STÖR National Tolerance in the Former Yugoslavia Randy Hodson, Dusko Sekulic, Garth Massey American Journal of Sociology, Volume 99, Issue 6 (May, 1994), 1534-1558. Stable URL: http://links.jstor.org/sici?sici=0002-9602%28199405%2999%3A6%3C1534%3ANTITFY%3E2.0.CO%3B2-I Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://uk.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. 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. American Journal of Sociology is published by The University of Chicago Press. Please contact the publisher for further permissions regarding the use of this work. Publisher contact information may be obtained at http://uk.jstor.org/journals/ucpress.html. American Journal of Sociology ©1994 The University of Chicago Press JSTOR and the JSTOR logo are trademarks of JSTOR, and are Registered in the U.S. Patent and Trademark Office. For more information on JSTOR contactjstor@mimas.ac.uk. ©2002 JSTOR http://uk.jstor.org/ Tue Sep 10 12:07:21 2002 National Tolerance in the Former Yugoslavia Randy Hodson Indiana University Duško Sekulič George Mason University Garth Massey University of Wyoming This article analyzes patterns of tolerance among nationalities in the former Yugoslavia. Greater tolerance among urban residents, those from nationally diverse republics, and those with nationally mixed parentage and less tolerance among religious people strongly support the modernization theory of ethnic relations. The association of unemployment with intolerance and outbreaks of violence in areas with greater national diversity support theories of ethnic competition. Factors associated with modernism produce greater tolerance but increase the possibility of ethnic conflict. Humanity's dilemma is how to preserve the benefits of modernism for increased intergroup contact and tolerance while avoiding its potentially tragic implications. The demise of Communist party hegemony in Eastern Europe and the Soviet Union has been followed by the recrudescence of ethnic and nationalist political activity (Dobbs 1991; Pfaff 1992). While ethnic and national identities are providing a basis for political solidarity, they are at the same time underpinning conflict among nationalities in several parts of the region (Voirst 1991). This conflict has been particularly visible in the former republics of Yugoslavia, particularly Slovenia, Croatia, and Bosnia-Herzegovina. Over 200,000 casualties and the creation of more than two million refugees have resulted from fighting between rival groups identifying themselves along nationalist lines in these republics. 1 This research was supported in part by a grant from the National Science Foundation through the Yugoslav-United States Joint Fund for Scientific and Technological Cooperation. We would like to thank the reviewers for the American Journal of Sociology for their helpful comments on earlier versions of this article. Please address all correspondence to Randy Hodson, Department of Sociology, Indiana University, Bloomington, Indiana 47405. © 1994 by The University of Chicago. All rights reserved. 0002-9602/94/9906-0004$01.50 1534 AJS Volume 99 Number 6 (May 1994): 1534-58 Yugoslavia The popular press typically attributes these conflicts to long-standing but politically suppressed ethnic hostilities. British Foreign Secretary Douglas Hurd was quoted in a July 20, 1992, Associated Press story as saying, "One of the things one learns from actually being here [Yugoslavia] is that the fears and hatreds which have been unleashed are absolutely formidable." Similarly, Elizabeth Drew (1992, p. 70) has written that "the disappearance of the Iron Curtain allowed long-suppressed— but no less bitter—ethnic hatreds to break out once more." It is not surprising that, once armed conflict begins, many people develop feelings of fear, and in some cases hatred, toward other nationalities (Smith 1981). But it is not at all certain that ethnic hatred is the key operational factor in explaining current conflicts in the former Yugoslavia. Nor is it clear that such animus has been fully present but unexpressed for several decades or centuries as is often implied in the popular press. The concept of "long-suppressed hatreds" carries with it assumptions about Communist party rule and recent political mobilization that may distort rather than illuminate the reasons for the ongoing conflict (see Glenny 1992; Magaš 1989; Suny 1990). The undocumented nature of the assumption that these hatreds were both widely held and a leading cause of current conflicts should encourage skepticism among social scientists. Arguments based on primordial hatreds should not be accepted without an examination of the possible structural underpinnings of current animosities. Sociological theories of modernization and ethnic competition suggest structural explanations for ethnic antagonisms without recourse to theories of primordial hatreds (Deutsch 1961; Hechter 1978). The analysis presented below examines the factors contributing to variations in national tolerance immediately prior to the outbreak of civil and military hostilities in the former Yugoslavia. It explores structural explanations for tolerance that are derived from modernization and ethnic-competition perspectives on ethnicity and ethnic relations. The findings provide substantial support for the modernization theory of ethnic relations. Theories of ethnic competition also receive some support. It appears, however, that theories of ethnic competition are better suited to predicting the subsequent outbreak of violence than to predicting levels of intergroup tolerance. These findings suggest a dilemma for the modern world: Modernization leads to greater intergroup tolerance but also creates the preconditions for subsequent competition and conflict. THE STRUCTURAL BASIS FOR TOLERANCE Modernization Theories Modernization theories see industrialization and its increasingly complex division of labor, enhanced communication and transportation, urbaniza- 1535 American Journal of Sociology tion, and rationalization of social institutions as leading to more univer-salistic principles and more cosmopolitan identities and allegiances (Schermerhorn 1970; Tudjman 1981). This perspective, derived from the American functionalist tradition of the 1950s, was developed against the backdrop of Third World modernization and development in the 1960s and 1970s. Modernization theory treats ethnic identification as premodern, provincial, traditional, and particularistic. According to this theory, ethnic identification's structural basis is the village; its structural support is the persistence of a cultural, political, and economic way of life that reinforces ethnicity as part of a value system lending coherence and consensus to the community (Isaacs 1975; Seton-Watson 1977, chap. 4). In the course of modernization the village ceases to be the focal point of social life, while more inclusive cultural, political, and economic systems come to dominate the social landscape. This vision is also consistent with Marxist views that see class as superseding ethnic relations in the process of industrialization. Where ethnic mobilization and ethnic nationalism occur with development, it is a prelude to class-based social organizations. Increased national diversity and mixing result from industrial development, urbanism, and population movements, and these are seen as important causes of increased tolerance in modernization theory. Traditional ethnic boundaries are more easily maintained when there is little contact between groups (Allport 1958; Belanger and Pinard 1991); the greatest intolerance is expected where there is the least contact between persons of different nationalities (see Lipset and Rokkan 1967). Young urban residents, those people with greater occupational status and higher educational attainment, should express greater tolerance. Persons involved in nontraditional organizations and associations should also be more tolerant, as should those most exposed to the wider world (e.g., consumers of printed and electronic media). Strong religious and traditional values are seen as providing support for intolerance and nationalist political agendas (Hannan 1979, p. 255; Ragin 1979). Ethnic-Competition Theories The ethnic-competition perspective emphasizes the elements of industrialization that foster and intensify ethnic identification and mobilization and encourage ethnic intolerance. As Hannan argues, "The primordial identity view suggests that ethnic distinctions will disappear as processes of modernization increase in scope and intensity. The reactive identity view suggests that this will not be the case as long as power and other institutional differences persist" (Hannan 1979, pp. 254-Í5). The predic- 1536 Yugoslavia tion that modernization will intensify ethnic identification and mobilization clearly differentiates the ethnic-competition model from the modernization perspective. Far from being an aberration, ethnic conflict is seen in the competition model as stimulated by the process of industrialization when diverse groups are thrown into conflict over scarce resources (Ahmad 1991; Barth 1969). The ethnic-competition perspective has spawned several notions about how increased ethnic competition takes place, notions that highlight both labor-market and political processes (Belanger and Pinard 1991). A variety of mechanisms may facilitate the development of ethnic competition alongside modernization: racially split labor markets (Bonacich 1972), employment segregation (Bonacich and Modell 1980), the cultural division of labor (Hechter 1978), center-periphery conflicts (Hechter 1976), the replacement of local control with more inclusive domains of control and sovereignty (Hannan 1979), and the expansion of state systems (Nagel 1986; Nielsen 1985). All of these developments hold the possibility that ethnic competition for jobs, housing, and political power will intensify with modernization (Kposowa and Jenkins 1993). In the ethnic-competition perspective, there is an underlying theme that heightened ethnic identity in ethnically diverse states is a reaction to the economic disruptions and political opportunities of industrialization and modernization (Meadwell 1989; Rogowski 1985). The ethnic-competition model is supported by studies of ethnic mobilization that show that conflict between ethnically distinct peoples may be impelled less by sentiments of intolerance rooted in ethnic or national culture than by the persuasive power of elites pursuing economic and political interests for themselves and their constituents (Ragin 1979). The state's role in promoting nationality as a basis for political mobilization is highlighted in Nageľs (1986, p. 102) proposition that "ethnic mobilization is most likely when political policies are implemented that recognize and institutionalize ethnic differences." Finally, Belanger and Pinard (1991, p. 450; see also Smith 1979) argue that the maintenance of ethnic identities provides an important basis for political action. Olzak (1983, p. 363) argues that "competition theories explain the causal link between modernization and ethnic mobilization," a link unanticipated in the modernization perspective. The ethnic-competition perspective suggests several specific factors that may influence national tolerance. Olzak (1983, p. 358) specifies a rough equality of size among ethnic or national populations as important for fostering ethnic mobilization. Where there are multiple competing groups, intolerance should be greater and may be especially high among members of the dominant group because of the power base provided by numeric dominance (Barth 1969, p. 19; Belanger and Pinard 1991, p. 1537 American Journal of Sociology 448; Korpi 1974; Kposowa and Jenkins 1993). This proposition directly contradicts the "contact hypothesis" of Allport (1958), which argues that tolerance will be fostered when different groups have contact under conditions of proximate equality and interdependent goals—an argument that has been at the core of the modernization perspective. The ethnic-competition model also suggests that population mobility and economic contraction will increase ethnic competition (Olzak 1992, p. 37). Those most affected by economic contraction, such as the unemployed, should exhibit the least tolerance toward other nationalities. Increasing urbanization is also expected to exacerbate nationalist sentiments (Olzak 1983, pp. 367-68). In contrast, modernization theory suggests that urbanization will increase tolerance. To summarize, modernization theory predicts that national diversity and urbanism will lead to greater tolerance. Additional factors modernization theory identifies as contributing to increased tolerance include nationally mixed family structures, participation in nontraditional social and political organizations, contact with the institutions of mass media, higher education, white-collar occupational positions, and youth. Modernization theory predicts that religion will foster intolerance toward other groups. In contrast to modernization theory, ethnic-competition theory predicts that national diversity and urbanism will foster competition and intolerance rather than greater tolerance. Unemployment is also expected to contribute significantly to competition and intolerance. Ethnic-competition theory also suggests that a numerically dominant group may develop intolerance toward other groups as part of a rationale for its aspirations for dominance (Brass 1985; Jackman 1978; Jenkins and Kposowa 1990). FOUNDATIONS OF NATIONALISM IN THE FORMER YUGOSLAVIA The strategy of the League of Yugoslav Communists (LYC) was neither to abolish national identities nor to discourage participation in political life based on nationality (Ramet 1982, 1984a; Rusinow 1985). The LYC did, however, seek to elevate the Yugoslav state to a political status exceeding that of any nationality and sought to relegate national and cultural differences to the area of lifestyle rather than political economy (Denitch 1976, chap. 6; McKay and Verdoodt 1975). Among the various ways the party and state leaders countered claims of nationalities over the state were the 1971 suppression of the Croatian nationalist movement, chastisement for expressions of national superiority (e.g., Slovenia's rationale for opposing the poorer-region-development plans), and efforts to allay fears by some nationalities that they would be subsumed within a Greater Serbia (Ramet 1984a; Rusinow 1977; Stanovčič 1988). 1538 Yugoslavia Many in the Yugoslav regime, including Slovenian Edvard Kardelj (1960), were convinced that modernization and the growing importance of institutions of a more rational character would weaken the hold of nationalist identities. In this vision, an effective educational system, geographic mobility, and increased communication and commerce would undermine the particularistic sentiments of nationalism (see also Connor 1984; Shoup 1968). These political actors implicitly held a modernization view of the eventual demise of ethnic conflict in Yugoslavia. In the former Yugoslavia, economic development and per capita gross domestic product roughly followed a north-south trajectory, with Slovenia in the north being the most affluent republic, and Kosovo and Macedonia in the south the most impoverished (Schrenk, Ardalan, and el Tatawy 1979). The policies of decentralized planning and development (Bičanič 1988; Burg 1983, 1988) make it difficult, however, to apply a core-periphery scheme to the former Yugoslavia. An internal colonial model is equally inappropriate (Burkett 1983; Tyson 1980), especially in light of Yugoslavia's federal policy on underdeveloped regions, which since 1945 sought to shift resources from the wealthier areas (esp. Slovenia and Croatia) to the poorer areas of the country (Burg 1983; Ramet 1984a). High levels of intolerance in both Slovenia and Kosovo, revealed by the analysis below, indicate that economic disparities and relations between poorer and more prosperous regions by themselves are not an adequate explanation of national tolerance. The Communist party implemented the principle of federation in Yugoslavia after taking power in 1945. At the beginning, the federal model was largely a subterfuge for real power concentrated in the Central Committee. Gradually, however, the central power within the party and the government structures was decentralized. More and more influence went not from the center to the republics but from the republican Communist parties to the center. The best example of this important change was the fact that in 1968 the congresses of the republican Communist parties were held for the first time before the federal congress. This step was important in reversing the tendency of the republican congresses to simply ratify and adopt the policies decided by the federal congress. Subsequent to 1968, the decisions of the federal congress increasingly came to be simply the negotiated sum of the decisions and platforms of the various republican congresses, taking into account, of course, the possibility of Tito exercising veto power over decisions at the federal level. Changes similar to those taking place in the Communist party were also introduced in government structures. In 1953 the Chamber of Nationalities was abandoned as a separate parliamentary body and it became an almost indistinguishable part of the Federal Chamber (Savezno 1539 American Journal of Sociology Vijece). With Amendment 8 of the constitution (April 18, 1967), however, the Chamber of Nationalities was reestablished as a visible structure of the federal congress with even greater power. In the same year, the principle of "national parity" in representation was introduced in leading party organs. This principle was extended to all organs of the federal government in 1971. With these amendments and changes, the presidency of Yugoslavia was constituted as a body reflecting the "multinational structure" of the Yugoslav federation. The most significant element of dispersed republican power, introduced in 1971, was de facto veto power of every federal unit (including autonomous provinces) on any decision made on the federal level. From the mid-1960s forward, the national principle became deeply institutionalized. The process of institutionalization was finalized in the 1974 constitution, where the national principle became the primary dimension on which the political system was organized (Bilandžič 1985). The goal of the federal Communist party from the mid-1960s forward was to foster genuine decentralization (federalism in the government and in party structures) but without the mobilization of national groups. The fulfillment of national rights was to be satisfied by giving increased rights to the republican Communist parties to represent their respective nations. In socialist Yugoslavia the economic and political policies of the League of Yugoslav Communists explicitly sought to diminish the cultural division of labor between nationalities (Denitch 1990, chap. 7; Mas-sey, Hodson, and Sekulič 1992). The dynamics generated by state policies supporting a multinational state were quite complex and contained many contradictory elements. The expression of nationalist sentiments was allowed and even facilitated by these policies. Old national slogans and traditional national sentiments became openly expressed by intellectuals and other groups outside the Communist party. Traditional national goals also quickly became parts of republican Communist party ideologies and strategies. But the political center, dominated by Tito, was careful in distinguishing between decentralization under Communist party control and spontaneous national mobilization not controlled by the party. Federalism under party control was meant to satisfy aspirations for national autonomy and expressions of national identity. However, it was unacceptable for the republican Communist parties to seek alliances with traditional nationalist groups in pursuit of national goals (Warwick and Cohen 1985). A good example of this unwillingness to allow traditional nationalist groups to enter the political arena was the strong federal reaction to the popular nationalist movement in Croatia in 1971. When Tito concluded that the Communist party of Croatia was making too many concessions to Croatian nationalists, he purged the higher ranks of the party in 1540 Yugoslavia Croatia. The party elite in the republics were always in danger of being pressured by the center if they made movements to "democratize" and allowed nationalist movements to play an independent role in the political sphere. Similarly, Tito's policies of independence for the newly designated nations was always within the framework of the Communist party. The decision to allow Muslims a separate official national identification was a decision negotiated within the federal party, not in Bosnia, where most Muslims lived. Similarly, increased autonomy for Kosovo was based on a close collaboration of a new party elite of Albanians in Kosovo with the federal party. The central contradiction generated by the policies of the federal Communist party leadership was that decentralization through federalism produced a fragmented political structure with eight Communist parties, each having monopoly over its own territory and veto power over federal decisions. Each republican party quickly came to rely on traditional national goals to articulate and legitimate its bargaining position relative to other republican Communist parties (Kourvetaris 1993, p. 5). This tendency was checked during Tito's life, because the center was still in a position to suppress such "nationalist deviations." With Tito's death, the political center lost the power to effectively check these centrifugal forces. In addition, starting in the 1970s and accelerating in the 1980s, an economic crisis was eroding peoples' standard of living, unemployment was increasing, and many people doubted the prospects for a reversal of economic fortunes (Lydall 1989). These political and economic forces combined to set the stage for the dissolution of the Yugoslav state in the 1990s (Sekulič, Massey, and Hodson 1994).2 National Groups in Yugoslavia Most of the people of the former Yugoslavia are ethnic Slavs. While writers in the West often refer to them as if they were members of different ethnic groups, in fact most people have a common ancestry, speak the same language, and share in dress, food, and lifestyle a similar culture. There are religious differences to be sure, but there is little evidence that religious differences per se provide a sufficient basis for differing nationalist claims (Ramet 19846). In addition to Slavic peoples, there are non-Slavic peoples (e.g., Gypsies and Jews) and other national groups within what were the borders of Yugoslavia. These include Albanian Muslims in Kosovo and Macedonia, Hungarians, Slovaks, and Romani- 2 Failed federalist strategies and economic crisis have similarly been identified as key precipitating factors in the eventual dissolution of the former Soviet Union (d'En-causse 1993). 1541 American Journal of Sociology ans in the Vojvodina, as well as several thousand Turks, Italians, Poles, Russians, Germans, and Ukrainians spread throughout the country (Petro vie 1983). The census in Yugoslavia asked persons to identify themselves with a particular nationality or group of national origin. The conventions followed by the census reflected the social consciousness of people in the former Yugoslavia. The vision of Yugoslavia as a multinational nation recognized foremost the historical experiences of groups of people who possessed distinct identities based only loosely on religious and other visible cultural manifestations but who saw themselves as having unique historical experiences that distinguished them from one another (Banac 1984; Cohen 1982; Warwick and Cohen 1985). In the census, persons who indicated that they were "Yugoslavs" were described as having no nationality. The Communist party rejected the idea of the creation of a new Yugoslav nationality that would replace the historically formed nations of Slovenes, Croats, and Serbs. The idea that these nations were just tribes that would provide the foundation of a new unified nationality had been preeminent during the period of the first Yugoslav state. During that period, however, such a vision was strongly discredited among non-Serbs as a cover for the formation of a Greater Serbia. The Communist party wanted to avoid repeating the instability of the first Yugoslavia, which was characterized by political conflict principally between Croats and Serbs. Tito and the communists were hoping that industrialization and modernization would erode old national divisions, but they had concrete historic evidence that any attempt to force Yugoslavism could provoke even greater instability. Their formula for the transition period was federalism, equal rights for all nations, and a vision that in the future, with the final ascension of communism and economic development, national affiliations would be eclipsed in importance. In adopting a policy of legitimating national identifications, the Yugoslav communists were adopting a policy toward nationalities analogous to Stalin's famous slogan that in the first phase of socialism the class struggle should intensify in order to wither away later. The Yugoslav Communist party promoted the ideology that the national feelings should be allowed to develop to their full potential so that they would provide the foundation for later communist internationalism. The results of this policy were that Macedonians got full recognition of their nationhood only in communist Yugoslavia and that the national feelings of Bosnian Muslims were recognized for the first time in communist Yugoslavia. The reluctance to recognize a Yugoslav national identification on an equal footing with other national identifications was reflected in the fact that in the census the Yugoslav national identification was subscripted 1542 Yugoslavia with the explanation, "having no identifiable nationality." In addition, an elaborate official vocabulary was introduced to describe nationality. Narod meant Slav nations having only Yugoslavia as their mother state: Slovenes, Croats, Muslims, Serbs, Macedonians, and Montenegrins were such nations. Národnost meant national minority: Albanians, Hungarians, Turks, Italians, and others living in Yugoslavia but having some other mother state. Gypsies were included in this latter category. Commitment to the Yugoslav nation-state was identified through the concept of citizenship (drzavljanstvo). Thus, Croats, Serbs, other nations, and other national minorities were all Yugoslav citizens having separate national identities or belonging to some minority nationality but all having a commitment to the Yugoslav state. Serbs made up 36.3% of the nearly 23 million people in Yugoslavia in 1981. The majority (59.8%) lived in Serbia proper, 16.2% lived in Bosnia, 6.5% lived in Croatia, and 16.2% lived in autonomous provinces that are now under Serbia's control: the Vojvodina (13.6%) and Kosovo (2.6%). Of the 1.3 million Muslims in Yugoslavia, 81.5% lived in Bosnia. Another 7.6% lived in Serbia proper. Slovenes were heavily concentrated in Slovenia: 97.7% of Yugoslavia's 1.7 million Slovenes lived in Slovenia in 1981. Of 4.5 million Croats, nearly four out of five (78%) lived in Croatia, 17.1% lived in Bosnia, and 3.4% lived in Serbia. Among the 1.3 million Albanians, 70.9% lived in Kosovo and 21.8% lived in Macedonia. Concentrated similarly to Slovenes in Slovenia, 95.5% of all Macedonians (1.2 million total) lived in Macedonia. The Vojvodina and Bosnia were the most nationally heterogeneous regions of Yugoslavia, with no single nationality making up more than 54.4% (Serbs in the Vojvodina) and 39.5% (Muslims in Bosnia) of their populations, respectively. The most homogeneous regions were Slovenia, with Slovenes constituting more than 90% of the population, and Kosovo, where 90% of the population was estimated to be Albanian. Between these extremes were Montenegro, with 68.5% of the population identifying themselves as Montenegrin, Croatia, with three-fourths of the population being Croats, Macedonia, with two-thirds of the population being Macedonian, and Serbia proper, where 85% of the population identified themselves as Serbs (Štatistickí Godisnjak Jugoslavije 1987, p. 449). The analysis below focuses on the structural factors that contributed to different levels of tolerance and intolerance among the many nationalities in the former Yugoslavia—many of which reflect the dramatic economic, social, and political transformations of socialist Yugoslavia in the post-World War II period. That people were willing and able to identify themselves in terms of nationality and the widespread belief that national intolerance was a significant factor precipitating the ongoing conflicts in 1543 American Journal of Sociology the former Yugoslavia makes an understanding of these factors doubly important. RESEARCH DESIGN AND MEASUREMENT In late 1989 and early 1990 the Consortium of Social Research Institutes of Yugoslavia conducted door-to-door interviews in all the republics of Yugoslavia, utilizing a random sample of households. The completed sample was composed of 13,422 adults above the age of 18 and was distributed across republics in accord with their populations. The sampling design resulted in a disproportionate number of male respondents. The occupational, educational, and age distributions of the sample, however, closely approximate those reported in the 1981 census. To adjust the sample to be representative of the gender distribution of Yugoslavs above the age of 18, male respondents were weighted by a factor of .769133 and female respondents were weighted by a factor of 1.400747. This weighting is derived from the 1981 census of Yugoslavia and reproduces in the analysis sample the gender distribution reported in the census. The questionnaire asked over three hundred items, resulting in a wealth of data and allowing the construction of indices of national tolerance and religiosity. National Tolerance Respondents were asked their level of agreement on a five-point scale with six propositions concerning tolerance toward other nationalities: nationality should be a central factor in choosing a marriage partner; nationally mixed marriages are more unstable than other marriages; every nation should have its own state; people can feel completely safe only when the majority belong to their nation; among nations it is possible to create cooperation but not full trust; without leaders every nation is like a man without a head. All items had item-total correlations above .4 except the last item, which scaled badly (61.5% of respondents were grouped in the "strongly agree" category). Accordingly, this item was dropped from the scale. The remaining items were reverse scored so that high scores indicate greater tolerance. The resulting five-item scale has a reliability index of .72. Explanatory Variables Two ecological variables are utilized in the analysis: national diversity in each republic and percentage representation of national groups within 1544 Yugoslavia republics. Modernization theory leads to the expectation that greater national diversity and national mixing in a republic will be associated with greater tolerance toward other groups. National diversity is measured with the Index of Qualitative Variation (Bohrnstedt and Knoke 1988, p. 76-77): Index of Qualitative Variation = ——----- , (K — 1)1 K where K = the number of categories and p = the proportion of cases in the ith category. The index ranges from "1," indicating that the cases are spread evenly over the categories, to "0," indicating that all cases are in a single category. The percentage representation of groups within republics taps the potential power base of each national group in a republic. Group power is seen in the ethnic-competition model as a basis of group mobilization. Each national group's power base in a republic is measured by the percentage representation of the group in the republic (Petro vie 1983). Four additional sets of characteristics are also expected to influence tolerance toward other nationalities: demographic factors, social status, social participation, and religiosity. Demographic factors include age, birth residence, current residence, and nationally mixed parentage and marriage. Modernization theories of ethnic tolerance suggest that older people and rural residents are less tolerant, because young people and urban residents will be the ones most exposed to the forces of modernism. (For a contrasting view based on the ethnic-competition model, see Olzak [1983], pp. 367-68.) Respondents were asked if they had been born in a village, village center, town, town center, city, or regional center and were also asked a similar question about their current residence. The modal birth residence was village with the median being between village center and town. The modal current residence was also a village, but the percentage living in towns increased and the percentage living in cities increased dramatically, rivaling village as the modal category. Nationally mixed family structures are expected under modernization theory to lead to greater tolerance (Bonacich and Modell 1980). Nationally mixed parentage was ascertained by comparing the nationalities of respondents' parents. Nationally mixed marriage was ascertained by comparing the nationality of a respondent with the nationality of his or her spouse. Approximately 8% of respondents were the offspring of nationally mixed parentage and 8% were in nationally mixed marriages. 1545 American Journal of Sociology Gender and marital status are included as controls; these are coded as binary variables (male = 1; married = l).3 Approximately 71% of respondents were classified as married. Two aspects of social stratification, education and occupation, are central to modernization theory. More educated people and people in white-collar occupations are expected to be more tolerant under the modernization model. Education is coded as years of schooling completed and averages just over 10 years. Occupational position is coded as nine binary variables specifying the categories of professionals, managers, clerical workers, police, retail workers, industrial and construction workers, peasants, unemployed people, and retired people. According to ethnic-competition theory, people who are unemployed can be expected to have increased feelings of intolerance toward other national groups because of intensified competition for jobs: "Economic contraction in combination with high immigration flows raises levels of ethnic competition, which in turn increases rates of ethnic collective action" (Olzak 1992, p. 37). Participation in political organizations and involvement in the broader society through reading the news are expected by modernization theory to increase the level of national tolerance and dilute allegiances built narrowly on ethnic solidarity. We use three sources of data on political involvement: membership in the League of Yugoslav Communists (LYC), office holding in workplace organizations, and office holding in community organizations. Membership in the LYC was a widely dispersed status. Individuals who were members of the LYC did not necessarily hold elite positions and did not necessarily disproportionally enjoy the privileges such power might provide. Nevertheless, membership in the LYC can be expected to imply a greater commitment to explicitly articulated national goals, among which was tolerance for different nationalities and support for a pluralist, multinational state. The binary party membership variable is coded "yes = 1" for those who either report currently being members of the LYC or having been members in the past. About 34% of respondents reported either being in the LYC currently or having been members in the past. The other two measures of political involvement are participation in political organizations in either the workplace or the community. Respondents were asked if they occupied an elected position at their workplace (yes = 1). Respondents were also asked if they held any elected positions in community organizations or if they were active in community organi- 3 Unmarried people currently cohabitating were also coded as married, under the assumption that their living arrangements would be more consequential for their attitudes of tolerance than the legal distinction between marriage and cohabitation. 1546 Yugoslavia zations (yes = 1). About 18% and 20% of respondents, respectively, reported active participation in work or community organizations. Literacy and newspaper reading provide greater contact with the world and are expected to increase national tolerance under the modernization theory of ethnic relations. In addition, newspapers and television were directly under LYC control prior to 1989, and the explicit agenda of the party was to encourage tolerance among nationalities. Other media such as radio and magazines were only slightly less controlled. Regularly reading the news can thus be expected to increase national tolerance. Respondents were asked whether they read the newspaper daily, weekly, monthly, or never (coded 4, 3, 2, or 1, respectively). Later in the questionnaire, respondents were asked to identify their three most common leisure activities. Respondents could identify "reading news" as their most important leisure activity, their second most important, their third most important, or not at all important (coded 4, 3, 2, or 1, respectively). Responses to these two questions were summed to create a seven-point scale of reading the news with scores ranging from two to eight. Level of religiosity is seen as an important negative influence on tolerance by modernization theorists. Five questions were asked about religiosity. Respondents were asked to identify, on a three-point scale, their level of belief in God, in life after death, and in the idea that God created people. Also, they were asked how often they attended religious services (never, monthly, weekly, or daily) and if their children attended religious schools. All items scaled positively with item-total correlations above .4. The resulting five-item standardized scale has a reliability index of .86. RESULTS Average tolerance levels are strongly differentiated between republics, ranging between 1.71 and 3.88 on a 5.00-point scale. Republics and autonomous regions are listed in table 1 by descending order of tolerance. People in Bosnia and the Vojvodina had the highest levels of tolerance. The average level of tolerance in Serbia, the largest republic, was identical to the average for the country as a whole. Croatians were more tolerant than Serbians but less tolerant than those who live in the most tolerant republics of Bosnia and the Vojvodina. Kosovo had the lowest average level of tolerance and is a true negative outlier in this regard. Average tolerance levels in the republics closely parallel the level of national diversity in each republic. Bosnia and the Vojvodina, which have the highest average tolerance, also have the greatest national diversity. Tolerance levels descend monotonically with declining diversity through the six most diverse republics. Only Macedonia and Kosovo, the two least tolerant republics, deviate from this pattern. In Macedonia 1547 American Journal of Sociology TABLE 1 Tolerance and National Diversity by Republic: Yugoslavia, 1989 Diversity Republic Tolerance Index N Bosnia .......................... 3.88*** .64 2,020 Vojvodina ..................... 3.83*** .61 1,315 Croatia ......................... 3.63*** .45 2,597 Montenegro ................... 3.45*** .45 909 Serbia ........................... 3.28 .27 3,321 Slovenia ........................ 2.67*** .19 1,299 Macedonia ..................... 2.53*** .41 973 Kosovo ......................... 1.71*** .39 988 Total ......................... 13,422 Average ..................... 3.28 .42 Note.—Statistical significance is measured by two-tailed i-tests; statistical tests for tolerance are based on contrasting the mean for each republic with the mean for all other republics. *P < .05. ** P < .01. *** P < .001. and Kosovo diversity levels are only slightly below the average across republics. The strong association of diversity and tolerance supports a central tenet of modernization theory that argues that diversity and increased intergroup contact lead to greater tolerance. Regression coefficients indicating the effects of the various structural variables on national tolerance are presented in table 2. Examination of these coefficients allows us to see which other structural variables beyond national diversity also have consequences for tolerance and the direction of their effects. The two ecological variables measuring the national diversity of republics and the majority status of groups within republics are both significant determinants of tolerance. As noted above, national diversity leads to greater tolerance; this is the second largest standardized effect in the model. Majority status in a republic leads to less tolerance and this is the third largest standardized effect in the model. The finding that majority status increases intolerance provides support for the competition theory, which proposes that dominant groups may express more ethnic or national intolerance in order to legitimate their aspirations for dominance. Age has a significant positive effect on tolerance: Older people are more tolerant than younger people. This finding is inconsistent with the view that the young are the carriers of the forces of modernization leading 1548 TABLE 2 Regression of Tolerance on National Composition, Demographic, Social Status, Participation, and Religiosity Variables: Yugoslavia, 1989 Independent Variable Mean SD b SE ß National composition: Diversity index ..............................48 .19 1.051*** .050 .19 Majority status ............................. 57.67 30.48 -.382*** .034 -.11 Demographic variables: Age ............................................ 40.63 14.56 .376*** .085 .05 Married........................................71 .45 -.074*** .020 -.03 Male ............................................49 .50 -.156*** .018 -.07 Urbanism (origins) ......................... 2.30 1.62 -.004 .006 -.01 Urbanism (current) ........................ 3.30 1.81 .030*** .006 .05 Mixed parentage ............................08 .28 .102*** .031 .03 Mixed marriage .............................08 .28 .349*** .033 .09 Social status: Education .................................... 10.07 3.96 .008* .003 .03 Professional...................................12 .32 -.066 .035 -.02 Manager .......................................06 .24 -.114** .043 -.03 Clerical ........................................09 .29 .019 .033 .01 Police ..........................................01 .11 .076 .073 .01 Retail worker ................................10 .29 .108*** .031 .03 Industrial or construction worker.......22 .42 .000 ...... Peasant ........................................15 .36 .097** .031 .03 Unemployed ..................................12 .32 -.285*** .031 -.09 Retired.........................................13 .34 -.035 .036 -.01 Social participation: Communist party ...........................34 .45 .025 .022 .01 Work organization ..........................20 .40 .095*** .024 .04 Civic organization ..........................18 .38 .058* .025 .02 Readsnews .................................. 4.56 1.30 -.027*** .007 -.03 Religiosity: Religiosity scale .............................00 1.00 -.309*** .009 -.29 Constant ........................................ 2.865*** R2 ................................................. .241*** Note.—Statistical significance is measured by two-tailed i-tests; industrial or construction worker serves as the reference category for estimating the regression coefficients for occupation; the regression coefficients and SEs for age and majority status have been multiplied by 100 for ease of presentation; N = 13,422. * P < .05. ** P < .01. *** P < .001. American Journal of Sociology to greater tolerance. Urban residents, however, are more tolerant than residents of villages and rural areas. There is no independent effect of rural or urban origins (as distinct from current residence). Nationally mixed parentage and marriages both tend to increase tolerance. Married people are less tolerant than unmarried people, and men are less tolerant than women. The social status variables have fewer and weaker effects on tolerance than the national composition, demographic, social participation, or religiosity factors. Education increases tolerance, but it is significant at only the .05 level, which is not a strong finding in a sample of 13,422 cases. Industrial and construction workers serve as the baseline for evaluating occupational differences in tolerance, and many of the other occupations and positions cannot be statistically distinguished from industrial and construction workers. Retail workers and peasants are somewhat more tolerant than industrial and construction workers, but managers are less tolerant. The strongest contrast in average tolerance levels across social positions is for unemployed persons, who evidence high levels of intolerance for other national groups. The average tolerance level of unemployed persons is more than a quarter of a point lower than that of employed industrial and construction workers. This finding provides additional support for the ethnic-competition model. Three of the four social participation variables have significant effects on tolerance. Participation in a civic or workplace organization increases tolerance, as suggested by modernization theory. Being in the Communist party, however, has no effect, either positive or negative, on tolerance. This finding contrasts strongly with official party rhetoric at the federal level and highlights the importance of nationalist sentiments and leaders in republic-level party politics in the period immediately prior to the dissolution of Yugoslavia. Reading the news regularly has a negative effect on tolerance. This effect again suggests the extent to which the control of key institutions at the republican level was held by those sympathetic to nationalist sentiments prior to the dissolution of Yugoslavia. Religiosity has the largest standardized effect on tolerance. Every standard deviation increase on the religiosity scale is associated with a decline of almost one-third of a point in tolerance. The powerful negative effect of religiosity on tolerance provides strong support for the modernization hypothesis that traditional religious beliefs provide a supportive ideological base for national and ethnic intolerance. The patterns of tolerance between the major nationalities within republics are reported in table 3. These coefficients allow us to examine the patterns of tolerance and intolerance for the major national groups within each republic. The tolerance levels reported in table 3 are stan- 1550 Yugoslavia TABLE 3 Adjusted Tolerance Levels by Majority/Minority Status within Republics: Yugoslavia, 1989 Republic Majority Group Principal Minority Group Bosnia...................................... 3.88 (Muslims) 3.82 (Serbs) Vojvodina ................................. 3.65* (Serbs) 3.86 (Hungarians) Croatia ..................................... 3.60*** (Croats) 3.93 (Serbs) Montenegro ............................... 3.32 (Montenegrins) 3.40 (Muslims) Serbia ...................................... 3.24 (Serbs) 3.39 (Muslims) Slovenia ................................... 2.71*** (Slovenes) 3.33 (Croats) Macedonia ................................ 2.61*** (Macedonians) 1.84 (Albanians) Kosovo ..................................... 1.79* (Albanians) 1.99 (Serbs) Note.—Statistical significance is measured by two-tailed i-tests; tolerance levels are adjusted for the demographic, social status, social participation, and religiosity characteristics analyzed in table 2; tests for majority vs. minority differences in tolerance contrast these groups within each republic; N = 13,422. *P < .05. ** P < .01. *** p < .001. dardized on the variables presented in table 2 with the exception of the two national composition variables. In other words, the adjusted tolerance levels reported in table 3 represent the differences that would exist if respondents of different nationality from each republic were identical in terms of demographics, social status characteristics, level of political participation, and religiosity. Remaining republic-level differences in tolerance are manifest in table 3 as differences between rows. Differences in tolerance between the majority group and the principal minority group in a republic that remain after standardization are revealed by differences between the two columns. Republics are listed in the same order as in table 1, from highest to lowest based on unadjusted mean levels of tolerance. Tests of statistical significance reported in table 3 evaluate the differences in average tolerance level between the majority group and the principal minority group in each republic. The pattern of tolerance levels in table 3 indicates that the majority group in a republic is generally less tolerant than the principal minority group even after standardizing on population characteristics. This pattern is significant for the Vojvodina, Croatia, Slovenia, and Kosovo. The minority group is significantly more intolerant than the majority national group only in Macedonia, where the principal minority group is Albanian. Albanians, whether in Kosovo, where they are the majority, or in Macedonia, where they are the principal minority, have the greatest intolerance toward other groups of any national group in the former Yugoslavia. 1551 American Journal of Sociology Consistent with the overall pattern, Serbs are more tolerant as a minority group in Bosnia and Croatia than they are in Serbia, where they are a strong majority. In Kosovo, however, the Serb minority is less tolerant than Serbs living as the majority group in Serbia. In Kosovo, it appears that the extremely high level of intolerance pulls all groups toward greater intolerance and overrides the general pattern of national groups being less tolerant as majorities than as minorities. In this highly volatile republic, the average tolerance level of Serbs, the largest minority group, drops more than a point below its level in Serbia and almost two points below its level in Croatia. A pattern similar to that in Kosovo exists for Croats in Croatia and Slovenia. Minority Croats in Slovenia are more tolerant than majority Slovenians, but less tolerant than Croats in their home republic of Croatia, which is more nationally diverse and tolerant. The most tolerant groups are Muslims in nationally diverse Bosnia and the Serb and Hungarian minorities in the ethnically diverse republics of Bosnia, the Vojvodina, and Croatia. Slovenia, Macedonia, and Kosovo, in which a single nationality composes more than 90% of the population, evidence the highest levels of intolerance even after standardization on population characteristics. Tolerance is affected by the characteristics of individuals, as we have seen in table 2, but not to such an extent that the patterns of differences between republics or between majority and minority national groups in republics are erased. These patterns reaffirm the findings from table 2 that the national diversity of republics and the majority or minority status of national groups within republics are powerful influences on tolerance toward other nationalities. Even adjusted patterns of tolerance continue to be strongly differentiated by republic of residence and by nationality within republics, with average levels of tolerance sometimes differing by a point or more between republics or nationalities. Tolerance is greater in nationally diverse republics than in more homogeneous republics. The contrasts between majority and minority groups within republics are also substantial though not generally as large at those between republics. Beyond the differences in tolerance levels evidenced in table 3 that are driven by republic-level differences in national diversity and majority or minority status differences, there are some notable differences between nationalities and republics that can only be explained by historical and cultural factors. High levels of intolerance in Kosovo are most notable in this regard and reflect a long history of unresolved rivalries between nationalities. Chronic low-level conflict between Serbs and Albanians has been the rule in Kosovo for decades, exacerbated by the emigration of several thousand Serbs in the 1980s and by Serbia's suspension of autonomous-region status in both Kosovo and the Vojvodina in 1989. The patterns in table 3 thus illustrate not only the powerful effects of national 1552 Yugoslavia diversity and majority or minority status in republics but also unique historical and cultural factors. DISCUSSION Both the modernization and ethnic-competition models of ethnic relations receive support from the above analysis. Support for the role of modernization in generating greater national tolerance is provided by the powerful effects of national diversity, urbanism, nationally mixed family structures, and participation in formal organizations at the civic and workplace levels. The strong association of religiosity with intolerance also supports a key tenet of modernization theory that identifies religious traditionalism with ethnic intolerance. The effects of social status characteristics provide less support for either modernization theory or ethnic-competition theory. Education has only a modest dampening effect on intolerance. And occupational differences in tolerance are remarkably flat. Intermarriage, urbanism, social participation, and the decline of religiosity are all more important for generating greater tolerance than the effects of industrialization on the mix of occupational positions. The prediction from modernization theory is that industrialization and its associated occupational transformations imparts greater universalism and acceptance of ethnic and national differences. It appears that the general effects of modernization on intermarriage and social participation are more important in generating increased tolerance than the effects of industrialization on education and the occupational structure. Investigators within the ethnic-competition perspective have also suggested that industrialization impels individuals into more rationally governed institutional contexts, which can enhance and generalize competition and, thus, increase intolerance (Dutch and Gibson 1992; Nielsen 1980; Olzak 1983, p. 362; 1992; Pinard and Hamilton 1978). The limited nature of occupational and educational effects thus fails to support either the modernization or ethnic-competition theories insofar as they envision a strong role of industrial transformation in determining levels of intergroup tolerance. Two factors in our model of national tolerance have effects inconsistent with the modernization theory of ethnic relations: Young people and those who read the news more frequently are less tolerant toward other nationalities. These effects may be unique to Yugoslavia, where relatively autonomous Communist parties controlled the mass media in each republic and where the older generations carried the dream of a unified Yugoslav state. The strong association of unemployment with intolerance provides support for that aspect of the ethnic-competition model that rests on a 1553 American Journal of Sociology vision of competition over scarce economic resources. The ethnic-competition theory of ethnic relations is also supported by the negative effect of majority status in a republic. The base of support provided by numeric dominance in a republic appears to have encouraged the growth of intolerance toward minority national groups (Enloe 1973; Roosens 1989; Young 1976). Numeric dominance may have been especially important in Yugoslavia, where political power was the chief resource over which groups struggled (Stanovčič 1988; Tudjman 1981). The association of newspaper reading with intolerance also supports the idea that intolerance toward other groups was fueled by local and regional political elites who controlled information and sought to cultivate and exploit sentiments of intolerance. These results raise questions about Belanger and Pinard's (1991, p. 447) conclusion that the accumulated "evidence for the presumed modernization/ethnic movement link is not strong." CONCLUSIONS Developments in Yugoslavia subsequent to the winter of 1989-90, when the survey data for our analysis were collected, allow some important questions to be asked about intolerance and the eventual outbreak of armed conflict. The implicit hypothesis underlying popular views of the fighting in the former Yugoslavia is that where nationalist intolerance is greatest conflict will be highest. This view would be supported if there was a close correspondence between these two phenomena. Such a correspondence would provide support for the theory that the resurgence of suppressed nationalist hatreds lies behind current conflicts. The above findings suggest, however, that there is little such correspondence. Intolerance was high in relatively homogeneous Slovenia, which seceded from Yugoslavia with a minimum of conflict. Ironically, in Croatia, where tolerance was greater, conflict was much more intense. And conflict has been most brutal and prolonged in Bosnia, where tolerance was at its highest level. Tolerance and armed conflict appear to be quite distinct phenomena, and they appear to have distinct patterns of causation. What then is their relationship and what have we learned from the above analysis about their causes? One possible explanation for observed patterns of tolerance and conflict that is consistent with the data is that heterogeneity leads to both tolerance and conflict (cf. Hechter 1987; Olzak and Nagel 1986). Heterogeneity provides the conditions fostering increased tolerance among individuals of diverse nationality through increased contact, but it also creates the conditions under which different national groups engage in competition over scarce resources. In heterogeneous settings national groupings can provide an organizational basis for mobilization. 1554 Yugoslavia Elites may find such situations conducive to mobilizing efforts in pursuit of ends that come to be defined in national terms (Kposowa and Jenkins 1993; Ragin 1979). These mobilized national groups may pursue these ends through armed conflict, attacking even those against whom they initially bear only limited animosities. Conversely, homogeneity appears to lead to greater intolerance but a lower likelihood of open conflict. Minorities may suffer prejudice and discrimination in such settings—as is suggested by the high levels of intolerance observed in the more homogeneous republics of the former Yugoslavia. But the preconditions for armed conflict are less present in more homogeneous settings—one group is clearly dominant and armed conflict to adjudicate this situation is unlikely. It is a great irony that some of the factors that most foster tolerance—such as the national diversity of regions and national intermarriage—are objects of attack under policies of ethnic cleansing. Yugoslavian state policies of modernization and controlled nationalism may have led to greater tolerance, but they also set the stage for the mobilization of groups around nationality, especially within nationally diverse and highly autonomous republics. The policy of multinationalism promoted by the LYC gave salience to national identities. The economic crisis of the 1980s and the decline of federal Communist party influence gave new life to political appeals based on nationality. In the absence of other forms of political cohesion, nationalism appears to be the most readily available unifying doctrine in much of Eastern Europe at the end of socialist formations. The dilemma facing the citizens and leaders of the former socialist nations, as well as significant other parts of the world, is how to realize the benefits that modernization provides for increasing intergroup tolerance while avoiding the potential it creates for heightened ethnic and national conflict. Many of the changes associated with modernization lead to greater tolerance. Other aspects of modernism, however, create the potential for heightened intergroup competition and conflict over scarce economic and political resources. These contradictory consequences of modernism, rather than primordial hatreds, appear to be at the root of current conflicts in the former Yugoslavia and elsewhere in Eastern Europe and the former Soviet Union. Bosnia enjoyed the highest level of tolerance of any Yugoslav republic, but this increased tolerance proved insufficient to outweigh the political forces emanating from its extremely diverse social fabric. REFERENCES Ahmad, Syed Nesar. 1991. Origins of Muslim Consciousness in India. Westport, Conn.: Greenwood. 1555 American Journal of Sociology Allport, Gordon. 1958. The Nature of Prejudice. New York: Doubleday. Banac, Ivo. 1984. The National Question in Yugoslavia. Ithaca, N.Y.: Cornell University Press. Barth, Fredrik. 1969. Introduction to Ethnic Groups and Boundaries, edited by Fred-rik Barth. Boston: Little, Brown. Belanger, Sarah, and Maurice Pinard. 1991. "Ethnic Movements and the Competition Model: Some Missing Links." American Sociological Review 56 (4): 446-57. Bičanič, Ivo. 1988. "The Fractured Economy." Pp. 120-42 in Yugoslavia: A Fractured Federalism, edited by Dennison Rusinow. Washington, D.C.: Wilson Center Press. Bilandžič, Dušan. 1985. Historija Socij'alisticke Federatívne Jugoslavije (The history of the Yugoslav Socialist Federation). Zagreb: Skolska Knjiga. Bohrnstedt, George W., and David Knoke. 1988. Statistics for Social Data Analysis, 2d ed. Itasca, 111.: F. E. Peacock. Bonacich, Edna. 1972. "A Theory of Ethnic Antagonism: The Split Labor Market." American Sociological Review 37:547-59. Bonacich, Edna, and J. Modell. 1980. The Economic Basis of Ethnic Solidarity: Small Business in the Japanese American Community. Berkeley and Los Angeles: University of California Press. Brass, Paul. 1985. "Ethnic Groups and the State." Pp. 1-57 in Ethnic Groups and the State, edited by Paul Brass. Totowa, N.J.: Barnes & Noble. Burg, Steven L. 1983. Conflict and Cohesion in Socialist Yugoslavia. Princeton, N.J.: Princeton University Press. ---------. 1988. "Political Structures." Pp. 9-22 in Yugoslavia: A Fractured Federalism, edited by Dennison Rusinow. Washington, D.C.: Wilson Center Press. Burkett, John P. 1983. The Effects of Economic Reform in Yugoslavia. Berkeley: Institute of International Studies, University of California. Cohen, Lenard. 1982. "Balkan Consociationalism: Ethnic Representation and Ethnic Distance in Yugoslav Elites." Pp. 29-62 in At the Brink of War and Peace: The Tito-Stalin Split in a Historic Perspective, edited by Wayne S. Vucinich. New York: Brooklyn College Press. Connor, Walter. 1984. The National Question in Marxist-Leninist Theory and Strategy. Princeton, N.J.: Princeton University Press. d'Encausse, Helene Carrere. 1993. The End of the Soviet Empire, translated by Franklin Philip. New York: Basic. Denitch, Bogdan. 1976. Legitimation of a Revolution: The Yugoslav Case. New Haven, Conn.: Yale University Press. ---------. 1990. Limits and Possibilities. Minneapolis: University of Minnesota Press. Deutsch, Karl W. 1966. Nationalism and Social Communication. Cambridge, Mass.: MIT Press. Dobbs, Michael. 1991. "The Eruption of Eastern Europe." Washington Post (November 18-24), national weekly edition, pp. 6-9. Drew, Elizabeth. 1992. "Letter from Washington." New Yorker 68 (July 6): 70-75. Dutch, Raymond M., and James L. Gibson. 1992. "'Putting up with' Fascists in Western Europe: A Comparative Cross-Level Analysis of Political Tolerance." Western Political Quarterly 45 (1): 237-74. Enloe, Cynthia. 1973. Ethnic Conflict and Political Development. Boston: Little, Brown. Glenny, Misha. 1992. The Fall of Yugoslavia. New York: Penguin. Hannan, Michael T. 1979. "The Dynamics of Ethnic Boundaries in Modern States." Pp. 253-75 in National Development and the World System, edited by J. W. Meyer and Michael T. Hannan. Chicago: University of Chicago Press. Hechter, Michael. 1976. "Ethnicity and Industrialization: On the Proliferation of the Cultural Division of Labor." Ethnicity 3:214-24. 1556 Yugoslavia ---------. 1978. "Group Formation and the Cultural Division of Labor." American Journal of Sociology 84 (2):293-318. ---------. 1987. Principles of Group Solidarity. Berkeley and Los Angeles: University of California Press. Isaacs, Harold. 1975. "Basic Group Identity." Pp. 29-52 in Ethnicity: Theory and Experience, edited by Daniel Patrick Moynihan and Nathan Glazer. Cambridge, Mass.: Harvard University Press. Jackman, Robert. 1978. "The Predictability of Coups d'Etat." American Political Science Review 72:1262-75. Jenkins, J. Craig, and Augustine Kposowa. 1990. "Explaining Military Coups d'Etat: Black Africa, 1957-84." American Sociological Review 55:861-75. Kardelj, Edvard. 1960. Razvoj Slovenačkog Nacionalnog Pitanja. Belgrade: Kultura. Korpi, Walter. 1974. "Conflict, Power and Relative Deprivation." American Political Science Review 68:1569-78. Kourvetaris, George A. 1993. "Ethnic Conflict in the Balkans: Causes and Implications of the Disintegration of Yugoslavia." Paper presented at the American Sociological Association meeting, Miami. Kposowa, Augustine J., and J. Craig Jenkins. 1993. "The Structural Sources of Military Coups in Postcolonial Africa, 1957-1984." American Journal of Sociology 99 (1): 126-63. Lipset, Seymour Martin, and Stein Rokkan. 1967. "Cleavage Structures, Party Systems and Voter Alignments." Pp. 1-64 in Party Systems and Voter Alignments, edited by Seymour M. Lipset and Stein Rokkan. New York: Free Press. Lydall, Harold. 1989. Yugoslavia in Crisis. London: Clarendon Press. Magaš, Branka. 1989. "Yugoslavia: The Spectre of Balkanization." New Left Review 174:3-31. Massey, Garth, Randy Hodson, and Duško Sekulič. 1992. "Political Affiliation and Social Mobility in Socialist Yugoslavia." Pp. 223-58 in Research in Social Stratification and Mobility, vol. 11. Edited by Robert Althauser and Michael Wallace. Greenwich, Conn.: JAI Press. McKay, William F., and Albert Verdoodt, eds. 1975. The Multinational Society. Rowley, Mass.: Newbury House. Meadwell, Hudson. 1989. "Ethnic Nationalism and Collective Choice Theory." Comparative Political Studies 22 (July): 139-54. Nagel, Joane. 1986. "The Political Construction of Ethnicity." Pp. 93-112 in Competitive Ethnic Relations, edited by Susan Olzak and Joane Nagel. Orlando, Fla.: Academic Press. Nielsen, Frangois. 1980. "The Flemish Movement in Belgium after World War II: A Dynamic Analysis." American Sociological Review 45 (1): 76-94. ---------. 1985. "Toward a Theory of Ethnic Solidarity in Modern Societies." American Sociological Review 50 (2): 133-49. Olzak, Susan. 1983. "Contemporary Ethnic Mobilization." Pp. 355-74 in Annual Review of Sociology, vol. 9. Edited by Ralph H. Turner and James F. Short. Palo Alto, Calif.: Annual Reviews. ---------. 1992. The Dynamics of Ethnic Competition and Conflict. Stanford, Calif.: Stanford University Press. Olzak, Susan, and Joane Nagel. 1986. Introduction to Competitive Ethnic Relations, edited by Susan Olzak and Joane Nagel. Orlando, Fla.: Academic Press. Petrovič, Ruža. 1983. "The National Composition of the Population." Yugoslav Survey 24 (August 3): 21-34. Pfaff, William. 1992. "The Absence of Empire." New Yorker 68 (August 10): 59-69. Pinard, Maurice, and Richard Hamilton. 1978. "The Parti Quebecois Comes to Power: An Analysis of the 1976 Quebec Election." Canadian Journal of Political Science 11 (December): 739-75. 1557 American Journal of Sociology Ragin, Charles. 1979. "Ethnic Political Mobilization: The Welsh Case." American Sociological Review 44 (4): 619-35. Ramet, Pedro. 1982. "Self-Management, Titoism and the Apotheosis of Praxis." Pp. 169-94 in At the Brink of War and Peace: The Tito-Stalin Split in a Historic Perspective, edited by Wayne S. Vucinich. New York: Brooklyn College Press. ---------. 1984a. Nationalism and Federalism in Yugoslavia, 1963-1983. Bloomington: Indiana University Press. ---------. 19846. "Religion and Nationalism in Yugoslavia." Pp. 144-69 in Religion and Nationalism in Soviet and East European Politics, edited by Pedro Ramet. Durham, N.C.: Duke University Press. Rogowski, Ronald. 1985. "Causes and Varieties of Nationalism: A Rationalist Account." Pp. 87-108 in New Nationalism of the Developed West, edited by Edward A. Tiryakian and Ronald Rogowski. London: Allen & Unwin. Roosens, Eugene. 1989. Creating Ethnicity. Beverly Hills, Calif.: Sage. Rusinow, Dennison. 1977. The Yugoslav Experiment, 1948-1977. Berkeley: University of California Press. ---------. 1985. "Nationalities Policy and the 'National Question.'" Pp. 131-65 in Yugoslavia in the 1980s, edited by Pedro Ramet. Boulder, Colo.: Westview Press. Schermerhorn, R. A. 1970. Comparative Ethnic Relations. New York: Random House. Schrenk, Martin, Cyrus Ardalan, and Nawal A. el Tatawy. 1979. Yugoslavia: Self-management, Socialism, and the Challenges of Development. Baltimore: Johns Hopkins University Press. Sekulič, Duško, Garth Massey, and Randy Hodson. 1994. "Who Were the Yugoslavs? Failed Sources of a Common Identity in the Former Yugoslavia." American Sociological Review 59 (February): 83-97. Seton-Watson, Hugh. 1977. Nations and States. London: Methuen. Shoup, Paul. 1968. Communism and the Yugoslav National Question. New York: Columbia University Press. Smith, Anthony D. 1979. "Towards a Theory of Ethnic Separatism." Ethnic and Racial Studies 2 (January): 21-37. ---------. 1981. "War and Ethnicity: The Role of Warfare in the Formation, Self-images and Cohesion of Ethnic Communities." Ethnic and Racial Studies 4 (October): 375-97. Stanovčič, Vojislav. 1988. "History and Status of Ethnic Conflicts." Pp. 23-40 in Yugoslavia: A Fractured Federalism, edited by Dennison Rusinow. Washington, D.C.: Wilson Center Press. Štatistickí Godišnjak Jugoslavije (Statistical yearbook of Yugoslavia). 1987. Belgrade: Savezni Závod za Statistiku. Suny, Ronald. 1990. "Revenge of the Past: Socialism and Ethnic Conflict in Transcaucasia." New Left Review 184 (November-December): 5-34. Tudjman, Franjo. 1981. Nationalism in Comparative Perspective. New York: Columbia University Press. Tyson, Laura D Andrea. 1980. The Yugoslav Economic System and Its Performance in the 1970s. Berkeley: Institute of International Studies, University of California. Voirst, Milton. 1991. "The Yugoslav Idea." New Yorker 67 (March 18): 58-79. Warwick, Paul V., and Lenard J. Cohen. 1985. "The Institutional Management of Cultural Diversity: An Analysis of the Yugoslav Experience." Pp. 160-201 in Ethnic Groups and the State, edited by Paul Brass. Totowa, N.J.: Barnes & Noble. Young, Crawford. 1976. The Politics of Political Pluralism. Madison: University of Wisconsin Press. 1558