Political Cohesion In A Fragile Mosaic

The Yugoslav Experience

Lenard J Cohen, Paul V Warwick

• Publisher: Routledge
• Year: 2019
• Print ISBN: 9780367283414
• eBook ISBN: 9781000307184
• Topic: History
• Subtopic: World History
• Index: History

This book represents the first comprehensive empirical investigation of political cohesion in the multi-ethnic state of Yugoslavia, covering the entire period from the nation's independence to the present. The authors base their analysis on an extensive body of aggregate voting data from elections during both the precommunist and communist periods

1
The Study of Political Cohesion: A Neglected Issue

The cohesion and survival of political systems have been central concerns of rulers, citizens, and political analysts throughout history. These concerns have assumed a special and very practical urgency in societies composed of highly diverse and conflicting cultural groups, where centrifugal pressures constantly test the merits of political structures, policies, and leaders. In such divided or "segmented" societies, the difficulties of ensuring the regime's continued existence often comprise a good deal of the political process.

Among contemporary states, there is no better example than Yugoslavia of the problems that extreme cultural diversity present for the attainment of political cohesion. The cultural complexity of that polity is legendary and has made the very survival of a single Yugoslav state from 1918 to the present day something of a minor miracle.¹ But the achievement has been a troubled one at best. Failure, or the risk of failure, to maintain the state has been an ever-present possibility to Yugoslav leaders throughout the six decades of the country's existence. Even today, after thirty-eight years of communist rule, the problems of cohesion and regime survival still occupy the attentions of the country's rulers and strongly influence their policies. Yugoslavia remains a volatile and uncertain experiment.

That the Yugoslav state continues to exist as a single entity is due in part to the ceaseless quest for solutions by her various political leaderships. As diverse as the country's cultural makeup is the array of institutions, policies, and actions that have been devised and introduced to enhance the political cohesiveness of the polity. Indeed, as we shall elaborate in the next chapter, every major regime-type -- from right-wing authoritarian dictatorship to liberal democracy to socialist one-party rule and, more importantly, virtually every extant strategy for coping with cultural diversity has been tried at one time or another in Yugoslavia's brief history. Yugoslavia has not wanted for political innovation. But with what success?

Oddly, this question, so central to the Yugoslav experience and potentially so instructive concerning the problems of many other new states, has not received comprehensive and systematic treatment to date. It is true that considerable attention has been devoted to the exposition of the various culturally-derived centrifugal forces that have manifested themselves during each distinct historical period in Yugoslavia, and some of these studies of the "national problem" do offer exceptionally useful treatments of the historical literature (Shoup, 1968; Hondius, 1967; Rusinow; 1977). Most, however, have been restricted in scope to a short period of time or to the analysis of one event or personality (Tito, Djilas, etc.), and/or have lacked a solid empirical basis for generalizations and conclusions of a broader nature. Regrettably, these studies have occasionally been marred by extreme value commitments on the part of ex-participants in the specific issues and events under consideration (Popovic, 1968; Djilas, 1969; Tudjman, 1981).

It is also true that information of a more systematic sort on political attitudes and behavior has been provided by public opinion surveys conducted, for the most part, during the 1960s (Zaninovich, 1970; Jacob and Teune, 1971; Verba et al., 1971, 1978). Although these studies have been quite valuable in themselves, they, too, lack the scope to address larger issues such as the influence of changing institutional structures and strategies on levels of political cohesion throughout the course of Yugoslav political history. Moreover, the survey studies and the voluminous workers^' council literature in particular have had the inevitable effect of focusing attention almost exclusively on individual orientations, leaving the reader with the impression that properties such as political cohesion and regime durability emanate directly from the political attitudes elicited from the man in the street or the factory. But this is patently not the case, for these properties are properties of the system, not of its components taken individually, and as such have meaning only at the level of the system. Even if survey researchers were free to ask their respondents questions in such sensitive areas as system support and legitimacy, and even if the responses were frank and accurate, their individual-level focus would condemn the evidence they provided to at best an indirect and supportive role in dealing with such questions. Elite studies, which also offer important insights on the political process, are equally inappropriate for our central concern. Political cohesiveness, although not an individual-level concept, is not an elite phenomenon either: it is a property of the system, not of its leadership.

How, then, can the issue of political cohesion in Yugoslavia be addressed? Cohesion, which in ordinary usage refers to the process of sticking together or agglutination of parts, can be indicated empirically by a variety of phenomena that point to its absence: acts or statements of dissent on the part of intellectuals or political leaders who reject, for nationalistic or ideological reasons, the state as presently constituted; the emergence of nationalist or separatist movements, politically-motivated strikes and demonstrations, or widespread evasion of citizen duties such as payment of taxes or military service; violent confrontations between subcommunities or subcultural groups within the society; and the like. Yugoslavia has experienced no shortfall in these types of non-cohesive behaviors. However, to concentrate solely on activities of this sort, which usually involve minorities and are sporadic and haphazard in their occurrence, would be to neglect the general state of cohesiveness in the polity in favor of occasional, more overt manifestations at its fringes. It would, moreover, involve establishing the existence of political cohesion in an exclusively negative fashion, i.e., by the dearth of indications to the contrary.

Accordingly, in this study the principal indicator of political cohesiveness shall be the extent of mass political incorporation that exists in the polity. By political incorporation we mean the degree to which various components (territorial units, cultural groups, social classes, or other sub-societal collectivities), especially those which have been excluded from, peripheral to, or even outside the boundaries of the state, function as parts of a single political entity. This conceptualization of political incorporation relates to "the generalized problem of holding a system together" which Weiner (1965:55) has suggested as the core meaning of the term "integration." We prefer to avoid this latter term, however, because it has been associated definitionally with a variety of political processes, including national identity formation, elite-mass integration, central control over peripheral units, and societal capacity to organize for common purposes, that are outside the focus of this study. Our concern is more simply with the degree to which mass political behavior is manifested in ways that are consistent with the sustained existence of a single political system.

This study focusses upon the degree of political incorporation, so defined, that has existed at various stages in Yugoslav history, and its consequences and implications for the future of the state. As the empirical source for evidence of incorporation, we shall rely upon the one political activity that has engaged the participation of most adult Yugoslav males (and, since 1945, females) at fairly frequent if irregular points over the course of Yugoslavia's existence; namely, general elections. Specifically, this study undertakes the quantitative analysis of twelve legislative elections² spanning three regimes and sixty-odd years of Yugoslav history, with a view to determining the effect of regimes, strategies and events upon the state of political incorporation in the polity.

In order to proceed in this direction, we require a theory that can indicate which kinds of voting behaviors can be considered as indicative of incorporation, and which can not. Such a theory should also have as its main focus the political system itself, since it is not our goal to use aggregate voting data simply to make inferences about individual voting behavior and motivations. Finally, the theory must permit the comparison of results across different elections, even if the particular parties and issues involved are not the same, and, given the scope of this study, across different institutional frameworks (i.e., regimes) as well.

These are demanding assumptions, for the usual tendency in aggregate data analyses of national elections is to generalize "downward" to subgroups of the electorate, rather than upward towards the state of the political system. Even when the concern is with the system itself, the generalizations tend to be specific to a particular election, or, at most, to the regime which sponsored it. Fortunately, there is one theoretical approach in the literature that attempts to break out of these restrictions. This approach was presented by Stephen Coleman in his book, The Measurement and Analysis of Political Systems (1975). Coleman's theory is mathematical and abstract^ and at first glance may seem rather removed from the problems of incorporation in a multi-cultural state. A brief discussion of its nature and implications at this point, however, can suggest the manner in which the theory links aggregate data on voting behavior to the state of political incorporation in the system.³

Coleman's (1975:22) basic premise is that in the analysis of social (and political) systems, "we are concerned with the predictability of social events and their rate of occurrence." By "events," Coleman means the alternative outcomes or choices that are possible in a given situation. To measure the predictability associated with a situation, Coleman borrows from information theory the concept of "entropy," a measure that ranges in value from zero (i.e., minimal uncertainty) when one event or outcome is inevitable to the logarithm of the number of events, which occurs when all of the events are equally probable and predictability is at a minimum for that situation.⁴

One important advantage of the entropy measure for comparative purposes is that it is a function, not of what the various events or alternatives in the situation or "event-set" are, but simply of their number and their relative sizes. This property of being content-free suggests Coleman's main hypothesis:

The entropy measurement gives the average social uncertainty about what will happen for events in event-sets in the social system, the entropy measurement is independent of whatever the content of the event is. Therefore, we must assume that we would find approximately the same entropy value in different event-sets within the system (Coleman, 1975:37).

Elections provide a suitable forum to analyze the validity of this hypothesis because they provide two separate, measurable event-sets. One event-set is constituted of the choice among the various parties or candidates; the other is the two-choice event-set of voting or abstaining. According to Coleman's reasoning, the entropy measurements of these two event-sets, referred to as party-choice entropy and turnout entropy respectively, should be equal once differences in their ranges have been controlled for. This constitutes the entropy hypothesis.

The importance of the entropy hypothesis for our purposes lies in its use of electoral data to make inferences, not about voters, but about the political system itself. For Coleman, an election is not just the aggregation of a large number of individual choices that can be analyzed best by means of public opinion surveys or other individual-level techniques, but rather a collective act expressing a community level of social complexity or uncertainty. It is this assumption that allows Coleman to postulate conditions for system survival or breakdown, which are of prime importance to us.

The entropy level of a society, since it reflects average social uncertainty, ideally should be uniform not only for all event-sets in the social system, but across the territory occupied by the social system as well. In practice, this is a rather unlikely situation. For instance, since processes associated with economic development, such as urbanization and industrialization, increase social complexity and thus uncertainty, one would expect to find that entropy levels are higher in more developed regions of any country. Therefore Coleman (1975:127-143) argues that in a well-integrated or cohesive political system, there should be, if not a uniform entropy level, at least a smooth, "harmonic" distribution of entropy values over the various electoral units such that the minimum and maximum values occur at the geographic boundaries of the system and other districts display patterns of graduated departure from these extreme values towards values that lie in-between. By way of contrast, systems which contain sharp internal discontinuities, such as a ridge of high (party-choice) entropy values adjoining a low entropy region, exhibit weak cohesion and are inherently unstable. Unless this pattern substantially changes, systems of this sort cannot be expected to survive.

The value of Coleman's approach to our objectives should now be clear: Coleman proposes the use of electoral data to assess the degree of political incorporation of the various geographical units of the political system and its likelihood of survival. Nonincorporation is related to survivability because in poorly incorporated polities, certain territorial units will display voting patterns distinct enough to be regarded as separate political subsystems for which secession becomes an attractive option. In this study, the entropy hypothesis will be employed to best effect in Chapter 3, which contains the analysis of the elections of the interwar era. The elections of the 1920s, the first two of which were basically "free" in the liberal-democratic sense, provide the most accurate measurement of the natural propensities of the political system. Our analysis of these elections will demonstrate a very high and increasing level of political non-cohesiveness in this period, a non-cohesiveness which grew to the point of system breakdown in the latter 1920s, despite the policies and electoral manipulations the authorities introduced to check it. The paralysis and threatened collapse of the system provided the context for the first major structural manipulation by the country's leaders: the introduction of a conservative, authoritarian political system in 1929. In the remainder of Chapter 3, we shall explore the efficacy of that regime's strategy by examining its effect upon mass electoral behavior in the 1935 election. The brief and often overlooked experiment in the late 1930s with aspects of what today would be known as a "consociational" strategy, which was doomed by the international developments of that period, will also be examined.

The quest for political cohesion took a radically different turn with the accession to power of the Communists under Marshall Tito. In Chapter 4 we shall discuss the evolution of the party's thinking on cultural diversity up to 1950, and the ways in which the party's own unbalanced ethnic and regional support affected the implementation of this thinking. This discussion of the party's approach to ethnic relations will set the stage for the empirical analysis, in Chapters 5 and 6, of the regime's postwar efforts to foster political incorporation.

The means by which we assess political incorporation must be altered to take account of the constrained context of Communist-sponsored elections. The Communist regime has gone to extreme lengths to avoid any possibility of ethnic considerations, which formed the basis for the politics of the interwar period, becoming a factor in postwar elections. With the divisive tendencies of inter-ethnic relations eliminated from the electoral process by structural means, they had hoped that elections would function as a legitimating and incorporating force, demonstrating and encouraging the alignment of the masses behind the leadership and goals of the party. Accordingly, the principal means by which non-incorporation in the political system is manifested in this period is through patterned dissent. Electoral manifestations of dissent take two forms: electoral abstention and deliberate ballot invalidation. There can, of course, be many idiosyncratic reasons for voters to exhibit either of these behaviors; therefore the crucial element in assessing dissent is the degree to which it is patterned along ethnic lines. In Chapter 5, we analyze the results of every federal election from 1950 to 196...

Table of contents

• Cover
• Half Title
• Dedication
• Title
• Copyright
• Contents
• List of Tables
• List of Figures
• Preface
• 1 THE STUDY OF POLITICAL COHESION: A NEGLECTED ISSUE
• 2 THE MANAGEMENT OF CULTURAL DIVERSITY: THEORY AND CASE STUDY
• 3 DIVERSITY AND DIVERGENCE: YUGOSLAVIA IN THE INTERWAR YEARS
• 4 THE COMMUNISTS AND CULTURAL DIVERSITY
• 5 INCORPORATION UNDER COMMUNISM, 1950-1969
• 6 ONE PARTY, MANY CHOICES: PLURALIST SOCIALISM IN THE 1960s
• 7 CONSOLIDATION WITHOUT COHESION: POLITICAL STABILITY IN A FRAGILE MOSAIC
• APPENDIXES
• REFERENCES
• INDEX