ethnogenesis, the formation of a new group from parts of other groups, or families, or individuals (for many examples, see Hill 1996). Ethnogenesis often simultaneously entails ethnocide, especially when two or more formerly distinct groups merge into a larger group. At other times, ethnogenesis is the result of one group splitting into two or more separate groups.

A problem here is that these processes are slow and often contested, sometimes violently, with a great deal of slippage back and forth along the various dimensions of change. Put alternatively, ethnic boundaries are nearly always fluid, fuzzy, and permeable. One consequence of the rise of the modern state with precisely defined geographic boundaries is that that precision has been metaphorically extended to all sorts of human groupings whose boundaries are anything but precise. Indeed, even those rare cases where ethnic boundaries do appear precise are really only a snapshot or freeze frame of larger, slower processes of continual change.

Finally, conflict itself can be ethnogenetic. Two groups, defined arbitrarily, or by the terms of a conflict if the conflict persists sufficiently long, may come to define themselves, and each other, as ethnically distinct. In the modern world (the last half millennium or so), this most often occurs with the splitting of states: Norway from Sweden, the two Germanies, Vietnams, or Koreas. As Germany illustrates, the splits can be temporary, and as Czechoslovakia illustrates, the reverse, amalgamations, can also be temporary. In premodern times, especially among nonstate societies, these processes were much more fluid in that there was little interest in precise demarcation of boundaries. None of this, however, is meant to imply that individuals were confused about who they were or their group membership(s), although they may often be contested.

Ethnic Change and Conflict are “Normal”

According to Ted Gurr and Barbara Harff (1994; Gurr 1993, 2000; Harff and Gurr 1998), ethnic conflict and ethnic violence have been the leading cause of casualties in warfare over the last few decades. Wilma Dunaway (2003) argues that the rate of increase of ethnic conflict has not grown since the end of the Cold War (1991), but that ethnic conflicts have become more costly for states and for capitalist enterprises. Still, for many observers this recent prevalence of ethnic conflicts is a puzzle. My argument here is that this puzzle arises from an incorrect framing of the problem. The real puzzle is why there is not far more conflict. Following Gurr’s (1993, 2000) pioneering work on ethnic conflict and his Minorities at Risk Project, there are on the order of 6,000 identifiable ethnic groups in the world today (at the beginning of the twenty-first century). Yet there are fewer than 200 recognized states in the world. That is to say that ethnic diversity is quite common, and hence at least the potential for conflict. Much of Gurr’s work has been directed at uncovering the situations that make conflict more, or less, likely. I return to this issue below. For now I continue to explore this diversity issue a bit.

If ethnicity were evenly distributed (a historically rash assumption), then every state would have approximately 30 groups. Still, as of the 1980s, the number of states with 95 percent or more of the population in the largest ethnic group is quite small, 30, with half those in Europe. Furthermore, given the movement of workers and refugees, the number is probably lower in the first decade of the twenty-first century (Laczko 2000:133). In other words, the so-called nation-state is a chimera. This then raises the puzzle, Why is the ethnically pure nation-state so widely held as an ideal? The answer is complex, rooted in both recent trends and deep historical processes.

One of the consequences of modernization, democratization, industrialization, and globalization has been an increasing uniformity in ideals of how to form a state. John Meyer and colleagues (1997) argue persuasively that were a new state to form today, we could predict in considerable detail how it would be organized politically and bureaucratically. This is a consequence of an emerging global culture that has accompanied the increasing tightening of global capitalism on the world economy. Since the modern state is ideally (ideally in the Weberian sense of an ideal type) ethnically uniform, that ideal has spread as well. But this begs the question of how that ideal arose.

In order to explore this question, it is necessary to revisit the long history of human states. One of the best sources on this is William H. McNeill (1986) in his series of lectures on “Polyethnicity.” Whether one follows more or less conventional world history (which is far from conventional and undergoing considerable intellectual foment; see Journal of World History over the last decade), or follows various accounts of long-term social change (Chase-Dunn and Hall 1997; Denemark et al. 2000; Frank and Gills 1993; Hall 1998; Lenski 1966, 1976; Nisbet 1969; Sanderson 1995, [1995] 1999), it is clear that since their first invention some 5,000 years ago, states have steadily expanded and progressively gobbled up neighboring groups, whether they be other states or any of a wide range of types of nonstate societies. Whether such expansionary conquests are a form of ethnic conflict is itself a minor puzzle. Historically, ethnic differences have not played a major role in analyses of state expansion. They have been seen as more of a side effect. Still, there is an ethnic aspect to such conquests, and especially to attempts to rebel and break loose from a conquering state.

In expanding, even those primary or autochthonous states quickly came to absorb peoples who were ethnically different. In McNeill’s terms, states quickly become polyethnic, or what today might be called ethnically plural or multicultural. Early states often were not concerned with ethnic or cultural uniformity, save possibly within a narrow elite. Rather, they focused on the efficient collection of tribute, but beyond that they tolerated a great deal of pluralism. Typically, however, ethnic groups within a state were ranked hierarchically (Laczko 1990), with the ethnicity of the dominating elite at the peak. Even without a policy directed toward doing so, this created some pressure for lower-ranked groups to assimilate into or acculturate toward the dominant culture. Other factors like location and occupational specialization, crop specialization, and so on created pressures to maintain differences. The net result is that the multiethnic state is “normal” in the statistical sense, and has been for five millennia. This deepens the puzzle of the chimera of the ethnically homogeneous nation-state.

Many writers, ancient and modern, have noted that a plural or polyethnic state is more difficult to administer. Uniformities in language, measures, monetary denominations, and cultural practices for commercial exchanges make exchange much easier. Indeed, one way of overcoming these obstacles was a trade diaspora (Curtin 1984) in which members of one ethnic group moved to another state and formed an enclave with which coethnics traded. Members of the enclave then conducted exchanges with the host society. In sum, for many millennia, states maintained a dynamic balance between the pressures for uniformity and for difference, and thus stayed plural.

Following McNeill (1986), this situation began to change in the seventeenth through nineteenth centuries. Two sets of factors contributed to this. One was more or less a historical accident. As European states were consolidating out of congeries of competing petty kingdoms and merging variant dialects, the pressures from administrators and merchants for uniformity increased (Bartlett 1993; Tilly 1975, 1990). Historians writing during those times turned to ancient writings to learn how to build states. They read how the ancients bemoaned the diversity of ethnic groups (usually described as nations or peoples) as these early classical states formed. This coincidence, of reading ancients at precisely the time in their own histories that they were undergoing similar problems, reinforced their view that ethnic uniformity was vital to building a successful state.

The second factor was a change in the nature of warfare. As states in Europe became larger and as guns began to come into common use, states needed larger and larger infantries. Eventually, the need was for armies larger than what could be drawn from elite classes or mercenaries. However, arming peasants has always been dangerous. Their loyalty needed to be ensured. One means of doing so was the conceptualization of citizenship and patriotism for the fatherland or motherland. But this is nigh impossible when loyalties are to localities, such as villages or small districts. Larger identities must be built. Thus begins a process of subsuming local identities within a larger state identity, and the birth of the nation-state as an ideal.

This is the process that Benedict Anderson (1991) has called the “imagined community” and that Eugene Weber (1976) described in Peasants into Frenchmen. Peter Sahlins (1989) shows in some detail how these processes worked differently in the Pyrenees as the boundary between France and Spain came to be precisely defined. All these, and many other accounts, demonstrate that this national unity is a very slow process, and that it inherently entails a good deal of ethnic conflict as identities are reshaped and reformed. That is, the rise of the nation-state as an ideal is itself one of the causes of ethnic conflict. Phrased alternatively, nationalism, or patriotism, inherently generates ethnic conflict.

Intellectually, as this ideal of the nation-state came to be seen as “normal” or “natural,” the persistence of ethnically distinct groups within states came to be seen as problematic. The ideal became so widely accepted as part of the modernization paradigm that many social thinkers expected ethnic conflict to decrease. Hence, its continuation, and indeed its perceived rise in recent decades, became a puzzle.

There are two levels of Eurocentrism embedded in this discussion. First, the emphasis on ethnic homogeneity clearly does emerge from the dominance of European and European-derived states in the modern world. Second, and more subtle, the relatively low attention among European and United States social scientists to ethnic processes in Asia, especially in the premodern world, may overemphasize the European versions of states in this analysis. One may consider a series of counterfactual scenarios. What if Chinese, Arab, or Indian sailors had “discovered” and colonized the Americas? Would the state today have such an emphasis on ethnic homogeneity? While fundamentally unanswerable, such a question does serve to highlight how both historical facts and the study of history may have distorted our understanding of the roles of states in ethnic conflict. One obvious remedy is more detailed, historical studies of state relations with surrounding peoples in Asia and elsewhere to develop a more robust understanding of the various roles of states and state organization (tributary, industrial, capitalist) in ethnic relations. A glimmer of an answer can be uncovered in closer examination of the history of state expansion into the territories of other peoples.

State Expansion and Ethnic Conflict

Before examining this rise in ethnic conflict further, it is necessary to reexamine state expansion, this time with a focus on interactions with nonstate societies or indigenous peoples (Hall and Nagel 2000). For a variety of reasons (that need not be examined in detail here), relations with nonstate societies, often glossed as “barbarians,” have not been considered under the heading of ethnic conflict, at least not until the twentieth century, when some indigenous populations, like the first nations in North America, have come to be seen as ethnic minorities. There are at least two major differences in how these groups have related to states through time.

First, they have been more subject to pressures to assimilate, to become “civilized,” that is, to take up the sedentary, agricultural life of people in state societies, including especially formal, bureaucratically organized political structures (for cogent review see Maybury-Lewis 2000). States from China (Barfield 1989, 1990, 2001), to early Indian states (Srinivas 1996), to Rome (Wells 1999a, b), to Spain in the Americas (Hill 1996; Weber 1992; Weber and Rausch 1994), to the United States (Hoxie 1984), and to Africa (Macharia 2003) have tended to view nonstate groups as politically unorganized, or “acephalous,” or without reason. States have a very difficult time dealing with the informal, charismatic, fluid organization of many nonstate societies. Consequently, they often pressured such groups to take on formal organizations or to assimilate, typically as peasants, into the state. Indigenous peoples often resisted fiercely, and not infrequently fled to hinterlands that states found undesirable or unusable. However, when they have been absorbed, and if they did take up the culture of the encroaching state, they have been destroyed as a separate ethnic group. This is ethnic conflict in the extreme, annihilating a group identity, even if the group members survive as individuals, a process labeled ethnocide.

Yet, there is a second, more subtle difference. The interactions with states, typically but not universally, have been violent and very conflictual and have had serious social structural effects far beyond the border or frontier zone of immediate fighting. Brian Ferguson and Neil Whitehead (1992 a, b) refer to this much larger zone of volatile change as the “tribal zone.” Here “tribal” needs to be in quotes because it has several meanings. Often the label “tribal,” like “primitive,” is a not so subtle justification for confiscation of land or mineral rights or seizure of some other resource. Often it is used as justification for forced assimilation into state culture. Incidentally, this is why many indigenous peoples in the United States have insisted on changing their names, such as from Navajo Tribe to Navajo Nation. It is also why many of these same groups now insist on using their name in their own language rather than the names by which they have been known, such as Navajo Community College becoming Diné College, Sioux returning to Lakota, Nakota, or Dakota, or the League of the Iroquois to Haudenosaunee.

Even where used analytically, “tribal” is a catchall term that is no more precise than “nonstate society.” Finally, even if “tribal” is precisely defined to refer to a form of social organization with semi-formal leadership, between band and chiefdom, the classification itself masks the historical nature of most such tribal organizations. Ferguson and Whitehead (1992a, b) follow the lead of Morton Fried (1967, 1975) in arguing that most “tribes” are in fact a product of intense interactions of nonstate societies with expanding states. “Tribes” may be formed by the breakup of organized chiefdoms, or the unification of groups of band societies. In short, most often they are not a “stage” in an evolutionary sequence, but a by-product of the expansion of states.

These kinds of transformations often ripple far beyond the zone of immediate contact with the expanding state. They disrupt traditional organizations and reorganize, often along new lines, the groups that exist in the “tribal zone.” Among other consequences, this means that even very early, first-hand observations of “tribal” life seldom reflect some pristine condition, but rather are products of long, intensive interactions. State-level observers who are literate and leave records seldom arrive in remote areas that have not already been incorporated into the tribal zone. David Anderson (1994), in reconstructing cycling among chiefdoms in the southeastern United States by reexamining documents from the DeSoto expedition in the sixteenth century, provides an exemplary model of how to interpret such accounts to overcome the tribal zone effect. The collections compiled by Ferguson and Whitehead (1992a) and Hill (1996) provide numerous examples of these effects, although those in Hill focus on ethnogenetic effects.

Thus, a great deal of ethnic change and ethnic conflict is engendered in the tribal zone by state expansion. Furthermore, much of this conflict is invisible, or nearly so, in the historical record. The accounts in Ferguson and Whitehead (1992a), Anderson (1994), and Hill (1996) are notable precisely because they are able to penetrate this veil by an astute use of documents often in combination with archaeological evidence. A key implication of these studies is that states everywhere, throughout their five millennia of existence, have had these kinds of effects on nonstate societies. Thus, this is probably the oldest and most common form of ethnic conflict. Furthermore, it has not solely been a result of European expansion. Let me hasten to add that by this I do not mean to claim that European expansion and colonialism or capitalism has been somehow less devastating than has often been argued. Rather, the root of such conflicts is in the state qua state. To be sure, the form of the state—tributary, mercantile, capitalist, or industrial—shapes in profound ways these general processes, but it is stateness that is key. If one compares the accounts of early Spanish colonialism in the collections compiled by Donna Guy and Thomas Sheridan (1998a, b) or David Weber and Jane Rausch (1994) with accounts of early Rome by Peter Wells (1999a, b, 1992), Dyson (1985), Mattingly (1992), or Miller (1993), one is immediately struck by the continuities in policies and strategies, and less by the differences. Differences emerge with the rise of modern, industrial, capitalist states and derive in no small part from the vastly superior power—military, political, and economic—in comparison to indigenous peoples (see, e.g., Hall 1989). Thomas Barfield (1989, 1990, 2001) finds similarities among the states that dealt with Central Asia steppe pastoralists over several millennia.

If these types of conflicts are ruled out of consideration with respect to ethnic conflict, the universe of conflicts becomes much smaller, and much more distorted. By examining such conflicts, wider similarities and differences can emerge. Furthermore, if analysts ignore these millennia-old processes, they run a high risk of assuming that the processes occurring in the last 500 years, or even the last 100 years, are typical and normal, and thus cannot analytically separate effects of states from effects that are due to the specific type of state: tributary, mercantile, industrial, or global.

Over these five millennia, some human groups have become quite large and have destroyed or absorbed many other, smaller groups. There are a variety of ways that this can happen, with probably more variation in the consequences for indigenous groups than for formerly autonomous states. It is useful to distinguish among different fates. Detailed histories over millennia of Chinese absorption of nonstate people and of analogous processes in South Asia would no doubt yield further insights into the varieties of these processes.

Genocide, Ethnocide, and Culturicide

There are many ways an ethnic or an indigenous group might be destroyed. Genocide, ethnocide, and culturicide all share an element of intentional destruction of a group. Genocide is the most familiar, and certainly the most brutal: the outright murder of all members of an identifiable descent group, or the attempt to do so. In contrast, ethnocide and culturicide involve attempts to destroy a group’s identity, and/or culture, without necessarily killing individual human beings.

Ethnocide is an attempt to destroy the identity of a group. In its ideal typical form, it would entail full assimilation of individuals into the dominant group, although some cultural elements might still persist (see, e.g., Ortiz 1984, 1985). A key feature here, besides the obvious internal contradiction of destroying an identity but allowing some of its “content” to remain, is that the group, qua a group, disappears. In contrast, culturicide is an attempt to kill a culture, whether or not its members survive and whether or not they retain a separate identity (Clastres 1980; Fenelon 1997, 1998, 2002). A notorious example is that of Richard Pratt, founder of the Carlisle Indian School, whose explicit goal for the school was “to kill the Indian, but save the man” (Adams 1995). While Pratt seems seriously retrograde at the beginning of the twenty-first century, he was a humanitarian reformer in the context of the late nineteenth century, when many still called for outright genocide (Adams 1995; Hoxie 1984). Here the separate identity may survive, but the cultural content is eliminated.

Ethnocide and culturicide are somewhat overlapping processes. Each process, and indeed which process operates, is largely conditioned on the degree to which group distinctions are racialized. Obviously, to the degree that a group is marked by readily visible phenotypically distinctive features, maintenance of identity in the face of destruction of the culture is easier and more common.

Ethnocide is closely similar to the older concept of assimilation, in which one group adjusts its culture to become progressively more like that of another group. The difference is the clear intent to eliminate the group identity. Culturicide, on the other hand, does not need to destroy the identity as long as the “content” of the identity becomes nearly the same as that of the dominant group and thus subordinate to the socioeconomic goals, practices, and ideologies of those in power.

So how is it that indigenous peoples have resisted attempts at ethnocide or culturicide? One way, albeit unplanned, is by remaining small, and therefore relatively nonthreatening, at least to the point that the costs of pursuing ethnocide or culturicide have not been worthwhile. Another has been via relative isolation. This, however, is most often an accident of history—location within a region of little interest to any states or lack of resources that are seen as valuable to states. Another is building upon, or using, a recognized land base to keep the community viable.

Some forms of resistance are covert, echoing Scott’s (1985) concept of “weapons of the weak.” Many forms of resistance transmute forms or mas-querade as something else. For instance, the events in Chiapas, Mexico, have often been cast in the light of a regional-, a peasant-(and hence a class-), or a caudillo-driven rebellion. They are less often discussed as an indigenous Mayan rebellion (but see Boswell and Chase-Dunn 2000; Collier 1999; Katzenberger 1995; McMichael 2000; Morton 2000). Movements in the United States, such as the American Indian Movement (AIM), are often seen solely in terms of localized ethnic, urban, or racial rebellions. Indigenous resisters are often far ahead of those who report about them—connected via the United Nations, a large variety of their own organizations, and the Internet (Langman, Morris, and Zalewski 2003; Smith and Ward 2000; Wilmer 1993). Anna Tsing’s (1993) In the Realm of the Diamond Queen can be read as an account of ways in which local people, in this case Dayaks in Kalimitan, resist state incorporation. Indeed, Tsing’s account and Stoler’s (1995) account of plantation resistance in Sumatra or Peluso’s (1992) account of forestry “management” in Java have as key components—if not the driving component—the struggle for survival of indigenous cultures, identities, organization, and economies. This applies to indigenous peoples throughout Southeast Asia (e.g., Sponsel 2000a, b; Steinberg 1987) and Asia in general (Barnes, Gray, and Kingsbury 1995). In other cases, traditional culture and organization itself is a resource that facilitates resistance and survival (Champagne 1989, 1992; Fenelon 1998). Virtually every place in the world indigenous resistance struggles continue (Gurr 1993; Hall and Fenelon 2003), even in Europe, as, for example, among the Saami (Eidheim 1969). Kurdish activities in West Asia and Miskito resistance in Nicaragua have long been noted as indigenous movements (Gurr and Harff 1994).

Another form of resistance has been conscious efforts to maintain “traditional culture” not as static and unchanging but rather as evolving according to the desires of group members resisting domination, rather than in accord with the desires or directions of outsiders. That is, “traditional culture,” like all other social forms and structures, evolves and changes continuously, if sporadically and unevenly (Fenelon, 1998:27-30, 72; Smith and Ward 2000). An extension of culture maintenance is culture building. For instance, there are 33 tribal colleges in the United States (American Indian Higher Education Consortium 2000; Boyer 1997).2 These are institutions of higher education, typically equivalents of junior colleges, run by various Native American groups. They differ from the typical U.S. junior college in the number of courses they offer that promote traditional culture, language, crafts, and customs. In some cases, language programs have been aimed at reviving or reinvigorating a language that has fallen out of use. Indeed, these are often their key missions. That is, tribal colleges are often one institutional means of preserving and enhancing “traditional cultures.”

Resistance can also take the form of building other localized institutions that conform to traditional cultural values. The Diné (Navajo) have several such institutions. The Navajo Nation Police, while acting much like any other rural police force in the United States, is also culturally sensitive to Navajo traditions and works within them. More direct are the “Peacemaker Courts,” which pursue resolution of disputes among Navajos through means that are in accord with Navajo concepts of harmony, avoiding the adversarial techniques of U.S. courts.

Other forms of resistance are less institutionalized, but nonetheless important. Carol Ward and colleagues (2000; Baird-Olsen and Ward 2000) analyze how women among the northern Cheyenne have adapted conventional 12-step programs that address alcohol abuse or spouse abuse to Cheyenne culture, promoting Cheyenne family values. Miller (1994) and Chiste (1994) discuss the ways in which Native women are producing new feminisms within changing tribal governments. Another common institution among Native Americans in the United States is the maintenance of matrilineal family systems, including especially ownership of property. This often comes at a great price, as missionaries and bureaucratic functionaries have repeatedly attacked matrilineality as “barbaric,” un-Christian, or chaotic. Native American feminism often organizes in ways that oppose more mainstream feminist movements. Typically, Native American feminists focus on issues of identity and cultural preservation as prior to more narrowly feminist concerns (Jaimes and Halsey 1992; Shoemaker 1995).

Religion can be yet another form of resistance. Maintenance of religious practices that anthropologists are wont to call shamanism, again over massive attempts to destroy these practices, asserts an entirely different way of approaching the supernatural and the sacred. Today, as New Agers have begun to practice various forms of shamanism, Indian groups have protested such attempts to appropriate Native traditions (Churchill 1994, 1996; Rose 1992).

The revival of older traditions, such as the Sun Dance (see, e.g., Fenelon 1998:114, 288-94; Jorgensen 1972), can be another form of this resistance. These revivals hark back to many revitalization movements: the Longhouse religion of the Iroquois (Wallace 1969), the Ghost Dance movement (Brown 1976; Champagne 1983; DeMallie 1982; Landsman 1979; Thornton 1981, 1986), and the Native American Church (Aberle 1982; La Barre 1964; Stewart 1987), and so on. These movements, all of which are somewhat syncretic, preserve many traditional values and have all met with some success in combating the destructiveness of incorporation into the capitalist world-system. The Longhouse religion has been a source of strength among Iroquois. The Native American Church (NAC; also known as the peyote religion) has been very successful in helping individuals recover from alcoholism. Also, NAC has won several court battles that allow members to use peyote (Iverson 1999:181-2).

All of these religious traditions are vastly different from various monotheisms found in the states of the modern world-system. Their survival and growth is an important form of resistance to the ideologies of the modern world-system and to pressures for increasing homogeneity of culture from various globalization processes.

Probably the most significant forms of resistance are the various ways that resources are managed collectively, for collective good. Here one must be careful not to read this as conventional “public goods” administration. This goes much further, in collective ownership of goods—land and livestock most commonly—that are typically individually, privately owned commodities in the capitalist world-system. Such challenges go to the heart of fundamental assumptions about capitalism, neoliberalism, and private property. Given these challenges, it is all the more surprising that stronger efforts to eliminate them have not occurred. Franke Wilmer (1993) argues persuasively that indigenous groups have succeeded in part because denial of their rights to sovereignty would entail dismantling the entire international relations regime based on the Peace of Westphalia (1648), which is the basis of the contemporary nation-state system. Sometimes too they succeed because these challenges are included or embedded in other challenges.

However, some of these forms of resistance preserve not only symbols but material practices that contradict how capitalism is practiced. They represent alternative ways of organizing human life. What is far from clear is whether these too can ultimately become “merely symbolic.” Is an American Indian nation that insists on tribal sovereignty and administers resources according to principles of collective rationality, yet externally participates in a capitalist world-system according to capitalist principles, resisting globalizing capitalism, or slowly evolving into an alternative form of capitalism? Clearly, all forms of human organization are changing under the pressures of global capitalism. There is no reason to expect that indigenous groups or that ethnic groups will not experience and attempt to shape the processes of change. In many cases, the nature of those changes will not be clear for some time. Furthermore, conflict both causes and is a consequence of these changes.

States, Ethnic Minorities, and Globalization

While the analysis of globalization is controversial in many ways (e.g., Hall 2002a; Manning 1999; Sklair 2002), there have been two modern waves of globalization of trade in the late nineteenth century spilling into the early twentieth century and the late twentieth century spilling into the twenty-first century (Chase-Dunn, Kawano, and Brewer 2000; Clark 2002; McMichael 2000; Rodrik 1997). Both have generated waves of immigration (Dunaway 2003; Gabaccia 1994). Furthermore, various conflicts, many of which were ethnic in character, have generated immense numbers of refugees (Dunaway 2003; Gurr and Harff 1994; UNHCR 2003). These, in turn, create potentials for many future conflicts as immigrants are subjected to pressures to assimilate, or return home, and as refugees transform from temporary “guests” into more or less permanent residents.

These conflicts and potential conflicts differ in many ways from conflicts surrounding indigenous peoples. The key difference is that immigrants and refugees have a home state that has the potential of supporting their claims for preservation or even autonomy in the international arena. As Gurr and Harff (1994) argue, the severity of such potential conflicts is a function of the relative power of the states involved, with stronger states posing more of a threat than weaker states. Concerns over potential refugees and potential rebellions play a significant role in global politics. In early 2003, a major sticking point in negotiations between Turkey and the United States over support for the war against Iraq revolved around the roles of Kurds during and after the war. Turkey is deeply threatened by any Kurdish independence movement, since so many Kurds live in Turkey. Within Iraq, the Kurds offer potential resistance to Saddam Hussein, but are problematic in a postwar Iraq. Their useful resistance could all too easily turn into a potentially destabilizing ethnic separatist movement or movements.

Movement of peoples, either as voluntary migrants seeking improved living conditions, semivoluntary migrants pushed from their home by poor economic conditions and attracted to other areas by the promise of work, or as refugees from various conflicts—ethnic or not—are leading to increasing ethnic diversity in many states. Such movements greatly increase the number of potential conflicts.

Additional problems in understanding ethnic conflicts stem from their mutability and efforts by some leaders to “ethnicize” conflicts. Franke Wilmer (2002) among others argues persuasively that the conflicts in the former Yugoslavia were not rooted in ancient ethnic antagonisms, but in fact were fomented by leaders who sought deeply emotional issues around which to mobilize their followers in order to maintain or enhance their political positions. Thus, what were at root political conflicts became ethnic, and in some cases cast as racial. According to Malcolm (1994), this is why libraries, monuments, and other historical repositories so often became military targets—in order to forcibly rewrite history to coincide with current political goals.

Many observers have noted too how Europeans and European-derived settler states became vitally concerned over the former Yugoslavia, yet were loathe to become involved in the even more intense and vicious ethnic conflict that took place in Rwanda. In both cases, it is arguable, if not completely demonstrable, that the conflicts were at root political and economic, not ethnic, but quickly became ethnicized. Why that is so remains problematic, although many have speculated that race has played a major role while others argue that it is the proximity of the Balkans to Europe. In many ways, these debates are emblematic of continuing debates about the origins and causes of ethnic conflict.

Causes of Ethnic Conflict: Current Research

Much of the older research on ethnic conflict focused on either nationalism (Connor 1994; Horowitz 1985) or ethnic competition and mobilization (Olzak 1992; Olzak and Nagel 1986; Pincus and Ehrlich 1994) or was a critique of international relations theory (Carment and James 1997). Current research on ethnic conflict can be divided into five types: (1) quantitatively based research, (2) world-systems analysis-derived research, (3) case studies, (4) studies of indigenous relations, and (5) studies that seek to bridge psychological and macrostructural approaches. In practice, these categories overlap extensively. In this very brief summary, I discuss and cite only very recent works, but ones that include extensive bibliographies to lead interested readers deeper into any or all of them.

Among the quantitative approaches, the most ambitious and global has been the work of Ted Robert Gurr and colleagues, called Minorities at Risk. In addition to works already cited (Gurr 1993, 2000; Gurr and Harff 1994; Harff and Gurr 1998), there is an elaborate Web site ( maintained at the University of Maryland devoted to the project that provides links to published papers, working papers, and several datasets. While much of the Minorities at Risk Project is directed at various quantitative studies, it is extensively supplemented with case studies, especially in Gurr (2000) and Gurr and Harff (1994). While the data compiled by the project is the most comprehensive compiled in one place, it does not claim to be universal. The disclaimers and codebooks make clear the limitations of the data.

Several studies bridge the quantitative and world-systems approaches. Especially intriguing is the work of Olzak and Tsutsui (1998). Using Minorities at Risk data, they show that the risk of ethnic mobilization is not uniform across peripheral areas as some world-systems analysts have argued. Rather, they show that “peripheral countries with more ties to intergovernmental organizations have significantly lower levels of ethnic violence” than those without such ties. Furthermore, those with “more memberships in international organization experience a significantly higher magnitude of ethnic nonviolent protest” (1998:691). This research robustly underscores the claims made by Gurr and Harff (1994) and world-systems analysts that ethnic conflicts cannot be understood as strictly local phenomena, but are intimately linked to international and global processes.

Wilma Dunaway (2003) argues that while there has not been a precipitous rise in ethnic conflicts since the end of the cold war, nonetheless, ethnic conflict is becoming more and more troublesome for the entire capitalist world economy. In major part that is because it is becoming more costly, both politically in terms of creating instabilities and economically by disrupting markets and commodity chains. Of equal significance, she argues that both indigenism and ethnification are inherent contradictions in the modern world-system and that both have the potential of accelerating counter-hegemonic movements and pushing the entire system toward transformation. This article links four of the five types of research together conceptually.

Other world-systems research focuses on the antiquity of ethnic conflicts and their roles in the transformations of older, precapitalist, or pre-ca. 1500 C.E. world-systems (Hall 1998). In two articles (Hall 2000c, 2002b) I link ancient and modern patterns of ethnic conflict by examining the complex roles of frontiers in ethnic processes—transformations, ethnocide, and ethnogenesis on frontiers. I argue that frontier zones, and their attendant ethnic conflicts, have often played an important role in system transformations, suggesting that Dunaway’s argument is not unique to the modern world-system.

Case studies of ethnic conflict abound. As noted, Gurr (2000; Gurr and Harff 1994) argues that quantitative studies must be supplemented with detailed case studies in order to understand the nuances of ethnic conflict. Furthermore, much of the Minorities at Risk data comes from culling numerous case studies of ethnic conflicts. A combination of quantitative and case studies provides a much more robust and subtle understanding of ethnic conflicts.

As noted, several studies emphasize indigenous conflicts as a form of ethnic conflict. While much of the current research sees indigenous conflicts as a significant part of ethnic conflict, there is also an extensive literature on indigenous peoples and their struggles. Al Gedicks (1993, 2001) emphasizes environmental struggles. Fenelon and I (Hall and Fenelon 2003) argue, much like Dunaway (2003), that indigenous struggles are particularly challenging to the contemporary world-system. Also, I (Hall 2000c, 2002b) argue that frontier regions are particularly salient for studying such processes. Indeed, some of the best world-system-based case studies focus on indigenous peoples (Dunaway 1994, 1996, 2000; Faiman-Silva 1997; Fenelon 1997, 1998, 2002; Hall 1989; Hämäläinen 1998; Harris 1990; Himmel 1999; Jones 1998; Meyer 1994; Pickering 2000, 2003; Volk 2000; Ward et al. 2000). Most of these studies focus on peoples indigenous to the Americas. Smith and Ward (2000) concentrate on Australia, but without a world-systems approach. Hill (1996) focuses on South America. Alison Brysk (1996) focuses on the “internationalization of indigenous rights.” In general, literature on indigenous peoples is strongly regionalized, in part due to the legacies of anthropological research and in part because of the extensive investment in time required by scholars to learn about a particular region.

Finally, there are studies that seek to bridge the psychological and the macrostructural approaches. Wilmer (2002) advances a complex thesis, at least with respect to the former Yugoslavia, that because identity is embedded in early childhood and because it is often entangled with separation anxieties of males from their mothers, ethnic identity issues carry both much more emotional baggage and have the potential under specific child-rearing practices to dichotomize and demonize “the other,” thus promoting extreme violence. Her argument is far too rich and nuanced to explicate fully here, but it is one way of linking very micro and social psychological processes with much larger macro interstate and interethnic processes.

Whether or not one follows Wilmer’s or others’ accounts, it is clear that conflict can become ethnicized, or can even be ethnogenetic. This complicates immensely the task of sorting out whether or not ethnic conflict has increased since the end of World War II or since the end of the cold war.

Dan Chirot and Martin Seligman (2001) provide other ways to bridge the micro and macro approaches. These tend to focus on the roles of identity and its linkages to economic and political processes. Articles examine the roles of the media and the creation of a climate of fear. The collection is notable for its interdisciplinary approach to ethnic conflict. Indeed, articles in this collection fall into or straddle all of these categories, including several interesting case studies.

While there is already a rich and complex literature on ethnic conflict, much work remains to be done. Globally, coverage is far from even. Many more case studies from other regions are needed, especially from Asia. As these become available, many of the existing theoretical approaches will need to be modified, if not replaced wholesale.


Ethnic conflict is an immensely complex social phenomenon, made all the more so by the common tendency of conflicts to have many causes, some of which appear to be other than what they are, or that are intentionally manipulated by various actors. What recent events do demonstrate is that ethnicity, and race, are issues that are not disappearing and becoming less important. If anything, ethnic conflicts will remain a major source of intergroup violence in the coming decades. Yet, framed in longer historical perspectives, this is not all that surprising.

First, states have been absorbing and transforming formerly autonomous groups for millennia. To be sure, the frequency, density, intensity, and means by which these changes have occurred have changed considerably throughout history. Thus, any ethnic conflict must be analyzed and understood in its deep historical context. Second, ethnic mixing is “normal,” not ethnic “purity.” Third, ethnicity can be both the cause of and a consequence of conflict. Fourth, to ignore conflicts with indigenous peoples as a form of ethnic conflict misunderstands the phenomenon fundamentally.

In the shorter term, recent processes of global change, often glossed under the term globalization—and whether globalization is seen as a process originating in the nineteenth century, or earlier, or solely in the late twentieth century—are radically and rapidly changing the contexts under which ethnic conflict arises. In short, ethnic conflicts are no longer, if they ever were, entirely local. Rather, they are embedded in larger global, or world-systemic, processes and cannot be understood without reference to those processes.