¹ Mayall, , “Nationalism and International Security after the Cold War,” Survival 34 (Spring 1992)CrossRefGoogle Scholar.

² For an exploration of the dissolution of the Soviet Union as an imperial collapse, see Dawisha, Karen and Parrott, Bruce, eds., The End of Empire: The Transformation of the USSR in Comparative Perspective (Armonk, N.Y.: M. E. Sharpe, 1997)Google Scholar.

³ Horowitz, Donald L., Ethnic Groups in Conflict (Berkeley: University of California Press, 1985)Google Scholar.

⁴ Linz, Juan and Stepan, Alfred, “Political Identities and Electoral Sequences: Spain, the Soviet Union, and Yugoslavia,” Daedalus 121 (Spring 1992), 123Google Scholar. For additional discussion of identity issues in democratic transitions, see Linz, and Stepan, , Problems of Democratic Transition and Consolidation: Southern Europe, South America and Post-Communist Europe (Baltimore: Johns Hopkins University Press, 1996)Google Scholar; Claus Offe, “Capitalism by Democratic Design? Democratic Theory Facing the Triple Transition in East Central Europe,” Social Research 58 (Winter 1991)Google Scholar; Przeworski, Adam et al. , Sustainable Democracy (New York: Cambridge University Press, 1995), 1–39CrossRefGoogle Scholar.

⁵ Even though neither the major elite actors nor the public at large favored independence, Czechoslovakia certainly had a stateness problem because of disagreement over the appropriate form of the state.

⁶ Rustow, Dankwart, “Transitions to Democracy: Toward a Dynamic Model,” Comparative Politics 2 (April 1970)CrossRefGoogle Scholar.

⁷ Linz and Stepan (fn. 4, 1996), 26. See also Binder, Leonard, ed., Crises and Sequences in Political Development (Princeton: Princeton University Press, 1971)Google Scholar.

⁸ Roeder, Philip G., “Soviet Federalism and Ethnic Mobilization,” World Politics 43 (January 1991)CrossRefGoogle Scholar.

⁹ Transition theorists such as Guillermo O'Donnell and Philippe Schmitter distinguish between liberalization and democratization. Whereas the former represents a relaxation of regime constraints on discussion and organization, a process that could well be reversed, the latter involves opening the system to political competition and hence the possible displacement of the ruling elite. Of course, liberalization can always spin out of control and generate a democratic opening or transition, and indeed that is not an uncommon scenario. See O'Donnell, and Schmitter, , Transitions from Authoritarian Rule: Tentative Conclusions about Uncertain Democracies (Baltimore: Johns Hopkins University Press, 1986)Google Scholar.

¹⁰ See Gregory Gleason for a representative discussion of how to understand both the limitations and the significance of Soviet federalism. Gleason, , Federalism and Nationalism: The Struggle for Republican Rights in the USSR (Boulder, Colo.: Westview Press, 1991), 2–5Google Scholar.

¹¹ On republican elites, see, for example, Karklins, Rasma, Ethnic Relations in the USSR' The Perspective from Below (Boston: Unwin Hyman, 1989)Google Scholar; and Leff, Carol Skalnik, National Conflict in Czechoslovakia: The Making and Remaking of a State (Princeton: Princeton University Press, 1988)CrossRefGoogle Scholar.

¹² All three states were marked by failed attempts to build an overarching common identity atop this ethnofederal structure: “Czechoslovak,” “Yugoslav” and “Soviet.”

¹³ Several important questions cannot be adequately addressed in the limited space of an article, notably the origins of ethnic conflict, a question that has been studied intensively elsewhere. The dynamics of federalism, of course, cannot be understood in isolation from the multinational complexity that encouraged a federal solution in the first place.

¹⁴ Vladimir Goati so depicts the Yugoslav case in passing. See Goati, , “The Challenge of Post- Communism,” in Seroka, Jim and Pavlovic, Vukasin, eds., The Tragedy of Yugoslavia (New York: M. E. Sharpe, 1992), 21Google Scholar.

¹⁵ Woodward, Susan, Balkan Tragedy: Chaos and Disintegration after the Cold War (Washington, D.C.: Brookings Institution, 1995), 77Google Scholar.

¹⁶ Because the democratization literature uses the term “liberalization” to connote an opening for freer discussion and critique of regime policy, I will continue to do so here, despite the ambiguities.

¹⁷ Restrictions on citizenship rights in the Baltics after independence in 1991 are the exception.

¹⁸ The absence in Czechoslovakia of a pretransition liberalization comparable to those in Yugoslavia and the Soviet Union is a legacy of the “normalization” period following the Soviet invasion of 1968. The leadership installed in 1969 to restore order remained in control until the Velvet Revolution, stringently policing the political agenda and reacting with profound malaise to the Gorbachev model, which bore a distressing resemblance to the limited agenda of the Prague Spring itself. Unlike Gorbachev, the incumbent leadership was unequivocally responsible for the earlier “era of stagnation”—its very survival in power depended on forestalling any serious political opening that would have allowed the articulation of the Slovak question and latent discontent over the communist federal model launched in 1968. See Leff, Carol Skalnik, “Soviet-Czechoslovak Relations in the Gorbachev Era,” in Staar, Richard F., ed., Soviet-East European Relations (New York: St. Martin's Press, 1991)Google Scholar. The synchrony of liberalization and democratization had two related effects. First, the abrupt onset of transition left the counterelite without a coherent program for transformation and with only scant time and resources to formulate one, on the national question or on any other for that matter. The second and related effect of telescoping the agendas was a degree of unjustified optimism about the ease with which the Slovak question might be resolved and a consequent rude shock as divergences in perspective unfolded. A prior liberalization period would have clarified these issues and perhaps even have invited preemptive action to forestall the conditions that later produced a stalemate. At the same time, however, experience elsewhere suggests that such a period would also have led the parties to dig in their heels.

¹⁹ At worst, they had no sanctioned institutional framework whatsoever, as was the case with the ethnic Turkish minority in Bulgaria, victims of a repressive assimilation campaign in the 1980s that stripped them of educational facilities, religious institutions, and even their Turkish names.

²⁰ Examples include the catalyzing effects on popular protest of the eviction of a Hungarian clergyman from his home in Romania and undermining the prestige of the Bulgarian communist regime as a result of the fiasco of the Turkish mass exodus in the summer of 1989.

²¹ Hayden, , “Constitutional Nationalism in the Formerly Yugoslav Republics,” Slavic Review 51 (Winter 1992)CrossRefGoogle Scholar.

²² Sartori, Giovanni, “European Political Parties: The Case of Polarized Pluralism,” in Palombara, Joseph and Weiner, Myron, eds., Political Parties and Political Development (Princeton: Princeton University Press, 1966)Google Scholar.

²³ See, for example, Hesli, Vicki L., “Political Institutions and Democratic Governance in Divided Societies,” in Grey, Robert D., ed., Democratic Theory and Post-Communist Change (Upper Saddle River, N.J.: Prentice Hall, 1997), 198–201Google Scholar.

²⁴ Tellingly, in each of these cases the dominant national groups, Russian and Czech, had no party organization of their own independent from the statewide communist party.

²⁵ Lytle, , “Electoral Transitions in Yugoslavia,” in Shain, Yossi and Linz, Juan, eds., Between States: Interim Governments and Democratic Transitions (Cambridge: Cambridge University Press, 1995), 239–40Google Scholar. See also Woodward (fn. 15), 21–46.

²⁶ Lenard Cohen points out that broadening the debate to encompass taboo national questions simultaneously narrowed the debate in other respects: the orthodoxy of nationalism tended to replace the orthodoxy of communism. See Cohen, , Regime Transition in a Disintegrating Yugoslavia: Laio-of-Rule vs. the Rule of Law, Carl Beck Papers, no. 908 (Pittsburgh, Pa.: University of Pittsburgh Center for Russian and East European Studies, 1992)Google Scholar.

²⁷ Ibid., 9–12.

²⁸ Lytle (fn. 25), 237.

²⁹ In the Soviet case this was a consequence of the limits on competition of the statewide elections of 1989 and of Gorbachev's decision to occupy the newly created Soviet presidency through parliamentary action rather than popular election.

³⁰ In Macedonia elite preferences shifted from confederation to independence with the defection of Slovenia and Croatia.

³¹ Whitehead, Laurence, “Three International Dimensions of Democratization,” in Whitehead, , ed., The International Dimensions of Democratization: Europe and the Americas (New York: Oxford University Press, 1996)Google Scholar.

³² Kitschelt, for example, suggests that we might regard Slovakia as having had a patrimonial regime, in contrast to the Czech bureaucratic authoritarian communism. Kitschelt, Herbert, “Formation of Party Cleavages in Post-Communist Democracies: Theoretical Propositions,” Party Politics 1 (October 1995)CrossRefGoogle Scholar.

³³ Scholars have questioned whether the international community was too ready to accept bargaining failure in Yugoslavia, that is, whether there should have been greater international pressure for a negotiated dissolution. Indeed, the consequences do suggest that asymmetrical democratization alone was not a viable policy.

³⁴ The attentiveness of external actors also constrained attempts by the Soviet central government to bring recalcitrant republics to heel; both at home and in Eastern Europe, Gorbachev was to some extent hostage to his own calls for more responsive governance and therefore unable to send out the tanks without considerable cost to the credibility of his international agenda.

³⁵ Whitehead (fn. 31), 15–21.

³⁶ This characterization corresponds to the valuable point made by Linz and Stepan about the importance of state capacity for democratization (and indeed, for all effective government): the state needs a monopoly of the effective use of force and the “effective capacity to command, regulate and extract.” Linz and Stepan (fn. 4, 1996), 11.

³⁷ Woodward (fn. 15), 69–74.

³⁸ See Cohen (fn. 26).

³⁹ Lijphart, Arend, “Democratization and Constitutional Choices in Czechoslovakia, Hungary and Poland, 1989–1991,” Journal of Theoretical Politics 4, no. 2 (1992)Google Scholar.

⁴⁰ Prague Television, November 4, 1990, in FBIS-EEU-90–216, November 7, 1990, p. 17.

⁴¹ See, for example, Tumovec, Frantisek, The Political Background of Economic Transformation in the Czech Republic, Discussion Paper no. 25 (Prague: Center for Economic Research and Graduate Education, Charles University, November 1993)Google Scholar.

⁴² The 1992 elections in the Czech and Slovak Republics produced the legislatures and governments that retained power in the new states until the elections of 1996 and 1994, respectively. The first Russian elections were held two years after independence, in 1993, and the multitude of post-Yugoslav elections also followed six months or more after independence.

⁴³ The Spanish case also differed in the relative weight of the substate challenge. Alternatives to the core Spanish identity were pressing only in the Catalon and Basque regions and not statewide.

⁴⁴ It is suggestive that in the unitary First Czechoslovak Republic (1918—38), the party system did indeed resemble Spain's, a hybrid of regionally based and statewide parties. The Second Czechoslovak Republic (1945—48) that provided for a Slovak legislature saw the complete segmentation of the party system and the disappearance of parties with statewide constituencies. David Olson has noted though that postcommunist laws governing elections to the Czechoslovak Federal Assembly privileged republic boundaries in two ways: no voting district boundaries crossed republic lines (an administrative as well as political choice that held in Yugoslavia and the USSR as well), and threshold requirements applied at the republican level but not at the state level. See Olson, , “The Sundered State: Federalism and Parliament in Czechoslovakia,” in Remington, Thomas F., ed., Parliaments in Transition: The New Legislative Politics in the Former USSR and Eastern Europe (Boulder, Colo.: Westview Press, 1994), 111Google Scholar.

⁴⁵ Historian John Lampe sees the critical point of no return as occurring much earlier, in the 1970s, when greater opening of the system might have reinforced public identification with the statewide system. See Lampe, , Yugoslavia as History: Twice There Was a Country (Cambridge: Cambridge University Press, 1996), 294Google Scholar.

⁴⁶ For an elaboration of the organizational difficulties, see Remington (fn. 44), 8–10.

⁴⁷ Linz and Stepan(fn. 4, 1996), 232.

⁴⁸ For a fuller analysis of this question, see Leff, , “Inevitability, Probability, Possibility: The Disintegration of Czechoslovakia” (Conference on the Dissolution of Czechoslovakia, CERG-IE, Prague, Charles University, June 27—29, 1996)Google Scholar. Many conference participants seemed skeptical that the federal government had the leeway to change the constitutional rules of the game; they focused instead on the short, two-year time frame mandated for the first parliament, which doubled as the constituent assembly.

⁴⁹ See Skilling, H. Gordon, Czechoslovakia's Interrupted Revolution (Princeton: Princeton University Press, 1977), 481–84Google Scholar.

⁵⁰ Although this article is primarily concerned with the process by which the three communist federations reached endgame, a brief comment is necessary on the much-discussed role of violence in state dissolution. It is less the strategic context of federal structures as such that dictated differential levels of violence than the territorial ethnonational distribution of the populations of the three states and the constructed historical meaning attached to that distribution. Yugoslavia and the Soviet Union are most clearly differentiated from Czechoslovakia by their more politically volatile ethnic intermixture. Since Czechs and Slovaks were territorially concentrated within their own republican boundaries, it was possible to dissolve the state without creating significant orphaned and threatened minorities stranded on the “wrong side” of new borders. This pattern of national population distribution reinforced the institutional divisions of the state but did not burden the dissolution with custody battles over the security of stranded minorities as in Yugoslavia and the USSR, both of which suffered from substantial crossboundary ethnic complications. This single factor is the most important in accounting for the presence or absence of violence acccompanying their breakups.

⁵¹ In Yugoslavia, Lampe (fn. 45) sees even this national assertion as “a substitute for democratic polities' (p. 293).

⁵² Pei, , “‘Creeping Democratization’ in China,” Journal of Democracy 6 (October 1995), 77—78CrossRefGoogle Scholar.

⁵³ The economic interests of the key republican elites have, of course, been intensively examined in the literature, with emphasis on the differential responsiveness to the dictates of marketization and on the relevance of central coordination to the effective management of economic liberalization and stabilization measures. Here, I have focused primarily on the politics of republican defection from central economic coordination.

⁵⁴ See especially Munck, Gerardo and Leff, Carol Skalnik, “Modes of Transition and Democratization: The South American and East European Cases in Comparative Perspective,” Comparative Politics 29 (April 1997)CrossRefGoogle Scholar.

⁵⁵ See, for example, Roeder, Philip, Red Sunset: The Failure of Soviet Politics (Princeton: Princeton University Press, 1993)Google Scholar; and Richard Snyder and James Mahoney, “The Missing Variable: Institutions and the Study of Regime Change,” Comparative Politics (forthcoming).