Imagined Communities and Real Victims: Self-Determination and Ethnic Cleansing in Yugoslavia ® Robert M. Hayden American Ethnologist, Vol. 23, No. 4 (Nov., 1996), 783-801. Stable URL: http://links.jstor.org/sici?sici=0094-(M96%28199611%2923%3A4%3C783%3AICARVS%3E2.0.CO%3B2-C American Ethnologist is currently published by American Anthropological Association. Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/anthro.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is an independent not-for-profit organization dedicated to creating and preserving a digital archive of scholarly journals. For more information regarding JSTOR, please contact support@jstor.org. http://www.j stor.org/ Mon Sep 5 10:59:50 2005 imagined communities and real victims: self-determination and ethnic cleansing in Yugoslavia ROBERT M. HAYDEN—University of Pittsburgh You know, I'm a Hegelian: I know that the suffering of individuals is irrelevant to the greater processes of history. —High official, "Republika Srpska," March 1994 Of course, it would be best to resolve problems with the minorities through negotiation, but we should never rule out military force. —High official, Committee for Human Rights and Rights of Minorities, parliament of the Republic of Croatia, March 1994 We will not become a nation until being a Serb is more important than living where your ancestors lived. —Radovan Karadzic, then President of "Republika Srpska," September 13, 1995 The collapse of the former Yugoslavia has been accompanied by violence that has shocked the world, particularly because it is happening in Europe, albeit in the Balkans.1 The horror and revulsion that anthropologists share with others (e.g., Kideckel 1995) may, however, obscure the logic of the wars of the Yugoslav secessions and succession, especially the fatal incompatibility of the objectified or reified cultures (Handler 1988:14-16; Kapferer 1988:4, 22) at the base of the several nationalist enterprises with the living cultures of the areas that have been the sites of the worst violence. The geography of the violence is an important consideration because the wars in the former Yugoslavia since 1991 have taken place almost entirely within regions that were among the most "mixed"—those in which the various nations of Yugoslavia were most intermingled. The extraordinary violence that has shattered these places was not the fury of nationalist passions long repressed by communism, as many journalists and politicians would have it. I argue instead that the wars have been about the forced unmixing of peoples whose continuing coexistence was counter to the political ideologies that won the free elections of 1990. Thus extreme nationalism in the former Yugoslavia has not been only a matter of imagining allegedly "primordial" communities, but rather of making existing heterogeneous ones unimaginable. In formal terms, the point has been to implement an essentialist definition of the nation and its In this article I view "ethnic cleansing" in terms of the structural logic advanced by Mary Douglas (1966) and manifested in the constitutions of the republics of the former Yugoslavia. These constitutions reify and objectify "culture" in ways that provide the conceptual, ideological, political, and legal justifications for processes of exclusion, from the denial of citizenship to expulsion and murder. The analysis is grounded in the texts of the constitutions read against local Yugoslav understandings of their terms, in the bureaucratic practices of granting and denying citizenship on an ethnic basis, and in the geography of the wars in the former Yugoslavia. [Serbia, Croatia, Bosnia-Hercegovina, nationalism, ethnicity, law, genocide! American Ethnologist 23(4):783-801. Copyright © 1996, American Anthropological Association. ethnic cleansing in Yugoslavia 783 state in regions where the intermingled population formed living disproof of its validity: the brutal negation of social reality in order to reconstruct it. It is this reconstruction that turns the imagination of community into a process that produces real victims. This is not a Cartesian distinction or the manifestation of an analytical attachment to a symbolic-materialist framework. The fortunate members of the imagined community are as material as the unfortunates who have been excluded. Instead, I wish to pursue the power of a system of reified, prescriptive culture to disrupt the patterns of social life—culture in an analytical sense—that would contradict them. The point is certainly structuralist as per Mary Douglas (1966; see also Herzfeld 1993:22), in that "ethnic cleansing" is (to describe bloodshed in a bloodless manner) the removal of specific kinds of human matter from particular places. At the same time "ethnic cleansing" may also be a corollary to an inverted Levi-Straussian myth of nation, a myth that does not provide a logical model capable of overcoming contradictions in existing social structure but instead proclaims that the existing social structure is contrary to logic and must therefore be destroyed. As a process of homogenization "ethnic cleansing" can take many forms. Within areas in which the sovereign group is already an overwhelming majority, homogenization can be brought about by legal and bureaucratic means, such as denying citizenship to those not of the right group, thus also inducing those members of minorities who can assimilate to do so while evicting those who cannot assimilate or refuse to do so. In more mixed areas, homogenization requires more drastic measures—the physical expulsion, removal, or extermination of the minority population. Although it is only the third of these processes that has come to be known as "ethnic cleansing" since the Yugoslav wars began, it is important to recognize that legal and bureaucratic discrimination is aimed at bringing about the same result: the elimination of the minority. Conceptually, the violence of ethnic cleansing may be seen as deriving from the clash of a prescriptive model of culture (culture-as-ideology) with what exists on the ground (culture-as-lived) but is not in accordance with the prescription. Phrasing the matter in this way is not to privilege the traditional subject matter of anthropology but, rather, to accommodate the current importance in "the West" of culture-as-ideology (what Verena Stolcke has termed "cultural fundamentalism" (1995]) as the key term in a rising political rhetoric of exclusion in Western Europe. A similar distinction, accommodating another rhetoric of exclusion and domination, is Ashis Nandy's differentiation of "faith," or "religion as way of life," from "ideology," "religion as . . . identifier of populations" in South Asia (1990:70). "Religious nationalism" in India (van der Veer 1994) is comparable analytically to the "cultural nationalism" of Europe.2 In both cases, what is contrasted is the difference between prescriptive views of what "culture" or "religion" must be and the ways in which people in particular places actually live. The imperative here is not only normative—what the culture should be—but also supposedly descriptive, thereby reproducing assumptions of the way the world really is, which is why the purported cultural deviation is abnormal. At a time when many anthropologists routinely challenge any form of empiricism, a contrast between "ideology" and "the way people actually live" may seem naive. Yet surely patterns of social life—the use of one script instead of another, rates of intermarriage, or rates of the utilization of lexical items—are observable and may often not be congruent with prescriptive views of what such patterns should be. It is this lack of congruence between the present reality of life as lived and theobjectification of life as it suddenly must be lived that produces the mortal horrors of ethnic cleansing. Thus the juxtaposition of "reality" and "imagination" in my title has more than a rhetorical bite. The point of this analysis is to show the logic of the translation of category violation into mass violence, to adapt Michael Herzfeld's (1993:33) comment on Peter Loizos's work (1988) on intercommunal killing in Cyprus. Where Loizos was concerned with explaining the violence 784 american ethnologist of certain individuals, however, I wish to consider the logic of the category system on which the ethnic nation-state is based as providing the inducement for mass violence. In this regard my point is similar to that of Herzfeld (1993). While the bureaucratic activities that Herzfeld analyzes include genocide, his emphasis is, however, elsewhere. My purpose is to address directly the mass violence that has so shocked observers, myself included. Further, the phenomenon I examine is not a matter of the production of "indifference," defined as "the rejection of common humanity" or as a "denial of identity" (Herzfeld 1993:1). On the contrary, the processes that I analyze recognize people as humans (albeit, perhaps, as inferior ones) and assign consequences to identities that the subjugated groupdoes, in fact, claim. Serbs in Croatia, for example, may have claimed that identity more frequently after 1990 than they did in the preceding few decades, when many identified themselves as "Yugoslav." The meaning of the identity, however, changed. constitutions as legitimation for ethnic cleansing In this article I look at the constitutions of the successor republics to the former Yugoslavia as manifesting and institutionalizing nationalist ideologies that aim to construct homogeneous nation-states in heterogeneous territories. I am concerned with the logic of the construction of a particular kind of state, the nation-state, when the word nation has connotations that Americans view as "ethnic," not primary in current American usage of the term. When Croatia is constitutionally defined as the "national state of the [ethnic] Croat people" (Constitution of the Republic of Croatia 1990, preamble) or Slovenia as the state ofthe sovereign Slovene people, "We, the people" has a very different meaning than it does in currently dominant American imagery. Constitutions are among the most important subjects for the study ofthe implementation of nationalist ideologies precisely because they are meant to be constitutive, providing not only the conceptual framework for the state, but also the institutional means to make the state conform to that model. When the states envisioned by the constitutions exclude many residents from the bodies political and social, as in the successor states to the former Yugoslavia, the seemingly bloodless media of constitutions and laws are socially violent and may often induce bloodshed. My initial goal in this article is thus to connect the cultural construction of "nation" with the legal constitution of states in the context ofthe former Yugoslavia and its successor republics. The analysis promises to be useful for other cases of ethnic nationalism, since some ofthe constitutional and legal phenomena found in the ex-Yugoslavia cases have close parallels elsewhere, particularly in Europe. The homogenization of a heterogeneous polity may be achieved through forced assimilation or expulsion, as well as through border revision (see Macartney 1934:427^49). While compulsory assimilation may be less overtly violent than what is now called "ethnic cleansing," the two processes are based on the same principles and seem to be merely different strategies to bring about the same end. Resorting to physical violence occurs where cultural geography is most heterogeneous, thus rendering domination by nonviolent means difficult (see Hayden 1995a). In this article I consider "bureaucratic ethnic cleansing" as well as direct violence, recognizing both as consequences ofthe same logic in different social settings. long-distance fieldwork: the ethnography of ideology Analyzing constitutions as mechanisms for turning nationalist ideologies into social practice is an enterprise for which traditional models of ethnography seem inapposite. The analysis of nationalist movements must be based on an analysis of texts produced by the proponents and ethnic cleansing in Yugoslavia 785 opponents of any particular nationalist vision (see, e.g., Handler 1988:27-29; Verdery 1991:19-20), a move that may push the post-Ceertzian metaphor of culture-as-text to its ultimate extreme. Yet these texts cannot be analyzed in isolation from the field of social relations in which they have been produced, read, and interpreted in thought and action (Verdery 1991:20). Fieldwork in the societies that are the referents of specific nationalist discourses seems to be a prerequisite for such a contextual analysis of nationalist texts. Certainly the meaning of a text varies with its audience but, in the study of nationalist ideologies, the range of meanings of texts that the authors and their primary audiences have in mind is ascertainable. To ascertain these meanings, however, requires a deep knowledge of the field of social relations that can only be achieved through protracted participation and observation in the society under study. Yet the fieldwork required may be of a kind that does not fit into the traditional anthropological mode of "being there." Once the ethnographer has acquired a substantial base of knowledge of the social field in which nationalist texts are produced, it is often possible to monitor developments in this social field from a distance. Texts travel in newspapers, on the radio, and often, these days, on e-mail, so that someone in America may have an electronic version of today's newspapers from India, the former Yugoslavia, or elsewhere via the Internet. In this regard there are e-mail networks centering on Serbs and Serbia, Croats and Croatia, Bosnia, Macedonia, and Slovenia, thereby making an extraordinary range of materials instantly available to researchers and other participant-observers throughout the world. Thus a knowledgeable reader can stay current on politics and ideological constructions in the former Yugoslavia without spending much time in the former Yugoslav republics. Long-distance fieldwork of this kind is simply a corollary to the "transnational" conditions that anthropologists have noted in recent years (e.g., Appadurai 1991; Basch et al. 1994). If an Indian-born American anthropologist on a field trip to South India discovers that the temple priest he wishes to see is in Texas (Appadurai 1991:201), it requires no stretching of the concept of fieldwork to suggest that the researcher could go to Texas to question the priest. Nor is this a new situation in anthropology. After all, Lewis Henry Morgan gathered much of his kinship material for Systems of Consanguinity and Affinity (1870) by questioning "natives"—Japanese and various American Indians—who happened to be where Morgan was, in Rochester, Albany, or New York. The possibility of doing long-distance fieldwork, however, may be predicated on first having done substantial field research of the more traditional anthropological variety, involving long-term residence in the society in question and linguistic fluency. Fieldwork from afar certainly benefits enormously from short visits to the location of focal concern. This has certainly been the case in the present project. My research on the links between nationalist ideologies and their constitutional expression in what was then Yugoslavia began in 1989 after I had already spent more than three of the preceding eight years working in the country on other projects. Since then, visits of four months in 1991, a few days in 1993, one of three weeks and another of two months in 1994, and ten days in 1995 have enabled me to augment my analysis of texts with focused interviews. the multinational federation and its demise Given the meanings of "Balkanization" in English (but see Bakic-Hayden and Hayden 1992; Todorova 1994) and the widespread assumption that the various Yugoslavs have always fought each other, it is necessary to substantiate the assertion that the former Yugoslavia, if not exactly a peaceable kingdom, was a state in which ethnic or nationalist tensions did not always—or even often—dominate daily life. Accordingly, in this section and the following part of this article I explore the community of Yugoslavia by examining evidence of heterogeneity in the territories of the country and the intermingling, in all senses, of its component peoples. 786 american ethnologist The Yugoslavia that existed from 1945 until 1991, a multinational state in which no single group comprised a majority, was premised on multiculturalism. Although it was composed of republics in which all but one had a clear majority of the group for which it was named (e.g., Serbs in Serbia, Croats in Croatia), all of these republics also had sizable minority populations. The republic of Bosnia and Hercegovina, the exception, had no majority group: in 1981 its population comprised Muslims (39.5%), Serbs (32.0%), Croats (18.4%), "Yugoslavs" (7.9%), and "others and unknown" (2.2%). In the 1991 census these proportions were, respectively, 43.7%, 31.4%, 17.3%, 5.5%, and 2.1% (Petrovic 1992:4). At the other end of the spectrum, the most homogeneous republic, Slovenia, had a population that was 90.5% Slovene in 1981 and 87.6% Slovene in 1991 (Petrovic 1992:9). The political geography of the country reflected these territorial concentrations. The Socialist Federal Republic of Yugoslavia (1945-91/92) was a federation of six republics (Bosnia-Herce-govina, Croatia, Macedonia, Montenegro, Serbia, and Slovenia) and two "autonomous provinces" within the republic of Serbia (Vojvodina and Kosovo). With the exception of Bosnia-Hercegovina, each republic or autonomous province was the area of the greatest territorial concentration of one of the major national groups that comprised Yugoslavia. Thus in 1991, 99.3 percent of the Slovenes in Yugoslavia lived in Slovenia, while 70.6 percent of the Montenegrins lived in Montenegro. In the free elections held in 1990 after the collapse of the League of Communists, the winning message in each republic was one of classic nationalism: Serbia for Serbs, Croatia for Croats, Slovenia for Slovenes, and Macedonia for Macedonians. In Bosnia-Hercegovina the vote resembled an ethnic census, with Muslim, Serb, and Croat nationalist parties accounting for about 80 percent of the total, in proportions only slightly less than those of each national group in the population of the republic; the most important party standing for a civil state of equal citizens, the Alliance of Reform Forces of Yugoslavia of the federal prime minister, received only 5.6 percent of the vote—less than the 6.0 percent received by the "reformed" communists (see Hayden 1993a). The victorious politicians in Serbia, Slovenia, and Croatia worked independently, and for their own reasons, to disable the federal government, thereby attaining the de facto state sovereignty mentioned earlier (Woodward 1 995; see also Jovic 1 995). Thus each republic except Bosnia-Hercegovina became a true nation-state based on the sovereignty of the majority national group. The separate nationalist political movements were justified on the grounds of "self-determination." This famous concept, however, had a specific meaning in Yugoslav politics and popular culture, one that had grim implications for any concept of a civil state of equal citizens. A statement in the first line of the 1974 Yugoslav Constitution about "the right of every nation to self-determination, including the right to secession"3 referred, not to the populations or citizens of republics, but to the nations, narodi (singular: narod), of Yugoslavia, ethnically defined. While these "nations" were recognized as having their several republics, it was the "nations," not the republics, that were described as having united to form the Yugoslav state; the Yugoslav republics, unlike those of the Soviet Union, did not have a right to secede. This seemingly arcane distinction between "nation" and "republic" as the bearer of rights was actually of vital political importance. The key to the separate nationalist political movements in Yugoslavia after 1989 was the explicit conflation of the "nation," ethnically defined, and the "state." Although this formulation was hardly new to European history, it did have sinister implications for minorities in states that were suddenly defined as the nation-states of their respective ethnic majorities. By definition, anyone not of the majority ethno-nation could only be a citizen of second class. The key to this distinction lay in the concept of sovereignty. As nationalist politicians came to power in the various Yugoslav republics after the elections of 1990, they rewrote their respective republican constitutions to justify the state on the sovereignty ethnic cleansing in Yugoslavia 787 of the ethnically defined nation (narod) in which others might be citizens but could not expect an equal right to participate in control of the state. The politics of nationalism in Yugoslavia in the late 1980s and early 1990s thus turned territories populated by concentrations of the various national groups into states in which the members of the majority nation were sovereign (see Denich 1994; Hayden 1992a). The presumption of the politics was that the various Yugoslav peoples could not live together and that therefore their common state had to be divided. The electoral success of this message meant the defeat of the "Yugoslav idea" of a common state of the south Slavic peoples, an ideology that had been devised as a counter and rival to the separate national ideologies of each group (see A. Djilas 1991). To reverse Benedict Anderson's evocative phrase (1983), the disintegration of Yugoslavia into its warring components in 1991-92 marked the failure of the imagination of a Yugoslav community. This failure of the imagination, however, had real and tragic consequences: the Yugoslav community that could not be maintained, and thus has become unimaginable, had actually existed in many parts of the country. Indeed, it is my argument that the spatial patterning of the war and its terrible ferocity are due to the fact that in some regions the various Yugoslav peoples were not only coexisting but also becoming increasingly intermingled. In a political situation premised on the incompatibility of its components, these mixed territories were both anomalous and threatening since they served as living disproof of the nationalist ideologies. For this reason, the mixed regions could not be permitted to survive as such, and their populations, which were mixing voluntarily, had to be separated militarily. heterogeneity, mixed marriages, and "Yugoslavs" Despite the maintenance of high levels of territorial concentration of the various national groups in their respective republics, the levels of ethnonational heterogeneity throughout most of Yugoslavia were increasing. In Slovenia, for example, the concentration of the Slovene population increased from 97.7 percent of Slovenes residing in Slovenia in 1981 to 99.3 percent in 1991 (Petrovic 1992:1 5). During this same decade, however, the homogeneity of Slovenia decreased: 90.5 percent of the population were Slovenes in 1981, compared with 87.6 percent in 1991 (Petrovic 1992:9). Nor was Slovenia unusual in this regard. From 1953 to 1981 almost all of the territories of Yugoslavia became increasingly heterogeneous (Petrovic 1987:48); that is, in almost all republics and provinces, the percentage of the population that was made up by the majority national group declined. The exceptions were the two autonomous provinces in Serbia, Vojvodina and Kosovo. In Vojvodina the Serbian majority increased, in part because of the low birthrate among the next largest group, the Hungarians. In Kosovo the Albanian majority increased, in part because of the high Albanian birthrate and the massive Serbian emigration from the province.4 Between 1981 and 1991, heterogeneity increased in Montenegro, Macedonia, Slovenia, and Serbia, but decreased in Croatia5 and Bosnia-Hercegovina (Petrovic 1992). Accompanying the increasing heterogeneity of most of the republics was an increase in the rates of intermarriage between members of the different national groups. Intermarriage is usually thought to indicate increasing assimilation and to increase integration of social groups (e.g., Blau et al. 1982). From the early 1950s through the 1980s, "mixed" marriages increased both in absolute numbers and in proportion to all marriages throughout most of Yugoslavia {Vreme 1991), but were particularly common between Serbs and Croats, and between Serbs and Muslims in Bosnia-Hercegovina. Not surprisingly, the highest rates of intermarriage occurred in the places in which the populations were the most intermingled: the large cities, the province of Vojvodina, Bosnia-Hercegovina, and the parts of Croatia that had large numbers of Serbs and Croats.6 If we consider the frequency of the claim that Serbs and Croats suffer from age-old hatreds, it is worth scrutinizing their increasingly close coexistence in Croatia after 1945 despite the 788 american ethnologist terrible massacres of Serbs by the fascist "Independent State of Croatia" from 1941 to 1945.7 According to the 1991 census, 12.2 percent of the population of Croatia were Serbs, primarily residing in Zagreb, but otherwise concentrated in several parts of the republic—specifically Slavonija, Banija, Kordun, and Lika. In Lika the population was almost entirely Serb, and there were few intermarriages. In areas where Serbs and Croats lived together, however, they intermarried in large numbers. For example, in the town of Petrinja in Banija, where the population was almost equally divided between Serbs and Croats, about 25 percent of the marriages were mixed, while in the major towns of Slavonija the percentages of mixed marriages climbed to 35 percent in the town of Pakrac (Borba 1991). Mixed marriages produce children of mixed background. Already by 1981, approximately one-third of the children born in Slavonijan towns such as Osijek were of mixed Serb-Croat background {Borba 1991). Bosnia-Hercegovina had the highest percentage of "mixed" children—15.9 percent overall—also concentrated in the most mixed areas. Even Slovenia, the republic with the highest concentration of its national minority group, had large numbers of "mixed" or "foreign" births: 7.9 percent issuing from mixed marriages, with another 19.0 percent from non-Slovene marriages, leaving only 73.1 percent of children issuing from "purely Slovene" marriages (Borba 1991). Another indicator of heterogeneity can be found in the figures on those who identified themselves in the censuses as "Yugoslavs" instead of as Serbs, Croats, Muslims, or any other national group. Between the 1971 and 1981 censuses the numbers of "Yugoslavs" increased sharply from 1.3 to 5.4 percent of the total population (Burg and Berbaum 1989). The distribution of these ethnic "Yugoslavs" was nevertheless far from consistent throughout Yugoslavia. In 1981 they lived primarily in Belgrade and the Vojvodina in Serbia, in the major industrial centers in Bosnia-Hercegovina, Istria, and some larger centers in Croatia, as well as in the "mixed" regions of Croatia (Petrovic 1987:152-153; Danas 1991). The age distribution of these Yugoslavs in 1981 indicated that this identity was preferred by younger people, which led some researchers to conclude tentatively (and subject to the rise of precisely the type of nationalist politics that destroyed Yugoslavia in the late 1980s) that Yugoslavia was developing an increasing sense of community and that support for the multinational community was likely to increase, as would self-identification as Yugoslavs (Burg and Berbaum 1989). Although these statistics do not indicate that national identity vanished, it is clear that national identity was not a primary focus of most people's concerns in the early 1980s. Ethnographers from mixed regions have consistently reported that while national differences were recognized, tensions were low in the 1980s until political events from outside of these regions overtook them (Bringa 1993 [Bosnia]; Jambresic 1993 [Banija]; Olsen 1993 [Slavonija]).8 The rise of mutually hostile nationalisms led to a sharp decline in the percentage of Yugoslavs throughout the country, from 5.4% in 1981 to 3.0% in 1991—a 41.3% drop. Again, the rates of decline by republic were not even. The percentage of Yugoslavs dropped most dramatically in Croatia, from 8.2 to 2.2% (a 72.3% drop) but also declined everywhere else: Bosnia-Hercegovina by 26.5%, Serbia by 28.1%, and Slovenia by 53.4% (Petrovic 1992). The percentages of Yugoslavs remained highest, however, in the most mixed regions: Bosnia-Hercegovina (5.5%) and the mixed areas of Croatia—where Yugoslavs had been most numerous in 1981 (Danas 1991). It should be noted that the decline in the number of self-identified "Yugoslavs" may often have represented a calculated assessment that continuing to identify oneself as such for official purposes was becoming increasingly hazardous. At the time of the census (April 1991) I was told by a number of people that they would prefer to continue to identify themselves as Yugoslavs but were afraid that doing so could cost them their jobs, and perhaps even their property, in the chauvinist political climate then dominant.9 ethnic cleansing in Yugoslavia 789 Through the early 1980s, then, most parts of Yugoslavia showed an increasing heterogeniza-tion of populations, accompanied by increasing numbers of mixed marriages and births of children of mixed parentage, and a rise in the percentage of those who identified themselves as "Yugoslav" rather than as members of any of the ethnonational categories. But the distribution of these factors was not random. Instead, heterogeneity was concentrated in the central part of the territory of Yugoslavia: the republic of Bosnia-Hercegovina, the parts of Croatia bordering Bosnia-Hercegovina and Vojvodina, and Vojvodina itself. In these parts of Yugoslavia, the idea that the Yugoslav peoples could not live peacefully together was empirical nonsense. It was perhaps because these regions constituted living disproof of the nationalist ideologies that became politically dominant after the late 1980s that, except for Vojvodina, those territories in which the intermingling of the populations was most complete have been the major theaters of the war. This is not to attribute causation to the structural argument; empirical priority must be given to the fact that the nationalists claimed the same "mixed" territories and were willing to fight over them. But the structural argument provides the basis for understanding how the empirical facts of "cleansing" become acceptable, even seemingly desirable.10 constitutionalizing nationalism Contrary to the official rhetoric both of winners and of most western observers, the free elections of 1990 in Yugoslavia did not replace state socialism with democracy. Instead, the transition was from regimes dedicated to advancing the interests of that part of the population defined constitutionally as "the working class and all working people" (see Hayden 1992a) to regimes dedicated to advancing the interests of that part of the population defined as the ethnonational majority. In this sense, the transition was from state socialism to state chauvinism, and socialism's "class enemy" was replaced by the national enemy identified by the particular local chauvinism (Hayden 1992a). Not surprisingly, these national enemies were primarily the members of the largest minority in each polity, along with any members of the majority who might try to support rights for the minority. Once in power, the victorious nationalists in each republic began to enact systems of constitutional nationalism, meaning constitutional and legal systems devised to ensure the dominance of the majority ethnonational group (see Hayden 1992a). Thus, for example, the constitution of Croatia (1990)11 gives in its preamble a capsule history of the efforts of the Croat "nation" (narod) to establish "full state sovereignty." After mentioning the "inalienable . . . right of the Croat nation to self-determination and state sovereignty," the Republic of Croatia is "established as the national state of the Croat nation and the state of the members of other nations and minorities that live within it" (Constitution of the Republic of Croatia 1990, preamble). In all these passages, "Croat nation" (Hrvatski narod) has an ethnic connotation and excludes those not ethnically Croat. This exclusionary definition of the bearer of sovereignty is reinforced by the emblems of the state—a flag and coat-of-arms bearing designs associated only with Croats (art. 11)—and the specification that the official language and script of Croatia are "the Croatian language and Latin script" (art. 12), thereby excluding the Serbian dialects and the Cyrillic alphabet customarily used to write them.12 Similar formulations of constitutional nationalism have arisen in other republics (Hayden 1992a:658-663). The transition from state socialism to state chauvinism is seen in the formulations of state identity and purpose contained in the various republican constitutions. Whereas the socialist constitutions grounded the state in the dual sovereignty of "the working class and all working people" and "the nations and nationalities" of Yugoslavia, the collapse of socialism left only one sovereign (Samardzic 1990:31). Furthermore, the formation of a state for each of these sovereign "nations" was justified by the right of self-determination. This is seen in the preambles or prefatory parts to the various constitutions (emphasis added in each case): 790 american ethnologist Proceeding from . .. the inalienable and inextinguishable right to self-determination and state sovereignty ofthe Croatian nation, the Republic of Croatia is established as the national state of the Croatian nation and the state of members of other nations and minorities who are its citizens. [Constitution ofthe Republic of Croatia 1990, preamble] Resting upon the historical, cultural, spiritual and statehood heritage of the Macedonian nation and upon their centuries' long struggle for national and social freedom, as well as for the creation of their own state .. . Macedonia is established as the national state ofthe Macedonian nation. [Constitution of the Republic of Macedonia 1991, preamble] On the basis of the historical right ofthe Montenegrin nation to its own state, established in centuries of struggle for freedom .. . the Parliament of Montenegro .. . enacts and proclaims the Constitution of the Republic of Montenegro. [Constitution of the Republic of Macedonia 1991, preamble] Proceeding from the centuries-long struggle ofthe Serbian nation for independence .. . determined to establish a democratic state ofthe Serbian nation . . . the citizens of Serbia enact the Constitution of the republic of Serbia. [Constitution of the Republic of Montenegro 1992, preamble] Proceeding from ... the basic and lasting right ofthe Slovene nation to self-determination and from the historical fact that Slovenes have, over centuries of struggle for national liberation formed their national identity and established their own statehood, the Parliament of the Republic of Slovenia enacts the Constitution ofthe Republic of Slovenia. [Constitution ofthe Republic of Serbia 1990, preamble] Although not internationally recognized, the "Republic of Serbian Krajina," the self-proclaimed Serbian state in Croatia, defined itself in its constitution in much the same terms as the recognized successor states above:'3 Proceeding from the right ofthe Serbian nation to self-determination . . . and the centuries-long struggle for freedom . . . , determined to establish a democratic state of the Serbian nation on its own historical and ethnic space, in which the other citizens are guaranteed the realization of their national rights, a state based on the sovereignty belonging to the Serbian nation and other citizens in it. . . the Serbian nation of the Republic of Serbian Krajina . . . enacts the Constitution of the republic of Serbian Krajina. [Constitution ofthe Republic of Serbian Krajina 1991, preamble] Similarly, the "Republika Srpska," the Serbian "entity" in Bosnia-Hercegovina proclaimed: Proceeding from the inalienable and untransferable natural right ofthe Serbian nation to self-determination, self-organization and association, on the bases of which it freely establishes its own political status and secures its economic, social and cultural development.... To proclaim [the Serbian nation's] determination to decide independently its own fate and to proclaim its firm will to establish its own sovereign and democratic state.. . . The Parliament of the Serbian nation in Bosnia and Hercegovina enacts the Constitution of the Republika Srpska. [Constitution of the Republika Srpska 1992, preamble] In each of these preambles, the word nation {narod in all ofthe languages involved) has an ethnic connotation; narod has the same root (rod) as the verb roditi (to give birth, to be born). When preceded by the ethnic adjective (Croatian, Macedonian, Montenegrin, Serbian, Slovenian), the constructions exclude those not of the specified ethnicity. From the excerpts above, and particularly the phrases emphasized, it is clear that the various formerly Yugoslav republics are considered to be manifestations ofthe right to self-determination—meaning the right to form one's own state—ofthe majority, titular nation (narod), even when some expression is given to the equality of minorities. Again, a contrast can be made with the Preamble ofthe U.S. Constitution, which provides simply that "We the People ofthe United States ... do ordain and establish this Constitution."14 Bosnia-Hercegovina, like the former Yugoslav federation itself, represents the failure of an attempt to define the state in such a way as to recognize the sovereignty of all of its constituent groups without privileging any of them. The last socialist constitution of Bosnia-Hercegovina (1974) defined the republic as a socialist democratic state and a socialist self-management democratic community of working people and citizens, the nations [narodi] of Bosnia and Hercegovina—Muslims, Serbs, Croats, and members of other nations and nationalities living within it, based on the rule and self-management of the working class and all working people and on the sovereignty and equality of the nations of Bosnia and Herzegovina and the members ofthe other nations and nationalities that live within it. [art. 1 ] ethnic cleansing in Yugoslavia 791 As socialism collapsed this definition was replaced by a constitutional amendment, so that the definition of the state in Article 1 read: The Socialist Republic of Bosnia and Herzegovina is a democratic sovereign state of equal citizens, of the nations of Bosnia and Herzegovina—Muslims, Serbs and Croats, and members of other nations and nationalities living within it.15 Yet this definition did not satisfy the aspirations of Serbian and Croatian political figures in Bosnia-Hercegovina. In part because of problems of defining the state, no new constitution for Bosnia-Hercegovina was ever passed, and, as Yugoslavia collapsed, the Serb and Croat leaders in Bosnia proclaimed their own self-determining regions within the republic. These regions quickly became quasi-states, closely linked to Serbia and Croatia, respectively, and were independent of the supposedly sovereign government of Bosnia-Hercegovina in Sarajevo (see Shoup 1994). The war that followed effected the partition of Bosnia-Hercegovina into regions that were meant to be, and are fast becoming, ethnically "pure" (see Hayden 1993a). This partition was inevitable once Yugoslavia collapsed because the self-determination of the Yugoslav nations (narodi), the political program that succeeded in 1990, meant that the Serbs and Croats of Bosnia-Hercegovina would be drawn inevitably toward union with their ethnic confreres.16 Thus "self-determination" brought on the civil war that destroyed Bosnia-Hercegovina. The Constitution of the Federation of Bosnia and Hercegovina that was written with the help of American diplomats and signed in Washington, D.C., by Croats and Muslims in March 1994, is based on a constitutional nationalism that excludes Serbs from the sovereign peoples of Bosnia-Hercegovina. While the preamble states, "The peoples and citizens of Bosnia and Herzegovina, determined to establish full national equality, democratic relations, and the highest standard of human rights and freedoms, hereby create a Federation," Article 1 then asserts that Bosniacs and Croats, as constituent peoples (along with others) and citizens of the republic of Bosnia and Hercegovina, in the exercise of their sovereign rights, transform the internal structure of the territories with a majority of Bosniac and Croat population in the Republic of Bosnia and Hercegovina into a Federation.17 The term Bosniac, an Anglicization of Bošnjak, has a purely Muslim referent and is not the same as Bosnian (Bosanac), and is a term for ethnic Muslims that avoids the specifically religious implications of Muslimani. In any event, this constitution excludes Serbs from the structure of the federation, apportioning executive offices to Muslims/Bosniacs and Croats (IV.B.1. arts. 2-5) and ensuring veto power in the legislature to Muslim/Bosniac and Croat delegations but not to others (IV.A.4 art. 18). The exclusion of Serbs became apparent immediately after the constitutional draft was signed in Washington, D.C., when a Sarajevo conference of Serbs who were loyal to the idea of a multiethnic Bosnian state asked to be included in negotiations. They were ignored (New York Times 1 994). The Dayton-Paris peace treaty that in December 1995 brought the Bosnian war to at least a temporary halt manifests constitutional nationalism in much the same way as the Federation constitution of 1994 by giving Muslims and Croats superior rights over all others in their portion of Bosnia, and the same to Serbs in theirs (see Hayden 1995b). citizenship: denaturalization as bureaucratic ethnic cleansing In popular speech and in the rhetoric of many international documents, the world is composed of nations. At the levels of law and politics, however, it is composed of states. Citizens of a state almost always possess rights that noncitizens do not share, and this was certainly true in the republics of the former Yugoslavia. As these states achieved independence their governments began to write rules to determine who could stay and who could not, who could 792 american ethnologist work and who could not, who could vote and who could not, who would receive medical insurance or other benefits and who would not, and who would be permitted to own real property and who would not. In each case, citizens were entitled to the rights and benefits; noncitizens were—at best—only temporarily entitled to them. Thus the question of citizenship in the successor states to the former Yugoslavia is one of utmost importance to the people living in them, since those who do not attain citizenship will be denied the rights essential for any kind of normal life. It must be emphasized that for many the question of citizenship was new. As noted earlier, the constitution of Yugoslavia had provided for a single, uniform Yugoslav citizenship and guaranteed the equality of Yugoslav citizens throughout the country. Suddenly, however, the citizenship of many residents in the newly independent states became questionable. New citizenship laws, written to privilege the members of the sovereign majority in each case, have worked to discriminate against residents who were not members of the majority groups. In essence, the new citizenship regimes have simultaneously extended citizenship to nonresident members of the majority ethnonation through easy naturalization while denying citizenship to many residents who are not of the right group. This last process turns residents who had been equal citizens of federal Yugoslavia into foreigners of their own republics, a process we might call denaturalization. Neither of these phenomena is unique to the formerly Yugoslav republics. The easy extension of citizenship to nonresident ethnonational-religious confreres is well known (as in Ireland and Israel), while the denial of citizenship to large numbers of people who until then were thought to have held it was the purpose of the 1981 British Nationality Act (Gilroy 1987). In this last case, however, many of the "denaturalized" potential citizens were not resident in Britain at that time. The combination of the easy naturalization of nonresidents with the denaturalization of residents seems uncommon, but is manifested now in the successor states of the former Yugoslavia and the former Soviet Union (see Brubaker 1992, 1993). The power of an imagined ethnic community (Anderson 1983) to break up actually existing communities in these post-communist settings is clear. With the demise of Yugoslavia, the immediate practical question for many citizens of those erstwhile states was citizenship in one of the successor states. Here, laws and policies have varied. At the most inclusive end, the Slovenian Citizenship Act of 1991 offered citizenship to all citizens of another Yugoslav republic who had resided in Slovenia on the day that the plebiscite on independence was held, and most applicants have been granted citizenship (Mazowiecki 1993a:44). Even so, approximately 50,000 citizens of Yugoslavia who were counted in the 1991 census as residing in Slovenia have become foreigners there since the independence of that republic (Vreme 1993:33). Other states have been far less accommodating. Unlike the Slovenian law, the law on Croatian Citizenship of 1991 made no special provision for citizens of other Yugoslav republics but instead rendered them all "foreigners" who were required seek to naturalization. Furthermore, Serbs in Croatia have complained that their requests for citizenship or for naturalization have been denied (see Mazowiecki 1992:22, 1993a:26-28). Although the Croatian authorities have denied discriminating against Serbs, relatively large numbers of requests for citizenship have been rejected (Vreme 1993:34). The Law on Croatian Citizenship permits the authorities there to reject a citizenship application even though the applicant has met all the criteria if they "are of the opinion that there are reasons in the interest of the Republic of Croatia for refusing the request for the acquisition ... of citizenship" (art. 27, sec. 2). The same article provides that these authorities need not state their reasons for rejecting an application (art. 27, sec. 3). Thus, as the Serbs have complained, the opportunity for discrimination certainly exists.18 The laws governing citizenship and naturalization are interesting because they are the mechanisms through which the imagination of an ethnonational community is made manifest ethnic cleansing in Yugoslavia 793 and actualized. Specifically, these laws provide the grounds for the acquisition of membership in the community, thus revealing the principles thought to define it. Again, the Law on Croatian Citizenship (1991) is interesting.19 Article 8 of this law stipulates the following: A foreign citizen who files a petition for acquiring Croatian citizenship may acquire Croatian citizenship by naturalization if he meets the following requirements: (1) [age requirement: 18] (2) [omitted] (3) that before the filing of the petition, he had a registered place of residence for a period of not less than five years uninterrupted on the territory of the Republic of Croatia. (4) that he is proficient in the Croatian language and Latin script. (5) that a conclusion can be drawn from his conduct that he adheres to the laws and customs prevailing in the Republic of Croatia and that he accepts Croatian culture. At first glance Sections 3 and 4 of this article do not seem to be overly controversial, but both open wide opportunities for discriminatory application. The residency requirement depends on the interpretation of the qualification "uninterrupted" (neprekidno). More interesting is the language qualification. The dialects of what has until now been known as Serbo-Croatian or Croato-Serbian are myriad and intermixed, with some Serbian populations speaking dialects similar to those spoken by most Croats, and some Croat populations speaking dialects similar to those spoken by most Serbs (see Hammel 1993:7-8). Serbs preferto use the Cyrillic alphabet, while Croats almost never use it. Thus the language criterion is problematic: is someone who speaks the Belgrade dialect proficient in the "Croatian language"? Who decides, and on what grounds? Would a "Serbian" dialect qualify if the speaker is an ethnic Croat but not otherwise? Section 5, however, is most revealing. What, exactly, does it mean to "accept Croatian culture," and how does one conduct oneself to show such acceptance? Since the primary distinguishing feature of Croatian culture is Roman Catholicism, must one convert to that faith? If not, what does accepting Croatian culture entail? This provision of the law takes a concept that anthropologists have regarded as descriptive and analytical and makes it prescriptive; yet the concept remains empty of specific content. It is this prescription of culture that turns it into an object (Kapferer 1988:2; cf. Handler 1988:14). The essentialism involved verges on racism when it entails viewing reified culture as somehow surviving transplantation into another country where the chosen people are a minority.20 In the Croatian case, these implications become clear in the special rules for emigrants and their descendants (art. 11) and for members of the Croatian nation (narod) who do not reside in Croatia (art. 16). In regard to both categories, Croatian citizenship can be acquired even though the applicant does not meet the requirements stated in Article 8, sections 1 -4, but these candidates must still meet the requirement of section. To an anthropologist the complete separation between language and culture inevitably seems odd; yet it is restated twice and so seems not to have been a slip of the drafter's pen. This provision provides a tool for extending citizenship only to ethnic Croats (e.g., the child of Croat emigres from Croatia or of ethnic Croats from Serbia) while denying it to others similarly situated (e.g., the child of Serb emigres from Croatia). Taken together, the naturalization provisions of the Law on Croatian Citizenship may lead to situations in which, for example, a Muslim from Bosnia, long resident in Croatia and a native speaker of a Croatian dialect of what used to be called Serbo-Croatian, is denied citizenship, while an ethnic Croat from the United States, who has never been to Croatia and who doesn't know the language, is granted Croatian citizenship. While the actual number of such cases in Croatia is unknown, it is interesting to note that the Slovenian provisions in regard to naturalization also privilege ethnic Slovenes, and that while 50,000 residents of Slovenia who were citizens of the former Yugoslavia have not acquired Slovenian citizenship, 25,000 ethnic Slovenes from outside of Slovenia have done so (Vreme 1993:34).21 Again the power of the imagined ethnic community to break up communities on the ground is apparent. 794 american ethnologist The new citizenship laws provide the legal means to exclude individuals from citizenship on ethnic grounds—in essence, bureaucratic ethnic cleansing. self-determination, homogenization, and "ethnic cleansing" The logic of "national self-determination" in Yugoslavia not only legitimates homogenization of the population but has also made that process so logical as to be irresistible. The course of the war has followed this logic of establishing the nation-state by eliminating minorities. What can be done bureaucratically by a majoritarian regime in a state with a numerically overwhelming majority, however, must be accomplished in other ways if the majority is not secure in its rule—specifically, military conquest and the subsequent expulsion of the unwanted population. The Serbs, who initially took by far the greatest amount of territory, have also committed by far the largest number of human rights violations. Nevertheless, the 1993 Croatian offensives to establish an ethnically pure Herceg-Bosna followed the same course in central Bosnia (Mazowiecki 1993a:8-10; 1994:6) and Mostar (Mazowiecki 1993a). "Population exchanges" have been part of this effort (Mazowiecki 1994:9-10). The result of the war as of late 1994 was the more or less complete exchange of populations outside of Sarajevo, as shown in Table 1. During the spring and summer of 1995, this process of expelling populations increased on the part of all parties. In May a Croatian offensive against the Serb enclave of Western Slavonia led to the expulsion of virtually all Serbs from that part of Croatia. In July Serb forces took two of the Muslim "safe areas" in eastern Bosnia and expelled or killed all residents. In August a Croatian offensive in the Krajina expelled close to 200,000 Serbs from Croatia, the single largest incidence of ethnic cleansing in the wars. Thus, between June 1991 and August 1995, more than 85 percent of the Serb population of Croatia was forced to leave the country (Vreme 1995). As in Croatia, the summer of 1995 brought even greater waves of "ethnic cleansing" by the various forces. In July—as I have just mentioned—the Bosnian Serbs captured two Muslim "safe areas" in eastern Bosnia and expelled or killed the inhabitants. In September the Muslims, with the support of the Croatian army, began an offensive in western Bosnia that drove tens of thousands of Serbs out of west-central Bosnia, just north of a line running from Jajce to Bihac. Before the war began much of this region had been populated almost exclusively by Serbs. The effects of all these military actions on population distribution are summarized in Table 2. Despite much-publicized protestations that the international community would never accept the ethnic partition of Bosnia, the Dayton-Paris peace treaty does precisely this by recognizing that Bosnia is composed of two "entities": the Croat-Muslim "federation" and "Republika Srpska," each under its own constitution (Hayden 1995b). Since, as I have shown above, these constitutions define their respective states in ethnic terms, this agreement under international Table 1. United Nations High Commission for Refugees [UNHCR] population estimates. 1991 Census November 1994 estimate Croat-Muslim federation Serbs 205,185 36,000 Muslims and Croats1 1,209,804 1,673,000 Serb-held territories Serbs 928,857 1,169,000 Muslims and Croats 838,190 73,000 Eastern enclaves Serbs 20,000 none Muslims 80,000 11 5,000 Source: Balkan War Report 1995:5 ethnic cleansing in Yugoslavia 795 Table 2. Population estimates for the areas of control in Bosnia and Herzegovina. Region3 Muslims Serbs Croats Total population Bihac region 200,000 5,000 6,000 211,000 Northern Bosnia 7,000- 660,000- 3,000 660,000-730,000 (Banja Luka region) 10,000 730,000 Central Bosnia 850,000- 20,000 130,000 850,000-1,000,000 1,000,000 Government-held 230,000 30,000 20,000 280,000 areas of Sarajevo Eastern Bosnia 5,000b 530,000- 535,000-560,000 (Serb-held) 560,000 Croat-held areas 100,000 5,000 550,000 655,000 (Western Hercegovina and Central Bosnia) aThe information in this table originated from a map issued by the U.S. Government's Bosnian Task Force. Source: U.S. Government 1995 bMuslims and Croats sponsorship legitimates the ethnic partition of Bosnia. But that partition had already been accomplished on the ground. The multiethnic Bosnia that was once actual, and for that reason prescriptive from the point of view of the international community, no longer exists and thus can no longer be prescriptive. from optimism of the intellect to pessimism of the will The analysis of ethnic cleansing as a manifestation of the incompatibility of the objectified or reified cultures at the base of the several nationalist enterprises with the living cultures of the areas that have been the sites of the worst violence is at once intellectually reassuring and deeply disturbing. It is encouraging intellectually to know that anthropological frameworks of analysis can explain the violence that has destroyed what had been the ethnically mixed regions of the former Yugoslavia. A rationalist might propose that since we know so much about the phenomena involved, perhaps we can prevent their recurrence in another place and time. Yet another stream of rational thought induces pessimism. The circumstance that induces ethnic cleansing is one of category violation. While it may be that contradictions are not resolved in myth and dream, in the realm of cultural politics the drive to make the world conform to a vision of the way it supposedly should be is powerful. That the vision is flawed empirically is irrelevant. Indeed, once the vision receives general support, its empirical falsity simply adds ferocity to the drive to accomplish it. A comparative look also gives further pause for thought. What we now call ethnic cleansing has been seen quite often in the 20th century, above all in Europe, but not only there. A look at some examples shows that the process has often succeeded in creating a new reality. For example, Poland expelled 6 million Germans in 1945, while 3 million Jews were eliminated from Poland in the period 1939-46, most being killed or deported in the Holocaust. The result has been the creation of one of the most ethnically pure states in Europe, a condition that is generally seen as to Poland's advantage in attaining postsocialist "democracy." Similarly, the expulsion of more than 3 million Germans from Czechoslovakia in 1945 has rendered the now independent Czech republic ethnically pure and thus, like Poland, ready for democracy. Hungary, the other leading candidate for the European Union and NATO, became ethnically pure after World War I through its exclusion of those territories where Hungarians and others 796 american ethnologist lived together. Thus Slovakia, Romania, and Serbia have internal ethnic tensions with Hungarians, but Hungary has none in its own territory. In the Yugoslav wars, Croatia's expulsion of its Serbs was viewed by the American Ambassador to Croatia as a positive step in resolving the Yugoslav conflicts (OMRI Daily Report 1995). "Ethnic cleansing" in Europe is thus a phenomenon that has proven successful both in recreating social reality and in gaining political acceptance. Faced with this historical experience and with that of the Yugoslav wars, perhaps I may be excused if I adapt and reverse Gramsci's famous dictum. We can now, as anthropologists, understand very well the processes that lead to ethnic cleansing; but we can also see how unlikely it is that, once started, they can be stopped. Optimism of the intellect here leads to pessimism of the will. notes Acknowledgments. The research reported here was supported in part by the National Council for Soviet and East European Research, but the opinions expressed are exclusively those of the author and cannot be attributed to the council. Earlier versions of this article were presented at the annual meetings of the American Association for the Advancement of Slavic Studies (Phoenix, AZ, November 1992), the American Anthropological Association (San Francisco, CA, December 1992), and the Law and Society Association (Chicago, IL, May 1993). The article has benefited very much from the comments of several anonymous reviewers for the American Ethnologist, Don Brenneis, and Michael Herzfeld. 1. Attempts to distinguish "the Balkans" from "Europe" have been central to much of the political discourse over the legitimacy or necessity of political acts concerning Yugoslavia's collapse and subsequent wars, both by Yugoslav politicians and by those on the world stage who have had to deal with them (see Bakic-Hayden 1995; Bakic-Hayden and Hayden 1992; Todorova 1994). Considering the extent of the devastation that Europeans have wrought on each other, to say nothing of the rest of the world, in what Günther Grass has called "the century of expulsions" (1992:27) such rhetorical exercises are suspect and are rejected here. 2. While I cannot explore the matter in this article, I suggest that what Gunnar Myrdal (1944) identified as the basis of the "American dilemma"—racism—is paralleled in different idioms elsewhere, such as a "European dilemma" of nationalism, or a South Asian one of "communalism." Note that in all cases, the dilemma is a moral one, caused by the persistent existence of supposedly "natural" distinctions in polities that profess aspirations toward democracy. 3. Constitution of the Socialist Federal Republic of Yugoslavia 1974: "Introductory Part, Basic Principles." 4. Bosnia-Hercegovina showed a rather different trend: the Serbian plurality recorded there in 1961 became a Muslim plurality in 1971, after the recognition of "Muslim" as a nationality in 1967 and the subsequent change in the declaration of nationality by many who had called themselves Serbs in 1961 (see Petrovic 1987:47). 5. The increase in the percentage of Croats in Croatia in the 1991 census was apparently the result of a shift by many who had identified themselves as "Yugoslav" in 1981, to "Croat." The number of "Yugoslavs" in Croatia declined by 72 percent between these two censuses: from 8.2 percent of the population in 1981 to 2.2 percent in 1991 (Petrovic 1992:7). 6. I am not convinced by the recent argument by Botev and Wagner (1 993) that intermarriage did not increase in Yugoslavia, an argument that considers aggregate data on the level of the republic and thus is not sensitive to regional variations. Furthermore, the symbolic value of what they view as small numbers of intermarriages was great. Contrary to their reasoning, Ivan Siber of the University of Zagreb has documented a sharp decline in intermarriages in Croatia since 1991 and interprets this as a sign of the homogenization of the population (Feral Tribune 1994). 7. The extent of these massacres became a topic of hot debate in the late 1980s, with Croatian historians attempting to minimize the numbers (see Boban 1990; for a discussion of this see Boban 1991; Hayden 1992b, 1993b, 1994). Croatian sensitivity on this topic can be seen in a ferocious attack—far in excess of normal standards of propriety in American scholarship—on Hayden's comments on Boban by a second Croat writer (Knezevic 1993; reply in Hayden 1993b). A recent, extremely careful analysis of the data on World War II victims in Yugoslavia (Bogosavljevic 1995) gives figures far lower than most Serbs imagine but far higher than most Croats will admit. 8. The transformation of the people in a mixed Muslim-Croat village from neighbors of different faiths into enemies of different nationalities is seen in Tone Bringa's stunning ethnographic film, Bosnia: We Are All Neighbors, broadcast in America on PBS in May 1994 (Bringa 1994). 9. Some respondents to the census registered a protest against the whole process by listing themselves as Eskimos, Bantus, American Indians, Citroens, lightbulbs, and refrigerators—among other fanciful categories. The deadly nature of the categories was brought home to participants at a seminar on "Beyond Genocide" at John Jay College in New York in April 1993, when a human rights group from the town of ethnic cleansing in Yugoslavia 797 Zenica in Bosnia-Hercegovina used leftover blank copies of the 1991 census forms as the paper for a book of pictures of atrocities committed on the Muslims of Bosnia-Hercegovina. 10. The phrasing of the last two sentences owes much to suggestions by Bette Denich. The reasonableness, in Western philosophical terms, of the competing nationalist claims is explained by Vladimir Cligorov (1995). 11. In this section of the article and the one that follows it, a great deal of emphasis is given to the analysis of Croatian constitutional and legal materials. Unfortunately, in the political climate surrounding the demise of the former Yugoslavia, the analysis of Croatian materials is frequently perceived by Croats as "anti-Croat," "pro-Serbian," or "disproportionate" if less space is devoted to the analysis of Serbian materials. Since this article deals primarily with constitutional and legal materials, however, it focuses on those documents that best exemplify the points under discussion, which are Croatian. Serbian materials are less revealing, not because Serbs manifest the phenomena at issue any less than do Croats, but rather because the Serbian regime of Slobodan Milosevic has put into place constitutional and legal structures that look progressive but that have little bearing on the actions of that authoritarian state (see also Hayden 1992a:660). The criticism is in any event misguided, since it is based on the assumption that Croatian materials should be immune to analysis because of the actions of the Serbs, a proposition that is difficult to defend in regard to academic work. 12. To be sure, this same constitutional article contains a second clause permitting the use, in particular local jurisdictions, of another language and script, "under conditions established by statute" (emphasis added). Both limitations, however, are suspect. If local jurisdictional lines are gerrymandered so that no minority is anywhere a local majority, the constitutional provision becomes meaningless. Furthermore, the subjugation of a supposed constitutional right to ordinary legislation vitiates the right. Thus, for example, a statute providing that one could use the "Serbian language in Cyrillic script" to write to the Minister for Religious Affairs, and only for that purpose, would be constitutional yet serve to deny, in a practical sense, the supposed "right." 13. The "Republic of Serbian Krajina" was destroyed by Croatian military action in August 1995 and almost all of its population was expelled from Croatia. 14. The U.S. Constitution as written in 1 787 did recognize a difference between "free persons" and "all other persons," and excluded "Indians not taxed" (art. 1, sec 2). Furthermore, American citizenship was limited by law to only "white persons" until after the Civil War, and even then naturalization was permitted only to "white persons" and "Africans or persons of African descent" until 1952 (see Gettys 1934). A more appropriate contrast might therefore be the Preamble to the Constitution of India (1950), designed to implement a democratic system in a polity fragmented along lines of caste, religion, and language, as well as social class: We, the people of India, having solemnly resolved to constitute India into a sovereign, secular, democratic republic and to secure to all its citizens: Justice. . . Liberty . . . Equality . . . Fraternity ... do hereby adopt, enact and give to ourselves this constitution. [Constitution of India 1950, preamble] 1 5. Amendment LX to the Constitution of the Socialist Republic of Bosnia and Hercegovina (Službeni List Socijalističke Republike Bosne i Hercegovine), 46:499, no. 21, July 31, 1990. 1 6. The Vance-Owen plan, which ostensibly was aimed at preserving a single Bosnia-Hercegovina, recognized this fact of political life by opposing the division of Bosnia-Hercegovina into only three ethnically determined regions, saying that, a confederation formed of three such states would be inherently unstable, for at least two would surely forge immediate and stronger connections with neighboring states of the former Yugoslavia than they would with the other two units of Bosnia and Herzegovina. [International Conference on the Former Yugoslavia 1992] Nevertheless, the Vance-Owen plan for dividing Bosnia-Hercegovina into ten completely autonomous regions was unrealistic, since it amounted to proclaiming a house divided to be a condominium despite the demonstrated willingness of many of the residents to raze the edifice (see Hayden 1993a). 1 7. From the Constitution of the Federation of Bosnia and Hercegovina (draft of March 13, 1994, 5 P.M.), obtained from the Embassy of Croatia, Washington, DC, in English as one of three (with Croatian and Bosnian) original languages. 18. As is the case with the constitutional provisions (see note 12 above), Serbia is less susceptible to analysis because that state, and the Federal Republic of Yugoslavia that contains it, is hardly a legal state at all. In the present instance, there is no new citizenship law in Serbia, and I am not aware of any analysis of Serbian practices in this regard. The bureaucratic requirements for obtaining citizenship in the new Yugoslavia (Vreme 1992) and the general pressure on minorities in that country (see Mazowiecki 1992:27-36; 1993a:?2^t2), however, indicate that the situation there is likely to be manipulated in order to discriminate against non-Serbs. 19. Zákon o hrvatskom državljanstvu, Národne Novine 1991 #53:1466-1469; amended in Národne Novine 1992, #28:659. 20. Stoicke (1995) distinguishes "cultural fundamentalism" from racism but considers only the political rhetoric surrounding immigration, not that linking emigres with the homeland. It is this latter link that must envision culture as an attribute of birth and thus as substance rather than as simply a code for conduct. 21. Again it is necessary to state that the situation in regard to the determination of Serbian citizenship is no different (Mazowiecki 1993b:26-27). Because Serbia has been an international pariah since 1992, 798 american ethnologist probably few are clamoring to acquire its citizenship. Indeed, I have met many Serbs who would like to acquire Croatian, Macedonian, or even Bosnian citizenship for purely pragmatic reasons, such as facilitating travel and emigration. Most have found this impossible to do, however, even when their parents were from those republics. references cited Anderson, Benedict 1983 Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso. Appadurai, Arjun 1991 Global Ethnoscapes: Notes and Queries for a Transnational Anthropology. In Recapturing Anthropology: Working in the Present. Richard Fox, ed. Pp. 191-210. Santa Fe, NM: School of American Research Press. Bakic-Hayden, Milica 1995 Nesting Orientalisms: The Case of Former Yugoslavia. Slavic Review 54:917-931. Bakič-Hayden, Milica, and Robert Hayden 1992 Orientalist Variations on the Theme "Balkans": Symbolic Geography in Recent Yugoslav Cultural Politics. Slavic Review 51:1-15. Balkan War Report 1995 December 1994-January 1995:5. Basch, Linda, Nina Click Schiller, and Cristina Szanton Blanc 1994 Nations Unbound: Transnational Projects, Postcolonial Predicaments and Deterritorialized Nation-States. Amsterdam: Gordon and Breach. Blau, Peter, Terry Blum, and Joseph Schwartz 1982 Heterogeneity and Intermarriage. American Sociological Review 47:45-61. Boban, Ljubo 1990 Jasenovac and the Manipulation of History. East European Politics and Societies 4:580-592. 1991 Still More Balance on Jasenovac and the Manipulation of History. East European Politics and Societies 6:213-217. Bogosavljevic, Srdjan 1995 Drugi Svetski Rat—Žrtve u Jugoslaviji. Republika (Belgrade), June 1-15, 1995: XI-XVI. Bogosavljevic, Srdjan, Vladimir Goati, Zdravko Grebo, Jasminka Hasanbegovič, Dušan Janjič, Branislava Jojič, and Paul Shoup 1992 Bosna i Hercegovina izmedju Rata i Mira. Beograd and Sarajevo: Forum za Etnické Odnose. Borba 1991 September 30:1 1 Botev, Nikolai, and Richard Wagner 1983 Seeing Past the Barricade: Ethnic Intermarriage in Yugoslavia during the Last Three Decades. Anthropology of East Europe Review 11 (1—2):27—34. Bringa, Tone 1993 National Categories, National Identification and Identity Formation in "Multinational" Bosnia. Anthropology of East Europe Review 11(1-2):69-76. 1994 Bosnia: We Are All Neighbors. London: Granada Television (Disappearing Worlds). Film. Brubaker, Rogers 1992 Citizenship Struggles in Soviet Successor States. International Migration Review 26:269-291. 1993 Nationhood and the National Question in the Soviet Union and Post-Soviet Eurasia: An Institu-tionalist Account. Theory and Society. Burg, Steven, and Michael Berbaum 1989 Community, Integration and Stability in Multinational Yugoslavia. American Political Science Review 83:535-554. Danas 1991 August 6:21 Denich, Bette 1994 Dismembering Yugoslavia: Nationalist Ideologies and the Symbolic Revival of Genocide. American Ethnologist 21:367-390. Djilas, Aleksa 1991 The Contested Country. Cambridge, MA: Harvard University Press. Douglas, Mary 1966 Purity and Danger: An Analysis of Concepts of Pollution and Taboo. London: Routledge. Feral Tribune 1994 January 11. Gettys, Luella 1934 The Law of Citizenship in the United States. Chicago: University of Chicago Press. Gilroy, Paul 1987 There Ain't No Black in the Union Jack. Chicago: University of Chicago Press. ethnic cleansing in Yugoslavia 799 Gligorov, Vladimir 1995 What if They Will Not Give Up? East European Politics and Societies 9:499-51 2. Grass, Günther 1992 The Call of the Toad. New York: Harcourt, Brace, Jovanovich. Hammel, Eugene A. 1993 Demography and the origins of the Yugoslav Civil War. Anthropology Today 9(1 ):4-9. Handler, Richard 1988 Nationalism and the Politics of Culture in Quebec. Madison: University of Wisconsin Press. Hayden, Robert M. 1992a Constitutional Nationalism in the Formerly Yugoslav Republics. Slavic Review 51:654-673. 1992b The Beginning of the End of Federal Yugoslavia: The Slovenia Amendment Crisis of 1989. University of Pittsburgh Center for Russian and East European Studies, The Carl Beck Papers no. 1001. 1993a The Partition of Bosnia and Herzegovina, 1990-93. RFE/RL Research, May 28, 1993:1-14. 1993b On Unbalanced Criticism. East European Politics and Societies 7:577-582. 1994 Recounting the Dead: The Rediscovery and Reinterpretation of Wartime Massacres in Late- and Post-Communist Yugoslavia. In Memory and Opposition under State Socialism. Rubie S. Watson, ed. Pp. 1 67-1 84. Santa Fe, NM: School of American Research. 1995a Constitutional Nationalism and the Logic of the Wars in Yugoslavia. Paper presented at conference on Post-Communism and Ethnic Mobilization, Cornell University, Ithaca, NY, April 21-23 1995. 1995b The 1995 Agreements on Bosnia and Herzogovina and the Dayton Constitution. East European Constitutional Review 4(4):59-68. Herzfeld, Michael 1993 The Social Production of Indifference: Exploring the Symbolic Roots of Western Bureaucracy. Chicago: University of Chicago Press. International Conference on the Former Yugoslavia 1992 Document STC/2/2. October 27:5. Jambresic, Renata 1993 Banija: An Analysis of Ethnonymic Polarization. In Fear, Death and Resistance: An Ethnography of War: Croatia 1991-1992. Lada Feldman, Ines Prica, and Reana Senjkovic, eds. Pp. 73-118. Zagreb: Institute of Ethnology and Folklore Research. Jovic, Borisav 1995 Poslednji Dani SFRJ. Beograd: Politika. Kapferer, Bruce 1988 Legends of People, Myths of State. Washington, DC: Smithsonian Institution Press. Kideckel, David 1995 Human Rights in Former Yugoslavia. Society for the Anthropology of Europe Bulletin 9(3): 1. Knezevic, Anto 1993 Some Questions about a "Balanced" Discussion. East European Politics and Societies 7:155-166. Law on Croatian Citizenship (Zakon o hrvatskom drzavljanstvu) 1991 Narodne Novine #53:1466-1469 1992 Narodne Novine #28:659. Loizos, Peter 1988 Intercommunal Killing in Cyprus. Man (n.s.) 23:639-653. Macartney, Carlile A. 1934 National States and National Minorities. London: Oxford University Press. Mazowiecki, Tadeusz 1992 Report on the Situation in the Territory of the Former Yugoslavia prepared by Mr. Tadeusz Mazowiecki, Special Rapporteur of the Commission on Human Rights. United Nations, General Assembly and Security Council, document A/47/666 and S/24809 (annex), November 1 7. 1993a Situation of Human Rights in the Territory of the Former Yugoslavia. United Nations, Economic and Social Council, document E/CN.4/1993/50, February 10, 1993. 1993b Fifth Periodic Report on the Situation of Human Rights in the Territory of the Former Yugoslavia. United Nations, Economic and Social Council, document E/CN.4/1994/47 November 17. 1994 Sixth Periodic Report on the Situation of Human Rights in the Territory of the Former Yugoslavia. UN document E/CN.4/1994/110 February 21. Morgan, Lewis Henry 1870 Systems of Consanguinity and Affinity of the Human Family. Washington, DC: Smithsonian Institution. Myrdal, Gunnar 1944 An American Dilemma: The Negro Problem and Modern Democracy. New York: Harper and Brothers. Nandy, Ashis 1990 The Politics of Secularism and the Recovery of Religious Tolerance. In Mirrors of Violence: Communities, Riots and Survivors in South Asia. Veena Das, ed. Pp. 69-93. Delhi: Oxford University Press. New York Times 1994 September 1 :A-15 800 american ethnologist Olsen, Mary Kay G. 1993 Bridge on the Sava: Ethnicity in Eastern Croatia, 1981-1991. Anthropology of East Europe Review 11(1-2):54-62. OMRI (Open Media Research Institute [formerly Radio Free Europe/Radio Liberty]) Daily Report 1995 Weekday news service release via e-mail. August 10, Part 2. Petrovié, Růža 1987 Migracije u Jugoslaviji i Etnički Aspekt. Beograd: SSO Srbije. 1992 The National Composition of Yugoslavia's Population, 1 991. Yugoslav Survey 1992(1 ):3-24. Samardzic, Slobodan 1990 Jugoslavia pred Iskušenjem Federalizma. Beograd: Stručná Knjiga. Shoup, Paul 1994 The Bosnian Crisisof 1992. In Beyond Yugoslavia. Sabrina Ramet, ed. Pp. 1 55-1 87. Boulder, CO: Westview Press. Stolcke, Verena 1995 Talking Culture: New Boundaries, New Rhetorics of Exclusion in Europe. Current Anthropology 36:1-24. Todorova, Maria 1994 The Balkans: From Discovery to Invention. Slavic Review 53:453-482. U.S. Government 1995 Population Estimates and Areas of Control in Bosnia and Herzegovina, November 8. Map 737639 (R00895). van der Veer, Peter 1994 Religious Nationalism: Hindus and Muslims in India. Berkeley: University of California Press. Verdery, Katherine 1991 Nationalist Ideology under Socialism: Identity and Cultural Politics in Ceausescou's Romania. Berkeley: University of California Press. Vreme 1991 March 11:31. 1992 August 3:1 6-1 7 1993 March 8:33-34. 1995 August 28:8-11 Woodward, Susan 1995 Balkan Tragedy: Chaos and Dissolution after the Cold War. Washington, DC: Brookings Institution. submitted June 21, 1993 revised version submitted May 26, 1994 revised version submitted September 22, 1995 accepted December 11, 1995 ethnic cleansing in Yugoslavia 801