The Myth of “Ethnic Conflict”, Beverly Crawford and Ronnie D. Lipschutz (eds.)
A Note on Unicode Characters Contained in this Essay
EXPLAINING CULTURAL CONFLICT IN EX-YUGOSLAVIA: INSTITUTIONAL WEAKNESS, ECONOMIC CRISIS, AND IDENTITY POLITICS
What are the root causes of the war in the former Yugoslavia? Why did the six republics fail to peacefully separate from one another? Why was Yugoslavia unable to go the way of Czechoslovakia, with its “velvet divorce,” or the Soviet Union, with its relatively peaceful demise? Why was Yugoslavia unable to persist as a state, and why was its dissolution so violent? Why did virulent “ethnic conflict” emerge in the wake of a collapsed federal state?
A virtual cottage industry of analysis has sprung up to meet the demand for answers to these questions. Most accounts focus on explanations for the violent dissolution of the state. But few recent works have specifically addressed the last question: why “ethnic” conflict defined the adversaries and the character of the war in Croatia and Bosnia, as opposed to the regional or ideological divisions that could have potentially been exploited, and why “ethnic” conflict did not break out in other parts of Yugoslavia as the federal state dissolved.
In retrospect it is clear that the Yugoslav federal state was long headed for dissolution. A quick survey of book titles from the 1980s tells that story: Political Cohesion in a Fragile Mosaic (1983), Yugoslavia: A Fractured Federalism (1988), The Improbable Survivor (1988), Yugoslavia in Crisis (1989), Descent into Chaos: Yugoslavia’s Worsening Crisis (1989). Throughout the decade, analysts grew increasingly pessimistic about the future of a united Yugoslavia. But the most shocking and puzzling question was how neighbors who had lived together peacefully for years in Croatia and Bosnia could turn on each other so viciously as Yugoslavia disintegrated. Analysts rushed to explain the ferocity of the violence by calling on “ancient hatreds” and long-festering historical grievances. Some blamed Serbian aggression, others blamed Croatian and Muslim nationalism, and still others pointed fingers at international forces.
In this contribution I offer an alternative explanation. I argue that the roots of “ethnic conflict” in the former Yugoslavia can be found in the institutional structure of the Yugoslav political and economic systems constructed after World War II.1 While the postwar institutional structure offered numerous incentives for identity with an integrated Yugoslav state, as well as incentives for regional (as opposed to ethnic) political loyalty, it also encouraged interethnic rivalry through its institutions of allocation, representation, and participation. As the federal state weakened, that institutional structure offered increasing incentives to political entrepreneurs to “play the ethnic card” in a bid for political power. Regional politicians used their access to resources to build a power base among local, culturally distinct populations.
As long as the federal state remained strong, ideological and regional loyalties competed with ethnic loyalties as a source of political identity. Federal institutions could adjudicate disputes among regional elites and provide for peaceful conflict resolution and repression of exclusive ethnic nationalist politics. But ironically, in order to maintain authority by deflecting criticism for economic hardship and political discrimination, the federal government decentralized its control over both the economy and the political system. Each move toward decentralization was a move toward fragmentation and the consequent erosion of federal authority. With deepening fragmentation, local elites had more resources to distribute in exchange for support and saw fewer reasons to maintain loyalty to the central Yugoslav government.
After 1989 these local elites could have mobilized around ideological appeals—like they did in the Czech Republic. Or they could have called for regional rather than ethnic autonomy, like winning politicians did in Macedonia, or like elites in Tatarstan, Ajaria, and Dagestan in the wake of the Soviet collapse.2 Why did elites in Serbia, Croatia, and Bosnia shun ideological and regional appeals and decide to engage in vivid displays of cultural symbolism that aroused ethnic emotions and provoked images of ethnic discrimination and privilege? I shall argue that their decision was largely shaped by institutional incentives created by federal Yugoslavia throughout the postwar period. As they became more deeply rooted, these institutional incentives discouraged coalitions that would assure moderation on divisive issues.
These incentives were reinforced by new institutional rules of participation and representation designed to accommodate multiparty elections in 1990. The rules discouraged issue-based or ideological coalitions across republican boundaries and encouraged the exclusive politics of cultural identity. Initial successful displays of ethnic symbolism—often artificially contrived—drew attention to those ethnic divisions perpetuated by past institutional incentives. In particular, acts of civil disobedience and even violence vividly recalled past grievances and created new ones. Acts of civil disobedience and violence both increased public support for politicians who played the ethnic card and encouraged more violence.
The political entrepreneurs who campaigned on ethnic nationalist platforms and won elections were tempted to oppress the minority losers to maintain their reputations, credibility, and political power. Minorities then organized around their own ethnic and religious identity to oppose the winners, and as a result, the odds of violence increased. In this way, ethnic entrepreneurs were able to eclipse other political entrepreneurs who offered alternative futures for Yugoslavia, and the “bandwagoning” and “balancing effects” of identity politics were created, particularly among Serbs and Croats, the two largest ethnic groups in Yugoslavia.
The remainder of this contribution presents the evidence to support this argument. It begins with a discussion and critique of alternative explanations for the cultural conflict in Yugoslavia. This section is followed by a discussion of the theoretical considerations that support the institutional account presented here. Section three presents a more detailed description of the institutions of federal Yugoslavia that both discouraged and encouraged the practice of identity politics. This section argues that ethnic conflict was not determined by “ancient hatreds,” but was shaped by institutional incentives. The fourth section explains why those institutions that promoted identity politics and ethnic conflict trumped the others as the central state disintegrated and the economy fragmented throughout the 1960s and 1970s. The fifth section describes how alternative cleavages collapsed into reinforcing ethnic divisions. The sixth section explains why political entrepreneurs in Serbia, Croatia, and Bosnia decided to play the ethnic card and why ethnic politics escalated to violence. It also explains why elites in Macedonia made the decision to minimize identity politics in favor of regional autonomy, thus avoiding violent cultural conflict in the short run. The final section argues that Western states should pursue a policy of “getting the institutions right” in post-Yugoslav states in order to prevent future cultural conflicts there.
The institutional argument presented here runs counter to those made in the recent flood of literature on the violent dissolution of Yugoslavia. That literature can be divided into three rival intellectual camps. The first has been labeled the essentialist or primordial perspective. Primordial explanations stress the role of the “Balkan temperament” and “ancient hatreds” unleashed by the collapse of communism.3
Essentialist arguments are difficult to discredit because they are nonfalsifiable.4 They link conflict with irrational and “natural” psychological and social tendencies to “belong” to a group and to reject the “other.” Although they do not explain why the central focus of belonging needs to be an ethnic or religious group, essentialists argue that this tendency emerges when it is no longer repressed. The introduction to this volume offers a critique of the broader literature upon which such essentialist or primordial arguments are based. Here I would simply suggest that there is ample evidence in the Yugoslav case to cast doubt on these claims.
Indeed the early fateful decision to decentralize political and economic power that led to federal weakness was a response to ethnic tensions. Nonetheless, it is clear that those tensions were often muted and could have been further reduced. Throughout Yugoslav history, intraethnic cultural differences shaped by regional dissimilarities were often greater than cultural differences between ethnic groups.5 Marriages between people of different ethnic groups in Yugoslavia were on the rise during the decade before the war.6 After Tito’s death in 1980, the percentage of the population that identified itself as “Yugoslav” as opposed to an ethnically defined nationality (e.g. Serb, Croat, Muslim) also grew significantly.7 There is abundant anecdotal evidence to suggest that many ordinary people did not know or care about the ethnic identity of their neighbors before hostilities began. As late as 1989, the majority of Serbs favored a liberal future for Yugoslavia, the preservation of the federal state, and Yugoslavia’s integration into “Europe.”8 Ante Marković, the last Yugoslav prime minister and free-market reformer and a Croat, was the most popular politician in all six republics. And during the course of the war, Serbs living in Serbia exhibited decreasing ethnic solidarity with Serbs in Krajina and Bosnia. The postwar Serb-dominated parliament of Yugoslavia even passed a law that disqualified Serb refugees from Bosnia and Croatia from becoming citizens of Yugoslavia.9 Finally, in 1996, after years of bloodshed in Bosnia, local Serb residents in former Muslim-dominated areas agreed to peaceful meetings with Muslims on the return of refugees to their homes until regional officials protested, organizing violent attacks on returning refugees.10 Certainly ethnic identity was highly politicized in Yugoslavia, but the evidence suggests that these politicized identities were not fixed, and were indeed quite malleable.11
At the other extreme are explanations for Yugoslavia’s violent dissolution that view international forces as central causes. There is a large body of both historical and current literature that blames Balkan war and its particular “ethnic” content on great power attempts to carve up Balkan states for their own advantage.12 Indeed the post-World War I order in the Balkans exacerbated and created ethnic tensions with arbitrary borders separating many people from their homelands and from their “ethnic brethren.” But forty years of peace in the region and the peaceful transition of other Balkan states from communism suggest that domestic institutions can rectify international failures and mitigate cultural conflict.
Some recent international approaches suggest that in a post-cold-war world, where the stability of superpower rivalry has disappeared, power positions are more fluid, and uncertainty is high about the source of the next international conflict. In a multipolar world, states may feel unprotected from one another, both because power is more symmetrical and because they are unsure about their neighbor’s power capabilities. They therefore are likely to rush to protect themselves from real or imagined threats.13 Journalistic accounts of the dissolution of Yugoslavia grounded in these assumptions have focused responsibility for the war on Germany’s diplomatic recognition of Croatia. They explain Germany’s unilateral recognition by pointing to these forces. One such account, for example, suggests that a more powerful Germany in a multipolar world without the military protection of the United States perceived the need to drive south toward the warm waters of the Adriatic Sea to protect its own security. A smaller Yugoslavia with Croatia and Slovenia as allies could realize this geopolitical aim.14 The more sinister version of this claim was that given its new international power position, Germany was attempting to recreate its World War II alliance with an independent Croatia and impose a divide-and-conquer strategy in the Balkans to protect its interests and enhance its relative power in the region.15
These accounts of the recent conflict are both easily discredited and do not get to the heart of the central concern here: the causes of cultural conflict. Although Germany’s recognition of Croatia and its utter disregard for the new government’s violation of human rights against Serbs living there clearly hastened the dissolution of Yugoslavia, the war had begun six months earlier. Indeed, the specter of 1914—when great power rivalry in the Balkans ignited war in Europe—haunted the great powers in the 1990s and led them to cooperate in an attempt to end the war in order to avoid conflict among themselves. In this war, unlike the Balkan conflicts that ignited World War I, the great powers worked together to end it.16
A more nuanced international-level explanation for the dissolution of Yugoslavia is offered by Susan Woodward in Balkan Tragedy. Woodward argues that the institutional structures of the Yugoslav state provided a basis for prewar political stability in Yugoslavia. But Yugoslavia’s pattern of global integration and domestic economic reform shaped by international financial institutions in the 1980s undermined those institutions. Liberalization and global integration required the weakening of those very state structures that had provided political stability.
Woodward’s argument is compelling but incomplete and is disputed at times by her own evidence. While its analytic focus on international causes of the dissolution of the federal state provides an important perspective overlooked in both public and scholarly debates, it obscures both domestic institutional incentives for cultural conflict and the role of agency in igniting violence. Indeed Yugoslavia’s unique geostrategic position in the cold war led to early integration in the international economy, and Yugoslavia’s integration into the international financial system was deeper than that of most socialist countries. The impact on domestic politics and institutional structures was a crucial cause of the later Yugoslav collapse. Nonetheless, Woodward contradicts her argument by showing that IMF policies in the 1980s were designed to strengthen federal institutions and that those policies were overwhelmingly rejected by domestic political forces.
Furthermore, a comparative perspective suggests that key causal elements are missing from Woodward’s account. Two examples illustrate. Bulgaria was mired in international debt and saddled with conditionality requirements for repayment in the 1980s, and its government collapsed after 1989. But although it had an ethnically mixed population—with similarities to that of Bosnia—it did not experience cultural violence in the aftermath of debt and disintegration. Social conflict erupted in other countries with multiethnic populations in the face of IMF austerity programs, but that conflict has not always taken the form of cultural violence. Brazil provides the prime example. While Woodward does a masterful job of explaining the causes of the collapse of the federal Yugoslav state, her overarching explanation does not account for the eruption of cultural conflict.
A third explanation for the Yugoslav conflict suggests that the causes were
instrumental. This literature places blame for the war not on primordial
urges within society or great power pretensions within a changing international
structure, but rather on “political entrepreneurs” like Slobodan
Miloševi
Instrumental explanations should not be rejected out of hand. Wars always
require leaders, political entrepreneurs able to mobilize populations for the
support of their aggressive or defensive military efforts. But instrumental
accounts of the Yugoslav war beg three essential questions. First, they do not
explain why political entrepreneurs made the decision to play the ethnic card
and why they were effective in their bid to promote ethnicity as a
cleavage for political advantage in Serbia and Croatia. The preservation of a
multinational Yugoslavia was for the majority of Serbs a preferable alternative
to ethnic disintegration. Second, instrumental accounts do not explain why other
political entrepreneurs, drawing on alternative social cleavages such as class,
ideology, or simply region and territory, did not gain sufficient social support
to eclipse ethnic entrepreneurs in those two republics after 1989 and in Bosnia
in 1992. As noted above, Ante Markovi
This last point raises the third issue: instrumental accounts do not explain why some leaders in Yugoslavia achieved political success by promoting alternative political programs. Indeed they do not explain Tito’s earlier success as a Yugoslav political entrepreneur who effectively muted cultural conflict. Accounts of early repression and terror to achieve stability miss the point. Tito was enormously popular in the larger population, even in Serbia, and most Yugoslavs endorsed his idea of the Yugoslav melting pot. Similarly, instrumental accounts do not explain why, despite intense ethnic cleavages in Macedonia, politically successful elites there made a bid for regional independence rather than ethnic autonomy in 1992 and why Albanian and Macedonian politicians there were able to form a stable coalition against the ethnic nationalists. A politician’s decision to exploit ethnic divisions or refrain from exploiting them and his success or failure in that effort must also be explained.
The introduction to this volume has detailed the institutional perspective
that informs the argument I make here. Philip Roeder’s chapter provides a
valuable link between the broad institutional approach and the rise of ethnic
entrepreneurs in Soviet successor states. I show here how his argument can also
be successfully illustrated in the Yugoslav case. Like the former Soviet Union,
Yugoslavia’
Their decision to play the ethnic card is shaped by institutional structures of accountability. When regional officials are still accountable to the central federal government and depend on central support to sustain them in office, they are unlikely to make extremist ethnic appeals in a bid for local support. When, however, central authority weakens and they become accountable to a local constituency, they calculate their chances of winning support with alternative political appeals. When their constituency is multiethnic, they may enter into coalitions that mute exclusive ethnic appeals and make political demands and promises that would benefit the population of their local region as a whole. But they may fear the loss of significant support to political entrepreneurs calling for the autonomy of a particular ethnic group. As noted in the introduction, this is because the political entrepreneur is sure that he can get the support of the targeted group but is less certain of the support of the wider population. Despite the persuasiveness of this logic, this explanation is not entirely satisfying. Even with the support of an “ethnic machine,” a regional leader’s decision to play the ethnic card does not automatically result in an enthusiastic response from the targeted population; nor will it automatically result in successful political mobilization. But ethnic appeals in the republics of Serbia, Croatia, and Bosnia after 1989 resonated with the populations in these republics. Why?
Strategic interaction theories of mass political action and theories of behavioral cascades and bandwagoning described in the introduction promise a fruitful explanation. Scholarship on previous Yugoslav crises has noted this effect.20 Recall too that political institutions can either encourage bandwagoning effects or inhibit them. In the Yugoslav case, one analyst has argued, with the federal government fatally weakened and loyalty to the center diminished, there were no incentives for political entrepreneurs to lower the tone of their agitating and provocative discourse. Nor were there incentives for intellectuals to abhor the expression of provocative nationalist sentiments.21 Finally, bandwagoning is related to both timing and ethnic alliances as further causes of ethnic conflict. In Yugoslavia, once nationalists had gained the upper hand in Serbia, incentives for ethnic nationalism rose in Croatia, and “sister” Serb and Croat nationalist parties were formed in Bosnia. As nationalist parties were formed and won elections and as they gained strength by forming alliances across republican borders, they crowded out other alternatives and narrowed elite political choices. By the time Bosnian elections were held, non-nationalist alternative parties did not stand a chance.22
This approach further suggests that if all-Yugoslav elections had been held before republican elections, incentives to appeal to a wider population would have been higher and political parties would have been more inclusive; ethnic bandwagons would have been slower to fill. In fact, nationalist politicians insisted on holding republican elections first; bandwagoning effects then precluded the possibility of elections at the national level. If the legacy of ethnofederalism is taken into account, however, by 1991 national elections—even following republican elections—would not have prevented the escalation to violence; people had no party organization to represent their interests outside of their own republic, and alliances between non-nationalist parties were precluded by spirals of mistrust brought on by early bandwagoning.
In sum, as we shall see in this case, postwar institutions of ethnofederalism cemented the logic of identity politics in the Yugoslav federal structure. That logic did not always dominate, particularly in the population at large, and it was more pervasive in some areas than in others. Nonetheless, as the federal system weakened and local officials became increasingly accountable to local ethnic constituencies rather than the central government, the logic of identity politics strengthened throughout the multiethnic Yugoslav republics. Institutional legacies favored the creation of ethnic nationalist parties and interethnic political rivalry; bandwagoning and balancing effects spread in Serbia, Croatia, and Bosnia. The weakness of those legacies in Macedonia accounts for the more conciliatory path taken there.
The central premise upon which this volume is based is that authoritative political institutions channel social conflict in the direction that institutional planners prefer; thus they determine the logic that will dominate political competition and provide the basis for political mobilization. Institutions do not treat all forms of conflict impartially; they constrain some forms of competition and mobilization and encourage others.23 They provide differential access to key resources, strengthening some actors and weakening others. In doing so, they shape political preferences and identities of both elites and publics. Institutional incentives, embedded in rules of accountability, representation, participation, and resource distribution can structure political struggle in ways that either moderate or encourage ethnic and sectarian political conflict.
After World War I, political elites in both Serbia and Croatia attempted to moderate interethnic conflict by imposing a system of pluralist political competition and an integrationist logic on the new Yugoslav state. They did so by constructing a unitary rather than a federal state system of representation and participation. That unitary state, they believed, would be based on a shared southern Slav identity and a common bond forged by the humiliation suffered at the hands of both the Ottoman and Hapsburg rulers.
Indeed some observers argue that many Croatian and Slovene elites (those who came from the bourgeoisie and the intelligentsia) had joined the effort to form a Yugoslav state because they saw political advantage in participating in the governance of a unitary state over receiving minority status and enduring the restrictive franchise in Austria, Hungary, or Italy. The 1921 constitution of the newly created state of Yugoslavia was a relatively liberal one, enshrining universal male suffrage and equal civil and political rights for all Yugoslav citizens.24
Not all Yugoslav elites, however, were happy with this new arrangement. Many Croatian nationalists felt that they had freed themselves from Hapsburg domination only to be newly saddled with Serbian hegemony in a unitary state. They mistrusted the new constitution, arguing that it masked Serbian control over Croatia and that a national Yugoslav identity could not be created under a Serbian king, his army, administration, and Orthodox religion. Indeed, argued such Croats, this “nation” really represented the submission of a Roman Catholic people on the periphery of civilized Europe to an inferior, Oriental culture.25
Threats of Croatian secession, and the fact that large sections of the Croatian population did not accept the constitutional basis of the Yugoslav state combined with the increasing centralization of power in Serbia to prevent the formation of interethnic political coalitions in representative institutions. Divisions were exacerbated when parliament—the forum where a clash of interests was aired—shut down in 1929. Debate ended, and a fraction of the Croat elite turned to violence. Nonetheless, these elite power plays did not trigger widespread interethnic conflict in the population as a whole. Indeed that conflict was nurtured by the breakdown of the state and by the war raging in the region after 1939. Croatian elites broke from Yugoslavia to ally with the Nazis and quickly carried out a German-led plan to massacre thousands of Serbs. The decision of the Bosnian Muslim elites to throw in their lot with Croatia transformed them into the enemy of the Serbs as well. With the German defeat in 1945, Croatia surrendered, and Serbia took its revenge by killing thousands of Croat and Muslim prisoners. This violence created a vast reservoir of culturally defined grievances that would shape the construction of postwar institutions.
FEDERAL INSTITUTIONS OF POSTWAR YUGOSLAVIA: BALANCING “NATIONAL” INTERESTS
With memories of mutual massacres still vivid, Tito believed that national integration was not possible in a unitary Yugoslav state. He thus established a federal system of ethnic republics after the war that would provide guarantees of national equality. Like any federation, authority was distributed between the central government and the governments of the constituent units, and the distribution of authority could not be changed without mutual consent. The Yugoslav federation held to three broad principles of federalism. First, within their respective spheres of operation, both the central government and the constituent units were independent, and neither was subordinated to the other. Second, the constituent units participated in the making of decisions at the federal level. Finally, important federal decisions required equal representation of all of the constituent units, regardless of their size and population.
Yugoslavia, however, was not a centralized federal system like that of the United States, the Federal Republic of Germany, or, in its most centralized form, the Soviet Union. Indeed it did not resemble most other federations, in which the central government could make many decisions without consulting the member governments of the constituent units. Instead because Yugoslavia was so divided as a result of the events of World War II, Tito created a noncentralized federalism in which the constituent units exercised a large degree of control and authority. Although the 1946 constitution placed all mineral wealth, power resources, means of communication, and foreign trade under state control, it also stipulated that the central government could make decisions in only a few narrowly restricted issue areas without obtaining the approval of the governments of the constituent units.26
Further, like the Soviet Union and Czechoslovakia, Yugoslavia was now governed by the institutions of ethnofederalism, which were intended to transform ethnically based political identities into cultural/administrative identities and thereby prevent the reemergence of extreme identity politics as a dominant political force.27 As Vesna Pešić argues, two kinds of national groupings were organized hierarchically in the constitution. Five culturally defined groups—Serbs, Slovenes, Croats, Macedonians, and Montenegrins were territorially organized in constituent republics in which, as the titular nationality, they held the status of “constituent nation.” The 1971 census recognized Muslims as a separate nation, and in 1971 Bosnia-Herzegovina was recognized under the national principle as a republic, consisting of three constitutive peoples: Serbs, Croats, and Muslims. Those not members of these six “nations”—e.g., Jews, Czechs, Romanians, Russians, Bulgarians, Romany, Vlachs, Albanians, and Hungarians—were called “national minorities” and later “nationalities.” These groups initially suffered from lower representative status than the constituent nations.28 Susan Bridge argues that the structure of formal political representation throughout the postwar period discouraged minority participation and representation through the single-member district in both party and government. But the single-member district worked to the advantage of minorities in two defined regions where the “nationality” was a majority of the population. As we shall see, after the constitutional changes of 1974, Kosovo, with a majority Albanian population, and Vojvodina, with a majority Hungarian population, gained increasing autonomy throughout the postwar period and enjoyed equal participation at the federal level with the same representative status as the constituent nations.29
A dual notion of federalism was embodied in representative institutions that in some ways resembled a traditional parliamentary democracy. The Federal Assembly was composed of a Federal Council, elected by citizens voting as Yugoslavs, and a Chamber of Nationalities, in which citizens were represented as nations and nationalities. This federal structure was intended to balance the interests of all the peoples of Yugoslavia. The importance of the equality of the constituent nations in representative institutions cannot be overstated. Indeed a territorially based federation was not considered fully adequate to provide an equal representation of Yugoslavia’s constituent “peoples” since most territorial units, even those with titular nationalities, had mixed populations. Therefore, territorial ethnofederalism was reinforced by a system of ethnic quotas or “keys” as a central principle for the allocation of political resources. All appointments to public office (including the military) were decided by a formula for the proportional representation, or in some cases equal representation, of individuals by constituent nation or nationality. The effort to maintain balance in public institutions went far beyond the intent of the quota system. For example, in an attempt to maintain balance even in the prosecution of politically motivated nationalist activities, central government authorities often went out of their way to balance a particular prosecution with charges against people from other ethnic groups.30
Tito established these institutions of ethnofederalism because he believed that if the resolution of disputes between national groups appeared to favor one group over the others, the federation’s internal balance would be upset and Yugoslavia would be destabilized. His goal was to preserve the central Yugoslav state. Given Serbia’s disproportionately large population and history as an independent state and given Croatian elites’ historic distrust of Serbs, this was not an easy task. Indeed some analysts argue that the constitution implied an unwritten agreement between Tito and Serbian political elites in which it would espouse Yugoslav unity and equality of representation in order to mitigate Croatian fears of Serb dominance in the state apparatus and thus cement Croatia’s loyalty to the center. One often heard the slogan, “Weak Serbia, strong Yugoslavia.”31
Soon the institutions of ethnofederalism would come to dominate all others in decisions of allocation, participation, representation, and accountability. At the outset, however, other powerful institutions were constructed to encourage solidarity and integration into the federal Yugoslav state. The two most prominent were the Communist Party and the army, supported by both socialist ideology and a system of privileges conferred upon those who had demonstrated loyalty to the central state. Successive constitutional and economic reforms created incentives for interregional competition among political elites over the means of economic development. These reforms also created socioeconomic divisions in the larger population that transcended ethnic cleavages, and they both encouraged and codified ideological conflict between conservatives and reformers in the party.
These divisions were potentially cross-cutting. That is, different ethnic groups who were part of the same socioeconomic class had more in common with each other than with their ethnic compatriots; different ethnic groups living in the same region potentially had common regional interests that transcended ethnic divisions; both ideological consensus and division could potentially overcome regional and cultural differences. Thus, as we shall see below, a plethora of institutions within the federal system held the potential to mitigate the importance of cultural divisions that had long plagued Yugoslavia. It is to a brief description of the alternative social cleavages created by these institutions that the discussion now turns.
INCENTIVES FOR INTEGRATION AND YUGOSLAV SOLIDARITY
Integrationist logic was initially encouraged by ideology and repression. It was further encouraged by partisan privilege as an important (though certainly not exclusive) allocative principle. First, Tito believed that the dominance of Communist ideology would reduce the salience of ethnicity as a source of political identity and replace it with a more cosmopolitan socialist one. Tito believed that the division of Yugoslavia into separate republics would be temporary and that once Marxist ideology became embedded in social practice, Yugoslavia could become an integrated, even unitary state.32
Initially repression bolstered this belief in the power of ideology; public debate on ethnic issues was largely forbidden, and, as in other Marxist regimes, the representation of grievances on the part of particular ethnic groups had to be articulated in economic and social terms since these were the only terms viewed by the state as legitimate.33 Policies suppressing religion and nationalist movements were designed to stifle interethnic competition and sectarian privilege in the interest of an integrated multinational state.34 These repressive measures were largely abandoned in the 1960s;35 nonetheless, the political expression of nationalism remained illegal and was prosecuted. Successive constitutions prohibited the propagation or practicing of national inequality and any incitement of national, racial, or religious hatred and intolerance.36
Two integrationist organizations dominated Yugoslavia’s political structure: the League of Communists of Yugoslavia (LCY) and the Yugoslav National Army (JNA). The LCY began as a highly centralized but carefully multinational institution which was quickly transformed after the break with the Soviet Union from an elite cadre to a mass organization. As party membership grew, it provided the only forum for political debate about alternative social interests and preferences. In fact, this horizontal political cleavage dominated all others for many years.37 It differed substantially from the other Communist parties of the Soviet Union and Eastern Europe in that central to its ideological core was the belief that its leading role in society would disappear with the development of democracy in Yugoslavia.
The JNA was organized to encourage integration and loyalty to the federal center through its rules of accountability, participation, and representation. It had a constitutional obligation to maintain the territorial integrity of Yugoslavia in the face of both internal and external threats, and it was beholden to the authority of the federal presidency. All army units were composed of a mix of officers and recruits drawn from throughout Yugoslavia. Its party organization was ethnically heterogeneous, and the ethnic quota system dominated the officer corps.38 The army’s political influence was substantial; it had a vote in the federal party presidency equal to the republics.
Finally, elite solidarity across national lines was initially encouraged by allocative policies that privileged partisans from all national groups who had fought against fascism. Pro-Partisan Serbs from predominantly pro-Chetnik areas, pro-Partisan Croatians and Serbs in Croatia, Partisan Muslims, Macedonians, Montenegrins, and Slovenes held top posts in the Communist Party and in government at all levels. They created local dynasties and operated politically much like a powerful lobbying group to put pressure on the central government to pursue integrationist policies and provide generous material support to their local communes and regions. These former Partisans and their families were united by ideological preferences rather than ethnic bonds; they too assumed that a Yugoslav identity would come to replace other national political identities.39
There are many indicators that these institutions were partially successful
in moving Yugoslavia toward integration. I listed several at the outset: a
rising Yugoslav national identity, especially among young people, the strong
preference among Serbs for the preservation of the federal state, and the
widespread political popularity of Ante Markovi
SOCIOECONOMIC CLEAVAGES
Incentives for integration were accompanied by both the intentional creation of channels for the political representation of various producer groups in the Federal Assembly and the unintended institutional creation of new socioeconomic cleavages in the population at large through state plans for resource allocation. In 1953 a new constitutional law attempted to strengthen the political power of producer groups composed of all nations and nationalities. Planners believed that this change would weaken representative divisions among ethnic groups. According to this law, of the two chambers in the Federal Assembly, only one-half of one would now be elected according to the nationality principle, while the other half would continue to be elected by the people at large. The second chamber would be elected by workers in “socially owned” enterprises. Professional workers, individual peasants, and private entrepreneurs were not represented. The law thus embodied the belief that political differences would be along class rather than ethnic lines.42
Workers were also given a strong participatory role in the economy. Beginning in the 1960s, the concept of self-management was introduced as the operational principle of economic management and allocation. Instead of state ownership, there would be social ownership of enterprises, governed by strong workers’ councils and powerful oversight committees.43 Behind the self-management principle was the belief that associations of workers had the right to participate in budgetary and managerial decisions at the workplace.
The 1965 economic reform increased enterprise autonomy further by removing regulatory burdens imposed by the central state: depreciation rates were increased and capital tax on fixed assets was cut. The reform also encouraged inter-republic enterprise relationships: enterprises were permitted to lend to other firms across republican and provincial borders directly and participate in joint ventures with them. In short, the implementation of the self-management principle, bolstered by the 1965 reform, would heighten producer political participation and thus further dilute ethnic divisions. In doing so, it would structure social competition along economic rather than along ethnic lines and thus increase loyalty to the federal state. Economic interests, however, were mediated by territorially based representative institutions; producer representatives were grouped first by commune and then by republic.44
Although equally represented and integrated into the management and decision-making structure of the economy, not all producer groups from all regions were treated equally when it came to allocative decisions directed from the center. In their attempt to modernize Yugoslavia, economic planners in the federal government favored some producers over others by giving some groups privileged access to material goods and services and denying it to others. Hardest hit by the new policies were the rural peasants. Yugoslavia possessed fertile agricultural land populated by an entrenched peasantry. The production of primary commodities had long been the mainstay of economic activity in the region. But the central economic goal of the postwar federal government was to transform the entire country from a backward agrarian nation into a modern industrial one. The federal government therefore poured its resources into industrial investment at the expense of the agricultural sector, thus driving peasants off the land and into the factories. Peasants who remained on the land lacked pension benefits and had little access to state housing, while they faced state-mandated prices for their produce and were provided with almost no investment capital.45
This policy created an important cleavage between the industrial workers and public-sector employees who had migrated to the cities, on the one hand, and those who remained on the land, on the other. Furthermore, the fastest growing economic group in postwar Yugoslavia was the white collar sector, reflecting rapid growth in the state bureaucracy at all levels of government. As long as the economy grew, the differences, particularly between white collar and blue collar workers, were resolved within both the system of self-management and the Federal Assembly. By the 1980s, however, when economic conditions took on crisis proportions, Susan Woodward argues, “The primary social divisions . . . in Yugoslav society were not defined by ethnicity but by job status. . . . In terms of how people saw themselves, ethnicity was less important than either occupation and the social status it conveyed or place of residence—urban or rural—and its related culture.”46 Whether Woodward is correct or not, there is ample evidence to suggest that economic cleavages in society (defined largely in terms of urban-rural splits) and socioeconomic identities were as strong if not stronger than ethnic cleavages and identities.
IDEOLOGICAL CONFLICTS
The initial ideological cleavages among political elites of the 1950s and 1960s were analyzed by Western observers as the usual division between ideological conservatives and liberal reformers. During the late 1960s reformers dominated politics at the federal center. They were confident that given the chance, the general population would endorse their liberal policies. They were particularly confident in their own electoral success over the conservatives because a middle class was growing in Yugoslavia that would be particularly receptive to their appeals. They thus resolved to further democratize the electoral process in order to reduce the presence of conservative opponents in the legislative branches of government at all levels. To the extent that reformers gained a foothold in political competition, the entrenched political establishment was pressured to widen the franchise. Much like democratizing Western Europe in the nineteenth century, groups within the establishment who expected to benefit from reform introduced more democratic procedures as a strategy to weaken powerful conservative opponents.47 Just before the 1967 elections, a wave of constitutional amendments increased the power of elected bodies at the federal level, particularly the Federal Assembly. And at all levels of government, elections were more hotly contested than ever before. Of course political rivalry was limited to intraparty ideological competition, and no thought was given to the creation of a multiparty electoral system. Liberalizers focused their political platform on the strengthening of market forces, freedom of speech, a merit-based system of promotions, and the withdrawal of the party from the arts and culture.
According to many observers, this path ended abruptly when some of the liberalizers in Croatia attempted to increase their popular support base against their conservative opponents by allying with nationalist political elements, who had become bolder and more openly critical with every move toward political liberalization. This coalition was a clear departure from the traditional coalition between reformers and conservatives against ethnic nationalists.48 As I discuss in more detail below, Tito crushed this alliance, and with the backing of the JNA, he purged the party of its nationalist and politically liberal elements and left centralizers firmly in power. Those centralizers, loyal to the federal state, acted quickly to suppress nationalist movements, and they were not to emerge openly again until the late 1980s.
According to Denison Rusinow and Steven Burg, with nationalist forces excluded from political participation, the early 1980s witnessed the formation of political cleavages along three separate ideological lines: the confederationists, the ideological conservatives, and the liberal reformers.49 Confederationists, found primarily in the leaderships of Slovenia, Croatia, and Vojvodina, wanted to expand their own political autonomy and economic power at the expense of the central government in Belgrade. The ideological conservatives, found primarily among the partisans, the JNA, and the poorer republics, argued against the establishment of a market economy and were sympathetic to a more centralized and egalitarian polity. The liberal reformers, found mostly in the Serbian party, defended the introduction of a market economy in Yugoslavia and argued for a centralized foreign currency market and the elimination of interregional economic barriers.
These ideological divisions among elites were deep, and they spread to the public at large. Indeed the evidence of strong liberal leanings in Serbia suggests that had ethnofederalism not been so entrenched and had the federal state and party organization not been so weak, Yugoslavia would have undergone a transition from communism similar to that of other Balkan and East European states. The primary elite division would have been between reformers and conservatives; international financial institutions would have bolstered the political power of the reformers, supporting the institutions of the federal center, and even conservatives in power would have been pressed to follow their mandates.50 Because federal institutions provided incentives for regional divisions, however, those divisions became more deeply rooted than either ideological or socioeconomic cleavages. It is to the issue of regional divisions that the discussion now turns.
CENTER-REGION AND INTERREGIONAL COMPETITION FOR RESOURCES
The twin economic goals of Yugoslav development policy had always been to reduce the disparities in regional living standards while maintaining a high rate of growth. As noted above, central economic planners believed that the overriding economic goal was to transform Yugoslavia into a modern industrialized nation. Priority was given to the development of industries that would contribute to the rapid growth of the country as a whole. But the individual republics had attained different levels of development, and as a strategy to encourage national Yugoslav integration, every five-year plan mandated the “equalization of conditions” between the developed and less developed regions.51
Resource limitations, however, produced deep tensions among the republics over the pursuit of these twin goals. Slovenia and Croatia, as the rich republics, preferred that funds be allocated by efficiency criteria; because they were most efficient, they would receive the bulk of the investment funds, and they resisted the transfer of resources to the poor. The poorer republics of Serbia, Bosnia, Macedonia, and Montenegro fought for funds as development subsidies. Because all funds were administered from a central General Investment Fund and because the central government regulated industrial development, divisions among the republics over investment took the form of center-region controversies. The poorer republics were dependent on the center for development funds, and the richer republics wanted autonomy from the center to free them from subsidies and regulation.
In addition to these conflicting pressures on central funds from the developed and less developed regions, partisan elites pressed for regional credit allocations based on political criteria rewarding partisan loyalty and punishing those who had opposed the partisans in the war.52 Indeed behind the scenes, regional politicians used political arguments to counter efficiency and development arguments, and they used patronage networks to lobby federal officials for regional investments; those investments would create jobs and income at home which in turn would bolster the regional power base of local politicians.53
These conflicting goals and criteria for the allocation of limited material
resources led to mutual resentments. No matter which criteria were used, the
republics who did not receive investment funds felt cheated. Three examples
illustrate. First, all investment funds to stimulate growth were distributed
directly to specific individual enterprises throughout Yugoslavia. These funds
were initially allocated through a central investment bank; after 1965 they were
allocated through a network of regional banks. Individual enterprises competed
for these funds on the basis of interest rates and repayment schedules through a
series of auctions. Enterprises in Croatia and Slovenia were always more
competitive according to the criteria. Furthermore, since priority was given to
industries that would contribute to rapid national growth, five-year plans
mandated that raw material inputs would be underpriced to make such industries
competitive. Again, priority industries were located in the developed regions of
Croatia and Slovenia; most of the raw materials came from Serbia, Montenegro,
Bosnia, and Macedonia. The predominance of low-priced raw materials in the less
developed regions and the absence of offsetting transfers meant that the poorer
regions were penalized for providing the raw materials necessary to
industrialize the country. The result, as one analyst put it, was that “both per
capita investments and the grants-in-aid allocated to the less developed regions
were consistently less than those allocated to the developed regions.” Serbian
political leaders complained about the huge transfers of industry from Serbia to
Croatia and Slovenia that had taken place between 1945 and 1951 in the name of
efficiency. And the introduction of market socialism in 1963, shifting economic
decision-making from local party and government elites to workers’
Second, as noted above, the wealthier regions complained bitterly about the huge income transfers required for the development of the poorer republics; they were particularly bitter about the forfeiture of hard currency earnings to federal treasuries. Croatian leaders complained that despite the fact that Croatia brought in half of all foreign capital as of 1969, it was allocated only about 15 percent of the total credits, an amount insufficient for its level of development. They further argued that Croatia produced most in foreign currency earnings and enterprise profits and received much less through the redistribution process. In 1971 the president of the Assembly of Croatia stated that Croatia would have to renegotiate the size of its contribution or abandon many of its own public works programs.55
Finally, in the eyes of the richer republican elites, political criteria seemed to work to the advantage of the poor republics. Croatian elites pointed out that Montenegro received the most investment per capita, while Croatia, the second most developed republic able to maximize output per investment, received what Croatian leaders believed to be barely its share. This perception was formed because policies providing partisan preferences seemed to make it clear that Croatia was being punished for its record in World War II, and regions with strong partisan groups were being rewarded. Critics in Croatia and Slovenia pointed out that Montenegro and Serbia received disproportionate investment credits, largely due to the strong patronage networks of partisans in these regions.
The pursuit of conflicting goals and the overlapping institutional structure by which investment funds were distributed from the center to the republics had two important cross-cutting consequences for the locus of political cleavages and loyalties. First, as we shall see below, the accumulated allocative disputes and resentments had an important impact on constitutional debates; those debates in turn exacerbated interregional conflicts as their resolution weakened the center. Heavily, directly, and transparently dependent on the central government, the poorer regions developed the most loyalty to the center and reinforced integrationist tendencies in those regions.56 The richer republics sought increasing decentralization of the federal system and more autonomy in economic decision-making.
A second and equally important consequence was the development of regional and republican loyalty over loyalty to the center. The institutions of economic allocation emphasized territorial over functional organization of the economy, and economic resources were distributed from the central government to the republics and regions rather than directly to individual enterprises. Thus republican loyalties replaced central loyalties, even in the poorer republics, because republican elites were responsible for procuring funds for their particular republics.
This last point is central to the argument: Because of the allocative structure, regional loyalties and divisions were created that were not necessarily congruent with “national” or “nationality” divisions. Two examples of regional loyalty (neither of which actually proves a causal linkage to allocative principles) illustrate. The first ironically can be found in the bitter complaints of the 1986 Memorandum written by a committee of Serb nationalist intellectuals in the Serbian Academy of Sciences. The Memorandum protested Yugoslav “assimilationist” policies, arguing that they were turning Serbs living in Croatia into Croats, and it claimed that ethnic Serb writers in Montenegro and Bosnia were writing “Montenegran” or “Bosnian” literature instead of Serb literature. And they complained that Macedonian Communists had simply Macedonized Serbs.57 With these complaints, the Memorandum suggested that republican political identities had indeed replaced ethnic political identities.
Second, even in war, Bosnian Serbs loyal to Bosnia joined in the Bosnian government and army to oppose Bosnian Serbs who fought the government. In the first elections there after 1990, 13 Serbs were elected to the assembly who were not members of the nationalist party. Indeed ethnic cleavages were not the only—nor were they necessarily the deepest—cleavages in federal Yugoslavia. As we shall see below, territorial loyalties increasingly overlapped with ethnic loyalties as the Yugoslav center collapsed. But territorial divisions unrelated to national divisions were real and they persisted even in war.
ALTERNATIVE CLEAVAGES AND ALTERNATIVE SCENARIOS FOR FEDERAL YUGOSLAVIA
In sum, in Yugoslavia, as in all federal systems, the preservation of the state depended on the strength of institutions that encouraged loyalty to the central government. It further depended on the subordination of cultural cleavages, ideological disputes, socioeconomic divisions, and center-region conflict to central government authority. As the narrative suggests, a complex pattern of political cleavage had evolved within both the party and government structures. Indeed this was evident by the 1960s.58 If institutional strength at the federal level had been maintained, political, socioeconomic, and regional divisions could have diffused social conflict, and Yugoslavia might have been preserved. But as we shall see below, the strength of federal institutions was slowly depleted and potential cross-cutting cleavages dissolved into ethnic divisions as ethnic representation was strengthened and as the mercantilist policies of the republics drained economic resources. The center would not hold.
If federal Yugoslavia could not be preserved, its peaceful dissolution
required that center-regional conflicts be resolved in favor of regional
autonomy. As we shall see below, Yugoslavia was indeed moving in this latter
direction. By 1974 the individual republics were close to becoming fully
sovereign states. And as late as the summer of 1991, Bosnian President Alia
Izetbegovi
The necessary condition for the dominance of cultural conflict in Yugoslavia was the entrenchment and expansion of ethnofederalism. Ethnofederalism politicized cultural identity, bolstered the power base of local elites, and thus deepened cultural divisions. The sufficient condition was the federal center’s decline in power and authority and the resulting economic decline. Economic decline and periodic crises triggered conflict over resources along the regional and ethnic lines that ethnofederalism had created. Constitutional changes and economic reforms throughout the life of Yugoslavia ensured that both of these conditions were met. A brief sketch of those changes reveals the resulting rising importance of ethnic divisions over purely territorial ones.
Distributive quarrels among republican elites in the Communist Party emerged from the outset. These quarrels were rarely played out between republics directly but were always directed toward the central government. Indeed there was no forum where the republics and provinces could negotiate directly with each other.60 As suggested above, the deepest discontent was in Croatia and Slovenia, the richest republics and the two most unhappy with central administrative controls. Fearful that this discontent would lead to calls for autonomy, aggravate ethnic conflict—particularly in mixed economic regions—and thus thwart the drive for Yugoslav integration, Tito attempted to deflect criticism and undermine autonomy demands by decentralizing most political and economic activity. The LCY thus undertook measures that conferred increasing political authority on the individual republics in the belief that more autonomy within the federal state would undermine divisive nationalisms. Indeed by the time of Tito’s death, there were very few funds, favors, or power resources left in the center to distribute; republican quarrels over central resources were fierce but bore little return for the winners.61 But as we shall see below, the decline in central powers led to a disintegration of the Yugoslav market and thus a decline in economic efficiency and growth. As the economic pie got smaller, competition over resources increased and national resentments deepened even further. As Rusinow argues, by the 1970s the disintegration of the Yugoslav market into “eight mercantilist and protectionist regional fiefdoms” exacerbated the tendency on the part of regional political elites to regard the republics as separate political and national communities.62
CONSTITUTIONAL REVISIONS AND ECONOMIC REFORMS
Successive constitutional revisions and economic reforms both codified and induced decentralization and representation by nationality or ethnic group. The 1953 constitutional amendment reduced the administrative role of the central state to five areas: foreign affairs, defense, internal security, and state administration. Although the “national economy” remained in the hands of the federal government, republics were given their own budgets over which they exercised independent control. In the search for a more impersonal allocative mechanism that would deflect criticism from the federal government, central authorities introduced administrative market socialism. Fixed wages were abolished, centralized planning mechanisms were weakened, and financial instruments replaced administrative rules in macroeconomic coordination. The republican governing bodies were made legally independent of the federal government.63
By 1963, however, economic conditions began to worsen. Recall that the industrialization drive had pushed peasants off the land and into the factories in urban areas. Wage earners pouring into the cities exerted pressure on demand for consumer goods that were in short supply. Expanding demand forced accelerated imports, and the balance of payments deficit dramatically increased. The IMF was called in to provide a structural adjustment loan; its conditionality requirement was that Yugoslavia further liberalize its economy. But as exporters, Croatia and Slovenia were suffering from the 1961 recession in Western Europe, and their exports sagged dangerously.
Squeezed both externally and internally, political elites in both republics continued to see the federal government as the target of their discontent. They saw federal fiscal policies intended to equalize levels of development among the republics as a transfer tax that would disproportionately benefit the less developed republics. Most dangerously, nationality came to be associated with locality and caused divisions within the party that cut across those between reformists and conservatives.64
Again to deflect attacks against the center as the cause of these economic problems and to weaken discontent, a new constitution was formulated in 1963. It further decentralized the economy. This time, however, decision-making authority and autonomy devolved to the communal level. According to some observers, it seems that the aim was to localize power without further diluting the authority of the central state since the federal government still had direct links to the communes without going through the republican assemblies.65
At the same time, the constitution expanded the meaning and practice of “ethnic balancing” in an attempt to further solidify loyalty to the center. The Chamber of Nationalities was given more legislative power. Its primary task was now to discuss and approve legislation of the assembly on an equal basis with the economic, education and culture, welfare and health, and organizational-political chambers which had developed out of the Council of Producers.66 The Chamber of Nationalities was upgraded again in 1967. Elected by republican and provincial assemblies, it now replaced the federal chamber as the senior of the five chambers, and the federal chamber was sharply downgraded.67
This organizational change had an important impact on the locus of political accountability and thus on the locus of political loyalty. Previously most members of the Federal Assembly had been responsible only to individual communal assemblies or to the voters at large in given electoral districts. With the upgrading of the Chamber of Nationalities, however, most members were now accountable to their republican and provincial assemblies. Because no federal legislation on any subject could be passed without the consent of the members of the Chamber of Nationalities and because members of that chamber were accountable to their regional assemblies, the republics sharply increased their power at the federal legislative level. In the 1971 census, the status of “nation” was conferred upon Yugoslav Muslims.68 Muslims could therefore be represented in the Chamber of Nationalities.
The strengthening of the Chamber of Nationalities not only deepened ethnic and regional political power, but it also exacerbated ethnic tensions. Representation in the Chamber of Nationalities was accorded by the principles of equality and proportionality. Twenty delegates to the Chamber of Nationalities were chosen by each of the six republics, regardless of population. This “balancing” effort bred resentment along ethnic and national lines. Serbs, for example, had 40 percent of the total population but had only 14 percent of the votes in the Chamber of Nationalities, while Slovenes represented 8.5 percent of the total population and had the same percentage of votes.69 Equality in representation benefiting the richer and smaller republics was intended to offset the disproportionate economic burden placed on them by regional development policy. But the policy backfired because it channeled resentments and privileges away from territorially defined republics and directed them toward specific ethnic groups who dominated those republics. Serbs began to resent Croats, not just Croatia, for both their wealth and its representative weight in the powerful Chamber of Nationalities. That representative weight, Serb elites believed, eschewed economic allocation to benefit Croats and was unfair to Serbs. Croats, in turn, resented the disproportionate representative weight of smaller nations which provided them with political clout to push for redistributive policies that would drain economic resources from richer republics like Croatia.
Ethnic tensions thus colored debates about regional development policy. Despite efforts to equalize levels of development among all of the republics, the gap between the rich and poor republics grew; elites in the richer republics saw little reason to continue to transfer resources to the poorer ones.70 They thus argued for allocation of investment funds based on profitability and efficiency. The poorer republics defended their position that they needed continued transfers in order to grow.
The result was a stalemate; there was no agreement on how investment funds should be allocated in the future. This immobilism at the center, combined with the growing deficit and pressure from the IMF for further liberalization and the representative weight of the rich but small republics in the Chamber of Nationalities, led to the economic reform of 1965. The reform itself suggests a triumph for the richer republics and the decentralizers as the center was further weakened; in the course of the debate, the central government was removed from its role as the provider of investment funds to the republics and a network of republic-level banks was created.71 They were authorized to take primary responsibility for investment finance. These banks had previously been simply the administers of government investment funds; now they were autonomous enterprises under regulatory control of the republican governments. Finally, the reform turned over most of the federal authority to raise taxes to the republics.
These changes meant an important power shift from the federal to the regional level and from territorially defined regions to ethnically defined republics. The shift in authority to the republic level doomed the regional development policy that was supposed to cement solidarity among the republics and loyalty to the federal center as it weakened the federal government even further. While the 1963 constitution had given more autonomy to the communes at the expense of the republics as a way of decreasing republican power, the 1965 reform returned authority to the republic level, and the commune’s economic authority was limited to attracting industry within its territorial boundaries. Here, Comisso argues, urban areas enjoyed immense advantages over rural communes, and this served to widen urban-rural social and economic divisions. The shifting of accountability from the center to the republics and the shifting of representative authority to the “nations” and nationalities shifted resentments away from the center and on to specific national groups.
1974: YUGOSLAVIA BECOMES A DE FACTO CONFEDERATION
Most analysts agree that the 1974 constitution was a watershed that turned Yugoslavia into a de facto confederation of sovereign states.72 The powers of the federal center were reduced to foreign policy, defense, the protection of national rights, and a minimum of economic instruments. Even in these realms, decisions had to be made by consensus among representatives of the republics and the autonomous provinces.73
The constitution further widened and deepened the system of ethnic and republican quotas to guarantee the smaller republics and nationalities that they would be equally represented. Where previously the quota system had generally followed the principle of proportional representation in federal appointments, it now stipulated that equal numbers from each republic regardless of population would be appointed to federal posts. Appointments to senior and mid-rank positions in federal and lower-level administrative and elective institutions, including the party, now came under the authority of republican and provincial party and state leaderships.
Finally and perhaps most important, the 1974 constitution changed the status of Kosovo and Vojvodina to autonomous provinces. They had been granted increasing authority over their investments and budgets after the 1965 reform. This constitutional change, however, moved them from a status of near-parity in the federal decision-making structure to complete equality with the republics. This meant that the Albanians in Kosovo now had de facto equal political status with the constituent national groups at the federal level. Now all six republics and the two provinces were equally represented in both chambers of the federal assembly regardless of their size. When collective leadership at the federal level was introduced, the two provinces joined in an eight-member presidency, in which each member had an equal vote.74 Indeed such an equal representation of the constituent units in both chambers of a bicameral federal legislative assembly is not found in any other contemporary federation. In comparative perspective, the small federal units in Yugoslavia were highly overrepresented, while large units were correspondingly underrepresented.75
Indeed, the 1974 Yugoslav constitution established a more decentralized system of industrial, political, and territorial decision-making than any other existing federation. Over the twenty-year period from 1953 to 1974, the constitutional move toward decentralization from the federal level toward republics and provinces gave legal status to the republics as power centers, making them in fact the highest self-governing communities in Yugoslavia. And the equal status conferred upon the republics and the provinces combined with the principle of unanimity in federal decision-making bodies to ensure immobilism at the center. Any representative who felt that the interests of his republic or province would not be met by a particular federal policy could block its implementation.76 Ethnic identity was given increasing political weight as ethnic representative bodies became more powerful and as the quota system was widened and deepened.
THE RESULTS: ECONOMIC FRAGMENTATION AND DECLINE
The result of this loss of power and authority at the center was increasing economic fragmentation of markets, duplication of investment projects, and a subsequent sharp decline in the economy as a whole. The complete story of Yugoslavia’s economic decline is a complex one, beyond the scope of this essay, and is yet to be written.77 I provide only a few examples here to illustrate the relationship between decentralization, the fragmentation of markets in Yugoslavia, and economic deterioration.
Once the regionalization of the economic policy was in place, the incentives for economic autarky increased.78 The regionalization of the banking sector witnessed the creation of as many banks as republics and regions. Bank authorities controlled allocation to individual firms, and regional regulatory authorities controlled banking practices. This regionalization of the banking structure made a nationwide monetary policy unattainable and blocked the possibility of interregional economic activity.79
With the regionalization of the banking system, preference in investment decisions was given to local objectives over the efficiency and profitability of the economy as a whole. Regional self-interest led to an increase in import substitution and the duplication of investment projects throughout Yugoslavia. In the period 1970-76, inter-republican trade in goods dropped from 27.7 percent to 23.1 percent of the national social product, while in 1981, 66 percent of all trade was intraregional and only 22 percent was interregional, with only 4 percent of all investment crossing republican and regional borders.80 Invisible but thick economic walls between the republics were gradually being constructed.
The devolution of authority to the republics to collect taxes worsened the economic situation further. It prevented the central government from having a coherent fiscal policy, and because the republican tax base was smaller, local and republic taxes on incomes were higher. Higher taxes reduced consumer purchasing power. By 1982, this, along with other problems associated with regional fragmentation, was reflected in a 36 percent drop in the volume of imports.81
As investment projects were duplicated and markets fragmented, overall economic growth ground to a halt. In 1982 real gross fixed investment fell by 37 percent. Labor productivity in the public sector fell by 20 percent, and public sector earnings fell by 25 percent. The average annual growth rate fell to 0.9 percent, a drop from an annual rate of 6.3 percent.82 As the economy worsened, regional fragmentation increased; the conduct of economic policy now depended on the wishes of the regional party organizations. Regional enterprises were subsidized as a part of patronage systems; patronage investments could only be financed by increased borrowing; increased borrowing deepened Yugoslavia’s external debt and worsened the economic system further.
As a result of uncoordinated investments, foreign reserve imbalances, and overborrowing in the 1970s, the 1980s witnessed permanent economic crisis in Yugoslavia. By mid-decade, inflation had reached 100 percent annually, while wages were frozen. The federal government faced a mounting debt obligation without any return on moneys spent. Unemployment rose from 600,000 in 1982 to 912,000 in 1983, not including the 700,000 who had been forced to emigrate abroad in order to find work. In 1981–85, unemployment in Serbia proper was 17–18 percent, and in Kosovo it was over 50 percent. By 1985 one million people were unemployed, and in all republics except Slovenia and Croatia the unemployment rate was above 20 percent.83
IMF structural adjustment loans only exacerbated regional tensions. For example, one requirement of the stabilization program was that the dinar be devalued. Bosnia was strongly opposed to devaluation because it was heavily dependent on imported intermediate goods from convertible currency areas. As the major exporters to the West, Croatia and Slovenia supported the decision to devalue. Because devaluations had to be approved by all republics, negotiations were time consuming, bitter, and divisive. Ultimately devaluation occurred, but exports failed to rise significantly and all economic indicators declined sharply. It was in this context of increasing fragmentation and permanent economic crisis that the centralizing organizations of party and army weakened, regional political entrepreneurs held sway as loyalty to the center dissipated, socioeconomic divisions dissolved into ethnic resentments, and center-region conflict gave way to national political rivalries. How and why ethnic divisions came to trump all others is the subject of the following section.
Above we saw that in order to maintain authority by deflecting criticism for economic hardship and political discrimination, the federal government decentralized its control over both the economy and the political system during the period 1953–74. With weakening power at the center, decentralization gave way to fragmentation, and fragmentation led to economic crisis and decline. As we shall see below, fragmentation also changed the rules of political accountability to make regional elites increasingly responsible to their local constituencies. Similar institutional changes increased constituents’ dependence on those elites and gave them more resources to distribute in exchange for support. Below I describe how, in the context of fragmentation and economic decline, the institutions of ethnofederalism permitted the logic of identity politics to shape the preferences of regional elites, weaken integrative institutions, and turn all potential social divisions into nationalist rivalries.
DECLINE OF LOYALTY TO THE CENTER
With the decentralization of political authority to the republics and the decline of central power came diminishing loyalty to the federal state. The devolution of power had an important negative impact on party loyalty at the federal level and on the cohesion of the army, the two institutional pillars of federal strength.
The more decentralized the system became, the more empowered were the regional party elites. As I have demonstrated above, as early as 1953, significant areas of political and economic authority had begun to devolve to the republics, and over the next ten years the republics gradually became important decision-making and patronage-dispensing centers.84 Recall that members of the Federal Assembly, previously accountable to communal assemblies, became accountable to the republican assemblies when the Chamber of Nationalities was upgraded in 1963 and 1967. With this institutional restructuring, ethnic and regional loyalties were bolstered and loyalty to the federal center weakened.
The regional party elites achieved key positions of power for two reasons: they were the most important economic actors and they were the most important party functionaries in the administration. The most powerful political leaders were those who had access to the state resources of the individual republics and the federal government. With those resources, politicians could create significant patronage machines. The deepening economic crisis made their role even more important because their aid became indispensable in keeping both enterprises and individuals afloat. As the central economic players, they controlled up to 70 percent of all federal investment funds, investing them in their own regions. In their role as regional party leaders, they made significant political and administrative appointments; for example, they controlled 25 percent of all employment in Kosovo; one out of four people was employed in administrative work within state-owned organizations there. In Slovenia one out of seven jobs was under the control of local and regional politicians.85
The changes in the 1974 constitution enhanced their power further. After 1974, when cadre selection was federalized, those with political ambitions knew that their careers were dependent on the approval of the republican and provincial bodies who sent them to Belgrade and knew that they would return to those bodies after federal service. The status of federal service was declining, and these career officials were often reluctant to accept a federal post. When they did enter federal service, they were always responsive to their home constituencies. The rules of accountability increased the power and attractiveness of local offices while reducing the power of the central ones. As some observers have noted, the party as such seemed to exist only for the duration of the party congress; by 1974 it had devolved into an umbrella organization, and regional LCY leaders viewed any effort to encourage Yugoslav integration as an attempt to undermine their respective power bases. As Cohen argues, “In place of the unified party elite that dominated the communist system in its initial postwar phase, the regime was now characterized by six republican and two regional party elites that skillfully utilized decentralized authority for their respective parochial interests.”86
In addition, as the center weakened, the allocative policies privileging partisans of all nationalities began to backfire. Recall that partisan privileges were intended to encourage Yugoslav integration and loyalty by cementing elite solidarity across national lines within the party. But integrationist goals were thwarted as these partisan elites created local dynasties and began to operate much like a powerful lobbying group, putting pressure on the central government to provide generous material support to their local communes and regions. In some areas, official veterans’ organizations exerted pressure on Belgrade to pursue policies favored by the local political machines of which they were a part. As the power of the center declined, partisan elites guarded their entrenched status jealously and were determined to ensure themselves the influence they felt was their due.87
The last pillar of Yugoslav loyalty to crumble under institutional incentives for decentralization was the JNA. Throughout the process of federal dissolution, the army had clung to its constitutional mandate to maintain the territorial integrity of Yugoslavia. It maintained its loyalty to the federal presidency. Even as the center disintegrated, it continued to pride itself on its multiethnic officer corps. But both external and internal pressures eroded the military’s integrationist function. First, the rich republics threatened its funding throughout the 1980s by continually grumbling over the size and destination of their contribution to the federal budget. The Croatian parliament voted to oppose federal financing for defense in general, and the Slovene parliament balked as well.88 Indeed because of its large share of the federal budget, the army became an important scapegoat for regional discontent against the center.
Internal problems in the army surfaced as well. Ethnic quotas in the appointment of officers had long fostered resentment, particularly among Serbs. Coming from the largest national group, they represented the majority of junior officers in the army but were restricted in opportunities for promotion by the quotas for national equality. After the Croatian crisis of 1971–72, Croatian soldiers balked at serving outside Croatia.89 Further, republican loyalty to the JNA was threatened when in 1987–88, the government of Slovenia supported demands that young men be allowed to do their military service at home in Slovenia rather than be sent to another republic. Tensions within the military were further heightened when the Slovene government called for the use of the Slovene language in all military communications and supported young people who campaigned for conscientious objector status.90 Both Djilas and Silber and Little suggest that by end of the 1980s the JNA began to mirror the weakness of the federal government as a whole. Shortly before Croatia and Slovenia moved toward secession, the officer corps was disproportionately Serb: with 40 percent of the population, Serbs represented 65 percent of the officer corps.91 By 1990, when Muslim and Croat youths ignored their induction notices completely, the army rank and file rapidly turned into a virtually all-Serb force.92
THE COLLAPSE OF SOCIOECONOMIC CLEAVAGES INTO REGIONAL DIVISIONS AND ETHNIC RESENTMENTS
Above I suggested that ethnic divisions had been partially dissolved through socioeconomic cleavages in the 1960s and 1970s. But again, decentralization, deregulation, and the ever-worsening economic situation transformed those cleavages into territorial and republican divisions. Three examples illustrate. First, producer associations were never organized functionally and thus were never able to enter into coalitions across republic lines. Recall that in the representative institutions of federal Yugoslavia, economic interests were mediated by territorially based institutions; producer representatives in the Federal Assembly were grouped first by commune and then by republic.93 Because they were subordinated to the republics and the communes, these associations never achieved autonomy; their functional interests were institutionally subordinated to the territorial interests of the republics. To the extent that ethnic divisions and political preferences granted according to ethnic identity coincided with territorial divisions, ethnic preferences and identity politics were reinforced by the representation of producer associations in the Federal Assembly.
Second, producers’ territorial dependence in representative institutions was reinforced by economic dependence on republican authorities. Recall that in the 1965 economic reform, all social investment funds, previously allocated to the enterprises directly from the central government, were transferred to communal banks.94 As a result, “extremely close” relationships developed among politicians, banks, and enterprises. As economic conditions declined and firms increasingly needed subsidies to stay afloat, their appeals to republican political authorities for favors multiplied. Local or republic governments would either aid the firms directly or, if they lacked the resources, would pressure the national government for more.95
With enterprises increasingly dependent on regional authorities, socioeconomic issues that could have transcended republican boundaries were increasingly translated into the long-standing center-region conflict. The more successful enterprises were concentrated in the more advanced republics of Croatia and Slovenia, and from the early 1960s onward, they increasingly found central government regulation constraining. They thus entered into informal coalitions with republican authorities to push for a decrease in federal control over their activities. On the other side were the centralizers, an informal coalition of politicians from the less developed republics, regions, and firms dependent on political subsidies and favoritism.96 Although the decentralizers triumphed in the 1965 economic reform, the patron-client relationships that lined up on either side of the conflict continued.
These relationships and the economic commitments they fostered resulted in the virtual absence of pan-Yugoslav economic integration at the firm level and a total lack of interrepublican investment and joint venture projects. The result was a dearth of countervailing pressures to diffuse the center-region conflict. A 1962 integration campaign had failed to produce any mergers across republic boundaries. As we saw above, interrepublican trade had dropped sharply by the late 1970s and never recovered. Indeed by the end of the 1970s, the Yugoslav market had disintegrated into eight separate mercantilist economies.97
Finally, by the 1980s, although occupation and resulting status differentials had the potential to create cleavages that crossed republican lines, the worsening economic situation and the subsequent collapse of the social welfare system led to a rise in the use of patronage networks, quotas, and cultural and ethnic bonds as the central mechanism by which scarce resources were allocated. Woodward writes that “in those poorer communities where job cuts were most severe and federal government subsidies and employment had been critical to the local economy, the employment requirement of proportionality and parity among national groups made ethnicity more salient rather than less.”98 In sum, at the elite level economic competition was subsumed in center-republic conflict, and for the public at large, economic decline and crisis fed ethnic resentments.
THE COLLAPSE OF IDEOLOGICAL CONFLICTS AND CENTER-REGION DISPUTES INTO NATIONALIST RIVALRIES
As we have seen, before the 1960s the central ideological dispute at the elite party level had emerged between the conservatives and liberal reformers. However, this division dissolved when liberals in Croatia allied with nationalists there to increase their political leverage against conservative forces. To gain the popular support of those who sympathized with nationalists, liberal reformers in the party began to issue increasingly vocal complaints about Croatia’s disadvantaged position in an “unfair” federal system. They began to call for an end to economic exploitation by Belgrade, reform of the banking and foreign currency systems, curbs on the wealth of Serbia’s export-import firms, and the redistribution of former federal assets that had been taken by Serbia after the reform. As a result, the liberal-nationalist coalition turned the initial liberal-conservative debate into a centralizing-decentralizing debate at the federal level.
In Croatia the liberal-nationalist alliance terrified the Serb minority and frightened potential liberal allies in other regions. Thus isolated, Croatia’s leadership relied on popular support and the increasingly bold alliance with nationalists within Croatia, further heightening tensions between Croats and Serbs both inside and outside the republic. Then in 1971 the liberalizers found that they could not end a strike at Zagreb University, organized by a militant group that they themselves had encouraged.99 Tito called in the JNA to quell the demonstration and, more important, to suppress the liberal-nationalist coalition. With backing from the JNA, Tito purged the party in Croatia of both its nationalist and liberalizing elements, leaving more conservative centralizers firmly in power.100
The “demonstration effect” then took hold. Liberalizing tendencies in the party had emerged throughout Yugoslavia, especially in Serbia.101 But the Croatian crisis suggested that an expansion of liberalism could open the door to nationalism. Thus in 1972 and 1973 liberals were removed from party leadership in all of the republics. By eliminating the liberal opposition in this way, the party ensured that in the case of its own demise, there would be no civil society to absorb the shocks of a transition.
While political liberalism had been crushed, economic liberals took sides in a fierce debate among the liberal reformers, ideological conservatives, and confederationists. The liberal reformers and defenders of a market economy were located primarily in the Serbian party. They argued for a unified Yugoslav market and the removal of economic barriers among Yugoslavia’s republics. Liberalizers were supported by IMF officials, who had stipulated a strengthening of federal institutions to unify Yugoslavia’s market. Confederalists, represented primarily by elites in Slovenia, Croatia, and Vojvodina, argued against the unification of the Yugoslav market; such unification would curtail the expansion of their own political autonomy and local power base. With their opposition to a market economy, they found unwitting allies in the ideological conservatives.
While always overlapping, ideological debates thus began to merge with conflicts between the center and the regions. Specific disputes between the centralizers and decentralizers took on new meaning; the centralizers—found mostly in Serbia and particularly in the Belgrade party—argued against the fragmentation of the national market and for the institutionalization of market mechanisms throughout Yugoslavia. The decentralizers—found primarily in Croatia, Slovenia, and Vojvodina—argued for the increasing use of self-management agreements on an enterprise (and thus a regional) basis in lieu of the market. Their rationale was that market mechanisms would constrain the decision-making rights of self-managed firms. By the end of the 1970s these center-region controversies began to be couched in veiled terms of national rivalries.102 In particular, anti-Serb rhetoric permeated the arguments of the decentralizers. But national rivalries would not break out in the open until Serb elites lost their loyalty to the center.
Indeed the Serbian party had always been on the side of the centralizers, in coalition with the poorer republics seeking subsidies from the federal government. But after the status of Kosovo and Vojvodina changed in the 1974 constitution, elites in both autonomous provinces argued on the side of the decentralizers, and Serb elites saw fewer reasons to remain loyal to the central Yugoslav government. By the late 1970s it appeared that the central state had ceased to serve the interests of Serb elites. With representatives in federal, republic, and party bodies from the national minority groups, both Kosovo and Vojvodina had the legal power to change the Serbian constitution and often voted against Serb preferences.103
This was to be the final blow to the center-region controversy. As we shall see below, Serb elites began to retreat from their support for federal institutions and openly encourage ethnic preferences for Serbs in response to Albanian discrimination against Serbs in Kosovo. Respect for minority rights was abandoned. At the federal level, Serb politicians began preparations to abolish the autonomous status of Vojvodina and Kosovo. Because of the increasing strength of republican party organizations, this move was entirely legal and politically possible. If the eight-man federal presidency were left in place after autonomy was abolished, Serbia would directly control three out of the eight votes. Other politicians would find this unacceptable, and they too would abandon federal institutions and retreat fully into republican sovereignty.104
In sum, by stealing all political loyalty from the center, fragmenting the organizations that propped up central authority, and providing local political entrepreneurs with resources and deepening ethnic resentments both among elites and in society at large, the institutions of ethnofederalism set the stage for identity politics to be played out in Yugoslavia. As we shall see in the following section, where ethnofederalism had been most entrenched, identity politics would be most vociferous. Where ethnofederal institutions were not well established, the destructive tendencies of identity politics did not take root.
The legacy of ethnofederalism in Serbia, Croatia, and Bosnia provided three incentives for ethnic bandwagoning and balancing to take hold and for regional politicians to play the ethnic card in their bid for political power. The first and most important effect was the demise of central power, which wiped out federal protection for national and minority rights and led to domination and discrimination of minority groups wherever one ethnic group enjoyed a majority. Domination and discrimination in one area prompted countermeasures in another, encouraging the escalation of open ethnic discrimination and violence. This in turn provided incentives for local politicians to exploit ethnic resentments for their own political advantage. Where the legacy of ethnofederalism was strongest, nationalist parties won the first “free” elections in federal Yugoslavia, held in 1990. Where they dominated republican governments, they created exclusive institutions and prevented losing ethnic groups from obtaining citizenship rights in their state, thus encouraging more secessionist violence.
Ethnofederalism’s second effect was to prevent the formation of political coalitions across ideological lines that could reverse this trend. It thus prevented the “pacted” and peaceful transition to democracy that had taken place in Latin America and Southern Europe.105 Third, by preventing political coalitions across regional lines, the legacy of ethnofederalism blocked liberal politicians from obtaining positions of power. To counter nationalist political forces, liberals needed pan-Yugoslav coalitions that regional fragmentation prevented. Given the absence of loyalty to the center, the absence of incentives for ideological and regional coalitions, and the presence of ethnic resentments spurred by institutions of accountability and representation, the dominance of identity politics and its escalation to violence in Croatia and Bosnia were assured.
BANDWAGONING AND BALANCING
The first move in the slippery slope toward ethnic violence was taken in Kosovo. Smoldering beneath the surface there—and encouraged by increasing autonomy—had been a radical Albanian move for republican status or even secession. Ethnic violence began to escalate as Albanians assaulted Serbs and vandalized their property. In 1981 riots broke out in Priština University in which Albanian students called for more autonomy; the JNA was called to intervene. Legal rights were weakened as Albanian officials hesitated to charge Albanians with hate crimes.106 Serbs and Montenegrins began to leave Kosovo by the thousands.
As Serbs continued to emigrate from Kosovo, economic hardship within Serbia deepened—partly as a result of the pressures of immigration. The immigration crisis, combined with the restriction of Serbia’s influence at the federal level by Kosovo and Vojvodina, pushed Serb elites to assert republican power over federal law and institutions. To halt immigration, the Serbian LCY implemented a series of affirmative discrimination measures favoring Serbs who stayed in Kosovo. It provided automatic admission of Serb students to Priština University, regardless of their qualifications. It prohibited the sale of land and buildings by Serbs and Montenegrins to Albanians. It promised jobs, housing, and schooling for Serbs and Montenegrins returning to Kosovo, and it built factories for Serb workers.107
The Kosovo crisis was interpreted in terms of ethnic discrimination and
privilege in the public debate; this interpretation opened the door for ethnic
nationalist sentiments to be freely expressed. As is now well documented in all
of the literature on the Yugoslav collapse, Slobodan Miloševi
With the open expression of nationalism now politically acceptable in Serbia and with the federal pillars of Yugoslav integration crumbling, the costs of using provocative nationalist rhetoric and engaging in ethnic violence were lowered throughout the Yugoslav republics. Extremist appeals crowded out moderate political platforms. Voters did not give ethnic entrepreneurs majorities in multiparty elections, but electoral rules combined with the political machines created under ethnofederalism and with incentives to bandwagon and balance at the elite level as more nationalist parties captured political space to escalate exclusive nationalist conflicts.
Slovenia was the first republic to hold multiparty elections in April 1990. The DEMOS, an anti-Communist six-party coalition, won 53 percent of the vote and took control of the parliament. Milan Kučan, the former Communist leader, won the presidential race.111 To balance what he saw as overwhelming Serb power at the federal level, Kučan had supported Albanian autonomy, publicly linking Albanian civil rights with the constitutional principle of territorial sovereignty and the right of secession. He portrayed Serbia as the enemy of Slovene democracy, as witnessed by its repression of Albanian rights, clearly heightening tensions between Serbia and Slovenia.112
Croatia was the next to hold multiparty elections. On 22 April 1990 the nationalist Croatian Democratic Union (HDZ) won the most votes in a majoritarian election and controlled two-thirds of the seats in the parliament. The electoral system underrepresented minorities and produced a legislative majority from a mere plurality of votes. The representation of minority parties, opinions, cleavages, and ethnic groups was thus artificially diminished. With only 41.5 percent of the vote, the HDZ got 58 percent of the seats in parliament. The single-member constituency electoral system further ensured that small parties were weeded out of any position of power or influence.113
The parliamentary majority of the HDZ permitted Croatia’s new president, Franjo Tudjman, to refuse minority rights to the 600,000-strong Serb population in Croatia, and the first constitution violated the Conference on Security and Cooperation in Europe (CSCE) principles on minority rights.114 Serbs were expelled from jobs because of their nationality. In Dalmatia, Croat gangs, often aided by the police, firebombed homes, smashed storefronts, and arrested Serb leaders. Croatian Serbs responded by demanding their civil and nationality rights. These demands fell on deaf ears; the federal government was now too weak to protect them. They thus held an autonomy referendum and built roadblocks around their areas to prevent Croatian interference.115 Croats living in mixed areas where Serbs began to mobilize saw this as Milošević’s hand stretching into Croatia.116 For Serbs in Croatia, these events gave credibility to the rising tone of nationalist rhetoric in Serbia. Local Serb leaders demanded that Serb-dominated territory be taken out of Croatia.117 Autonomy demands escalated to violence.
Bosnia-Herzegovina was the last republic to hold multiparty elections in December. On the surface it appeared that the elections would bring a successful multiethnic government to power. Although the three nationalist parties won the most votes, each from their own national group, none of the nationalist platforms was belligerent or aggressive. Although the Muslim party, the SDA, was represented by Islamic symbolism, its platform was a pluralist one. The Serbian party, the SDS, led by Radovan Karadžić, campaigned on a nationalist platform calling for the defense of Serb rights. But the party’s campaign did not call for the division of Bosnia. Thus those who voted for the SDA and the SDS were not voting for partition and war. Furthermore, unlike Croatia, Bosnia’s electoral rules followed the system of proportional representation. The system provided a close proportion of seats to votes, so that no one political party was underrepresented. Muslims gained ninety-nine seats in the assembly, Croats gained forty-nine, and Serbs from the SDS and other parties gained eighty-five. These seats closely represented the percentages of the vote gained by each party. Izetbegović formed a grand coalition among the three major parties, and government posts were divided among them.
This coalition, however, turned out to be a coalition of convenience in that it was created merely to form a government and not to achieve lasting accommodation, moderation, and compromise among the three dominant national groups.118 It quickly fell apart as Serbs began to declare large parts of the country “autonomous regions,” and SDS members of the republican presidency began to boycott presidency meetings. By October 1991 the SDS had left the assembly, which then promptly voted for Bosnian sovereignty. Several days later Karadžić set up a Serb Federal Assembly in Banja Luka. When a referendum was held on Bosnian independence, Karadžić’s SDS boycotted the election. On 26 April 1991 the Serbs of Bosnian Krajina created a separate assembly. Less than a year later, Serbs, Croats, and Muslims in Bosnia were at war.
Why did this happen? The Bosnian electoral system and government contained the key features of proportional representation and power-sharing that elsewhere have brought stable multiethnic governments to power in other divided societies.119 The system of proportional representation with very close proportionality was constructed to be fair to all constituent groups. It encouraged a proliferation of political parties so that all interests could be represented. Indeed forty-one parties, including the LCY, socialists, and Marković’s Alliance of Reformist Forces, took part in the electoral competition.
Part of the explanation for the political breakdown can be traced to the
timing of the Bosnian election and the ethnic alliances that had formed between
Serb nationalist political elites in Bosnia and Serbia and Croat nationalist
elites in Bosnia and Croatia. Bosnia was a latecomer, the last republic to hold
elections, and nationalist parties had formed and won elections throughout
Yugoslavia. Ethnic tensions had escalated in Croatia, lowering the cost of
jumping on ethnic bandwagons in other republics. Nationalists were firmly in
power in Serbia and Croatia. The success of Croat and Serb nationalists in their
titular republics induced Bosnian politicians to pursue exclusive ethnic or
religious nationalist strategies with the aid of their ethnic “brethren.”
Bosnian Muslims had been granted the status of nation and thus believed they
were justified in holding power as a nation. Indeed as Izetbegovi
Furthermore, ethnic alliances had formed across republican borders, bringing
material and symbolic resources to the Serb and Croat parties. Malcolm reports
that Miloševi
In short, ethnofederalism had prepared the way for these bandwagoning effects to induce political entrepreneurs to play the ethnic card. Slovenia and Croatia had long been the strongest advocates of decentralization and republican autonomy. By the 1990 elections, political and economic resources were in the hands of their regional and exclusive nationalist politicians. Serbia had long been a supporter of centralization but was pressured by new accountability rules in the 1974 constitution to relinquish political control over its territory. This intensified ethnofederalism induced Serb politicians to drop their support of the federal government and take control of territories populated by majority nationalities. Where other titular nationalities were making exclusive claims to territory, Bosnian Muslims also began to make territorial claims. The bandwagoning effect of exclusive national claims to territory reduced incenti