¹ Anderson, Perry, Lineages of the Absolutist State (London:Verso, 1979Google Scholar); Brenner, Robert, “Agrarian Class Structure and Economic Development in Pre-Industrial Europe,” Past and Present (No. 70, 1976), 30–75CrossRefGoogle Scholar.

² This is true of the most representative of neoinstitutionalist works, Evans, Peter B., Rueschemeyer, Dietrich, and Skocpol, Theda, eds., Bringing the State Back In (Cambridge:Cambridge University Press, 1985CrossRefGoogle Scholar), and of two primary practitioners—Theda Skocpol in her work on the New Deal and in States and Social Revolutions: A Comparative Analysis ofFrance, Russia and China (Cambridge:Cambridge University Press, 1979Google Scholar), and Skowronek, Stephen, Building a New American State: The Expansion of National Administrative Capacities 1877–1920 (Cambridge:Cambridge University Press, 1982CrossRefGoogle Scholar).

³ See Aston, T. H. and Philpin, C.H.E., The Brenner Debate: Agrarian Class Structure and Economic Development in Pre-Industrial Europe (Cambridge:Cambridge University Press, 1987Google Scholar); Janos, Andrew, The Politics of Backwardness in Hungary, 1825–1945 (Princeton:Princeton University Press, 1982Google Scholar); and the work of historians in Eastern Europe, especially the Hungarian Peter Hanak.

⁴ Much of what Marx wrote about the “state” was after all developed in response to the French “mix,” particularly in the 1840S-50S writings, Class Struggles in France, 1848–50 and The 18th Brumaire of Louis Bonaparte; the alternation between regimes and the central-regional conflicts, particularly in the political economic alliances they represented, in 19th-and early 20th-century France have characteristics strikingly similar to contemporary Yugoslavia.

⁵ The best of these is the work by Ivan Szelenyi; see, for example, “The Intelligentsia in the Class Structure of State-Socialist Societies,” in Burawoy, Michael and Skocpol, Theda, eds., Marxist Inquiries: Studies of Labor, Class, and States (Chicago:University of Chicago Press, 1982), 287–326Google Scholar.

⁶ Bauman, Zygmunt, “Officialdom and Class: Bases of Inequality in Socialist Society,” in Parkin, Frank, ed., The Social Analysis of Class Structure (London:Tavistock, 1974), 129Google Scholar–48.

⁷ The continuity of the Soviet state with Russian autocracy, of Bolshevism with the Christian church or with prerevolutionary absolutism, and the concept of Leninist regimes as an ideology of organization that is constant are examples; see Brzezinski, Zbigniew, “The Patterns of Autocracy,” in Black, Cyril E., ed., The Transformation of Russian Society: Aspects of Social Change since 1861 (Cambridge:Harvard University Press, 1960), 93–110Google Scholar; Cohen, Stephen F., “Bolshevism and Stalinism,” in Tucker, Robert C., ed., Stalinism: Essays in Historical Interpretation (New York:Norton, 1977), 3–19Google Scholar; and the work of Jowitt, Kenneth, esp. “An Organizational Approach to the Study of Political Culture in Marxist-Leninist Systems,” American Political Science Review 68 (September 1974), 1171CrossRefGoogle Scholar–91, The Leninist Response to National Dependency (Berkeley:Institute for International Affairs, University of California, 1978Google Scholar), and “Inclusion and Mobilization in European Leninist Regimes,” World Politics 28 (October 1975), 69–96CrossRefGoogle Scholar. On the social contract argument, see Pravda, Alex, “East-West Interdependence and the Social Compact in Eastern Europe,” in Bornstein, Morris et al., eds., East-West Relations and the Future of Eastern Europe (Boston:George Allen & Unwin, 1981), 162–187Google Scholar; Colton, Timothy, The Dilemma of Reform in the Soviet Union, 2d ed. (New York:Council on Foreign Relations, 1986Google Scholar); and Hauslohner, Peter, “Gorbachev's Social Contract,” Soviet Economy 3 (No. 1, 1987), 54–89Google Scholar.

⁸ This approach begins with Skilling, H. Gordon, “Interest Groups and Communist Politics,” World Politics 18 (April 1966), 435CrossRefGoogle Scholar–51, and has been furthered most by Jerry Hough. Two fine recent contributions are Nina Halpern, “Policy Communities, Garbage Cans, and the Chinese Economic Policy Process,” unpub., presented at the annual meeting of the American Political Science Association, Chicago, August 1987, and Solomon, Susan Gross, ed., Pluralism in the Soviet Union (New York:St. Martin's Press, 1983CrossRefGoogle Scholar). A useful attempt to deal directly with this problem is Prasenjit Duara's concept of state involution, a “variation of the state-making process wherein the formal structures of the state grow simultaneously with informal structures”; see his “State Involution: A Study of Local Finances in North China, 1911–1935,” Comparative Studies in Society and History 29 (January 1987), 132CrossRefGoogle Scholar–61.

⁹ This is the implication of Comisso's “State Structures and Political Processes outside the CMEA: A Comparison,” in Comisso, Ellen and D'Andrea Tyson, Laura, eds., Power, Purpose, and Collective Choice: Economic Strategy in Socialist States (Ithaca, NY:Cornell University Press, 1986), 400Google Scholar–21. The debate between A. Ross Johnson and William Zimmerman originates in different views of the state itself, which can be seen in their discussion of Yugoslav international alliances; for Johnson, the state boundary is military and foreign policy; for Zimmerman, it is foreign trade. See Johnson, , “Is Yugoslavia Leninist?” and Zimmerman, “Rejoinder,” in Studies in Comparative Communism 10 (Winter 1977), 403CrossRefGoogle Scholar–11. The Soviet reforms under Gorbachev make this Leninism of Yugoslav reform clearer.

¹⁰ See Dorothy Solinger's triadic model in Chinese Business under Socialism: The Politics of Domestic Commerce 1949–1980 (Berkeley and Los Angeles: University of California Press, 1985) for an account that more closely captures Yugoslav coalitions than the normal factional model of socialist state politics.

¹¹ This classification rises and falls on the placement of Serbia, which many erroneously assume to be less developed. A conservative is loosely conceived in all these works on the 1960s as a supporter of the status quo ante 1966 (as these authors portray it); the term could be more carefully defined.

¹² This is the standard argument of economic reformers in socialist systems, best known through the work of Janos Kornai; see, e.g., his Contradictions and Dilemmas: Studies on the Socialist Economy and Society (Budapest: Corvina, 1983). The literature tends to speak in metaphors and ideologies of market economies rather than to elaborate the how and why of the institutional change that is considered necessary for their operation. Nor are the economic and political aspects of their demands for greater autonomy as separate as some authors imply. Situated between this argument and Carter's is Ellen Comisso's thesis of market (autonomy) vs. plan (center) dialectic, which she posits for Yugoslav self-management in Workers' Control under Plan and Market (New Haven and London:Yale University Press, 1979Google Scholar).

¹³ Lijphart, Arend, The Politics of Accommodation: Pluralism and Democracy in the Netherlands, 2d ed., rev. (Berkeley and Los Angeles:University of California Press, 1975Google Scholar), and Democracy in Plural Societies: A Comparative Exploration (New Haven:Yale University Press, 1977Google Scholar).

¹⁴ Kaplan, , System and Process in International Politics (New York:John Wiley & SonsGoogle Scholar, Another of Ramet's arguments is that central planning cannot exist without agreement, and that, because the leadership could not obtain agreement among the republics, it had to decentralize; this does not fit the fact that the move away from central planning occurred in 1950–1952.

¹⁵ Tilly, Charles, ed., The Formation of National States in Western Europe (Princeton: Princeton University Press, 1975), 3–83Google Scholar. Cohen and Warwick could of course be applying unitarist Yugoslavism (narodno jedinstvo) instead; see Banac, throughout.

¹⁶ This tendency to periodize politics in communist party-states according to their leaders' tenure is the norm; Valerie Bunce takes it to the level of axiom in Do New Leaders Make a Difference? (Princeton:Princeton University Press, 1981Google Scholar); Cohen and Warwick also periodize the interwar period by administrations. The case of purges is particularly debatable; they may reflect leaders' calculations that dissident colleagues could call on a sizable popular following should they choose to continue opposition to the policy taken. This would make purges significant political events, if not causes of change; but the argument needs to be made.

¹⁷ Further,

Whereas the proponents of unitarism sought to obliterate all historically derived differences between the Serbs, Croats, and Slovenes by means of a strictly centralized state, the majority of Serbian political parties used centralism to further Serbian predominance. This meant that the partisans of a unitarist and a Great Serbian version of Yugoslavia were counterposed to the representatives of the non-Serb national movements, who either demanded a federated (or confederated) state structure or sought guarantees for their national aspirations outside the framework of Yugoslavia (Banac, 214–15).

¹⁸ In the elections just prior to the declaration of the dictatorship in 1929, Cohen and Warwick (p. 35) find a ridge of increasing entropy (measure of average social uncertainty for event-sets, in this case choice of party and turnout in parliamentary elections), or fragmentation into political subsystems, running through the region of mixed population and administrative systems of the former imperial borderlands that parallels Coleman's findings for North-South division in the United States in 1860. See Coleman, Stephen, Measurement and Analysis of Political Systems: A Science of Social Behavior (New York:Wiley-Interscience, 1975Google Scholar).

¹⁹ That is not to deny the extraordinary methodological problems of such analysis (which Cohen and Warwick elaborate), or their persuasive argument that electoral data are an underutilized source of information about Yugoslav politics; it is only to raise questions about the inferences they draw because of their assumptions about the data themselves and their choice of a contentless measure to make intertemporal comparisons. (Their hypothesis is that entropy measures for different regions will become more continuous and “harmonious” over time if the political system is becoming more stable and internally cohesive.)

²⁰ Because data are often gathered for political purposes, empirical analysis is not always a sure way around ideologies unless the content of those ideologies is taken into account; that is one of the many values of Banac's study. For example, the census of 1921 on which Cohen and Warwick rely for ethnic composition “reveals a good deal about the official ideology, [but] is not particularly helpful as a statistical guide to the size of each national community” (Banac, 49ff). Banac's discussion of the elections of the 1920s also casts doubt on a major assumption of their analysis: that, because the elections were held under a “liberal democratic regime” and were thus “free,” they “provide the most accurate measurement of the natural propensities of the political system” (Cohen and Warwick, 5). Banac, too, can succumb, however; see his use of data on taxation from the political pamphlet by Rudolf Bičanić for the Croatian Peasant Party, which initiated a famous debate on national discrimination in economic policy under the Kingdom in 1938–40; see Ekonomska Podloga Hrvatskog Pitanja [The economic basis of the Croatian question] (Zagreb:Vladko Maček, 1938Google Scholar).

²¹ Burg cites the report of the party commission set up to investigate Ranković's crimes (i.e., to legitimize the purge). There is substantial circumstantial evidence to be skeptical toward its claims, however, including the striking parallels with the party commissions to investigate leaders purged earlier (e.g., Hebrang and Zujovic in 1948), and later statements from political leaders, including Djilas, in Tito: The Story from Inside (New York:Harcourt Brace Jovanovich, 1980Google Scholar), and Rise and Fall (New York:Harcourt Brace Jovanovich, 1985Google Scholar). Burg's extensive use of party documents is a great strength; but, because these documents are also political acts, they need interpretation.

²² See Banac, Ivo, “Yugoslav Cominformist Organizations and Insurgent Activity: 1948–1954,” in Vucinich, Wayne S., ed., At the Brinks of War and Peace: The Tito-Stalin Split in a Historic Perspective (New York:Brooklyn College Press, 1982), 239Google Scholar–52; see also Banac's forthcoming monograph on the Cominformists.

²³ Banac (p. 2911) calculates that the 1838–1843 membership of Matica Ilirska, the Croat patriotic and cultural organization, consisted of administrators and public servants (27%), clergy (22%), free professionals (primarily lawyers) (18%), craftsmen and merchants (14%), teachers (7%), and noblemen (11%).

²⁴ Ethnographers, members of a still young profession, were included in national delegations to the negotiating tables to provide “scientific” evidence about the ethnic composition of territories; the most famous was Jovan Cvijić, who worked on behalf of the Serbs.

²⁵ A legally defined unit, not kin-based like the extended household in Serbia of the same name and ethnographic fame. See Byrnes, Robert F., ed., Communal Families in the Balkans: Essays by Philip E. Moseley and Essays in His Honor (Notre Dame, IN:University of Notre Dame Press, 1976Google Scholar).

²⁶ See Donia, Robert J., Islam under the Double Eagle: The Muslims of Bosnia and Hercegovina, 1878–1914 (Boulder, CO:East European Monographs, 1981Google Scholar). The situation in Dalmatia (which ended up under Habsburg control) shows yet another set of complexities, primarily because ofearlier Venetian policies. Verdery's argument on Romania is analytically very helpful: she identifies the “social-structural foundations” of group difference, where ethnic “cultures” are associated with different status groups, often territorially defined, and how shifts in status led to shifts in ethnic identity. See Verdery, Katherine, Transylvanian Villagers: Three Centuries of Political, Economic, and Ethnic Change (Berkeley and Los Angeles:University California Press, 1983Google Scholar).

²⁷ This is why, according to Banac, intellectuals create nationalist movements. Moreover, they remain a political force in this part of the world—and are perceived as a threat by ruling groups—because of their ability to keep the memory of previous political rights alive and to interpret that tradition under changing conditions. There is a similarity with the “literature of memory” tradition in Central and Eastern Europe, which should prove the usefulness of Banac's discussion of the South Slav case to other national movements in the area, such as the Poles and the Czechs, especially those originating within the Habsburg Empire. A contrasting approach is in Anderson, Benedict, Imagined Communities: Reflections on the Origin and Spread ofNationalism (London:Verso, 1983Google Scholar).

²⁸ Because the farming class in Sweden, for example, had its own estate separate from the nobility, it could organize independently; this influenced the alliances and policies possible in Sweden's 20th-century state. For an elaboration, see Esping-Andersen, Gøsta, Politics Against Markets: The Social Democratic Road to Power (Princeton:Princeton University Press, 1985Google Scholar).

²⁹ Further illustration of such difficulties is found in Banac's fascinating discussion (pp. 359–78) of the quandaries of the Muslim political organizations (above all the JMO) during the interwar period.

³⁰ The Allies offered much of Dalmatia to Italy if it would shift sides in the war; the Habs-burg threat of a separate peace included an independent Croatia.

³¹ Banac (p. 131) quotes a report by a National Council official in Slavonia (Croatia):

The people are in revolt. Total disorganization prevails. Only the army, moreover only the Serbian army, can restore order. The people are burning and destroying. .. . The mob is now pillaging the merchants, since all the landed estates have already been destroyed. Private fortunes are destroyed. The Serbian army is the only salvation.

³² Only when pressure by the British and French for liberal democratic institutions yielded in 1929 was the king relieved of any restraint on outright dictatorship. Four years later, to appease German militarism, Britain and France handed Yugoslavia, along with the rest of Eastern Europe, over to a German sphere of interest; the king was assassinated by Croat and Macedonian fascists paid by Mussolini; and the domestic political order under Premier Stojadinović took on ever more coercive tendencies and imitation of fascist institutions.

³³ See Kljaković, Vojmir, “The International Significance of the Second Session of AVNOJ,” Socialist Thought and Practice 23 (December 1983), 63–76Google Scholar.

³⁴ Samouprava, more often rendered as self-management.

³⁵ Although much is made of this as the Croatian program, its full details are in fact closer ' to the Slovene constitutional proposal of 1921—suggesting, as does the entire path of political construction in the postwar period, the dominant influence of Edvard Kardelj, who was a member of the wartime leadership, head of the Slovene Party, and holder of many governmental positions, from Minister of Foreign Affairs to Vice-President. See Beard, Charles and Radin, George, The Balkan Pivot: Yugoslavia (New York:Macmillan, 1929Google Scholar). The Slovene role also serves as reminder that there is more to the Yugoslav national question than a Croat-Serb conflict.

³⁶ See the speech before parliament on July 20, 1946, by the chair of the Economic Council, Boris Kidrič, “Obrazloženje Osnovnog Zakonao Državnim Privrednim Preduzećima” [Exposition of the basic law on state economic enterprises], reprinted in Kidric, Boris, Privredni Problemi FNRJ [Economic problems of the FPRY] (Belgrade:Kultura, 1948), 7–21Google Scholar.

³⁷ Including the creation of NATO, the final division of Germany, the neutralization of Austria, and the negotiations over Eastern security that led to the Warsaw Pact in May 1955.

³⁸ See Veljko Vlahović's speech to the Executive Committee (enlarged session) of the League of Communists on February 10, 1961, in A Step Backward (Belgrade:Jugoslavia, 1961Google Scholar), and Kardelj, Edvard, Socialism and War: A Survey of the Chinese Criticism of the Policy of Coexistence (New York:McGraw-Hill, 1960Google Scholar).

³⁹ For details, see the chapter on Yugoslavia in Bitterman, Henry J., The Refunding of International Debt (Durham, NC:Duke University Press, 1973Google Scholar).

⁴⁰ See Bebler, Anton, “Development of Sociology of Militaria in Yugoslavia,” Armed Forces and Society 3 (November 1976), 59–68CrossRefGoogle Scholar, for the dating of this policy. It was discussed again by the executive committee of the League of Communists of Yugoslavia in 1968, enacted in the Law on National Defense in February 1969, and publicly proclaimed at the Ninth Party Congress in April 1969. The doctrine of “general peoples' defense” was a return to the partisan war concept of total popular mobilization, on the one hand, and, on the other, a territorialization akin to the 1881 reform of the Habsburg army, which placed operations under the republics' military staffs, emphasized local postings, and aimed to integrate the army more fully into social and political life. See also Dean, Robert W., “Civil-Military Relations in Yugoslavia, 1971–75”, Armed Forces and Society 3 (November 1976), 17–58CrossRefGoogle Scholar.

⁴¹ Turčinović, Slobodan, “Financing Socio-Political Units, 1961–1967,” Yugoslav Survey 9 (May 1968), 59–74Google Scholar. The proportion of national income expended by governmental budgets fell from 31.4% in 1961 to 25.7% in 1964 and to 18.9% in 1966; the federal proportion of spent budgetary revenues decreased from 54.4% in 1964 to 50.3% in 1966, or, when subsidies to republics are deducted, from 50.6% in 1964 to 45.5% in 1966. The share of communes rose from 27.6% in 1964 to 31.9% in 1966 (Turčinović, 73–74).

⁴² Slovenia created a political crisis because it did not receive the portion of IBRD funds it wanted; Bosnia took its dissatisfaction with the criteria set for development aid by the federal fund to the parliament, with constitutional consequences. These examples contradict Burg's division of the republics into two groups on levels of development and attitudes toward redistribution, as does the assignment of Serbia to the less developed camp (which is true in some ways but not in others, and does not take account of the strength of Serb liberals).

⁴³ Economic Chambers are sectorally organized associations of enterprises (previously the directorates between ministry and enterprise in the planning hierarchy) that also participate in making and enforcing policy; they deserve far more study than the preoccupation with the federal system accords them.

⁴⁴ The same military border and marginal regional towns and a similar constituency—lawyers, schoolteachers, university students (but not the veterans)—gave early and strong support to the Croat cultural and patriotic organization Matica Hrvatska, which came to symbolize the nationalist “threat” in 1967–1971 while the official position of the Croat leadership echoed with early Yugoslavism; see fns. 18 and 23.

⁴⁵ Robert W. Dean reports that by 1972, of 437,709 veterans in Serbia only 32% were employed, while 66% were retired (about half of these were under the age of 55); in a 1971 national survey of the partisan elite “first fighters” (joining the partisans in 1941), more than half had been retired before pension age and 27% had incomes below the legal minimum and “complained most of social isolation, inactivity, and lack of prestige.” Dean, “Civil-Military Relations in Yugoslavia, 1971–45,” Armed Forces and Society 3 (November 1976), 17–58, at 54n.

⁴⁶ Party finances are based primarily on members' dues, which are assessed on a progressive, proportionate scale of individual incomes. Thus, a bias is built in against those at the lower end of the income scale—young people starting out, rural youth, and blue-collar workers—both in the selection process and in the retention of members. (The first cause for expulsion, which affects many workers, is nonpayment of dues.) The more party finances were decentralized and put on an economic accounting basis, the greater would be this bias in poorer localities.

⁴⁷ Some victories, as one of Carter's election stories shows (p. 147), represented genuine disagreement on policies: in one Serbian constituency, a Partisan general unseated the official candidate, the federal secretary for trade, by disparaging his opponent's war record and his policy of permitting the import of cheap apples to the detriment of local fruit growers, by appealing to disgruntled peasants who had lost land in the agricultural reforms of 1958–1966, and by promising locals he would win their struggle for a railway line.

⁴⁸ For the empirical evidence, see Estrin, chaps. 5 and 6. Estrin's interest is in testing some of the theoretical literature on labor-managed firms—“how firms will behave when the la-bour force plays the role of entrepreneur” (p. 1). He also argues—by indirection—that self-management explains allocative “imperfections” in Yugoslavia better than do the capital market theorists (the property rights school of Furubotn and Pejovich and the institutional school of Vanek and Jovičić).

⁴⁹ One reason the Albanian-Serb conflict in Kosovo is so intractable may be that Kosovar autonomists base their claims for territorially defined political rights (as they did in the inter-war period against Serbia) on “historicist” grounds; but these are lands that Serbs also claim on historicist grounds—the rights of “primary acquisition” (the locus of the medieval Serb empire)—rather than on assimilationist principles. The Albanians' tactics to become the numerical majority in the region—first through their birth rate and later by harassing non-Albanians to move—may be seen as a predictable outcome of this ideological stalemate.

⁵⁰ The delegate system, by which parliamentary representatives at all three levels of government represent a “base” within a self-managed unit, was introduced in 1967 and has been a constitutional principle since 1974. Persons elected are already employed and in no need of supplementary (out-of-budget) income and perquisites. Self-managed interest communities (sics) administer funds to provide social services that were transferred, with its bureaucracy, from a government budget to self-management; further financing is voted directly by enterprise representatives to a sic assembly. The sics' numbers grew exponentially after 1969.

⁵¹ By the fall of 1988, the parallel with the disturbances of 1969–1971 was particularly striking, even though the alignment of republic leaderships and the issues for students and workers had changed. Now Serbian leaders (headed by Slobodan Milosevic) were mobilizing popular nationalism in their bargaining for political reform, Slovenia was leading the fight over the foreign exchange regime, and Serb and Croat leaders were both neoconservatives. But the conflicts between proposals for economic and for political reform—those of liberal nationalism—have continued to divide the bargaining parties over the locus and criteria offiscaland political (courts, police) regimes, even though they now agree on economic market reform. That a compromise has been negotiated without the presence of Tito illustrates the systemic character of these politics, and the role of international finance is far more visible.

⁵² Estrin argues (p. 2) that “labour market forces are weak in a self-managed system” according to both theory and his empirical data; the entry of new firms “effectively ceased” after 1964 (p. 90); mergers “generally occurred in a commune or neighbourhood” (p. 95); and overall concentration ratios began high and rose throughout the 1960s. As in the interwar period, declining labor mobility should be considered a significant factor in the popular responsiveness to national appeals, independent of whether their political expression is permitted or not. Similarly, when urbanization, social and economic mobility, and/or war undercut the residential bonding of communal (religious, occupational, racial) identities and their political expression in votes, loyalties, and collective action in three of Lijphart's main cases—the Netherlands, Lebanon, and South Africa—his predictions of stable rule through elite concert failed. See Lijphart (fn. 13).