Document Details
Other Publication
Publication Year:
Federation as a Method of Ethnic Conflict Regulation

Federation as a Method of Ethnic Conflict Regulation1 John McGarry and Brendan O‘Leary Federalism is a normative political philosophy that recommends the use of federal principles, i.e. combining joint action and self-government (King 1982). ”Federal political systems‘ is a descriptive catchall term for all political organizations that combine what Daniel Elazar called ”shared rule and self-rule‘. Federal political systems, thus broadly construed, include federations, confederations, unions, federacies, associated states, condominiums, leagues, and cross-border functional authorities (Elazar 1987). Federations are very distinct federal political systems (Watts 1998; Watts 1987). In a genuinely democratic federation there is a compound sovereign state, in which at least two governmental units, the federal and the regional, enjoy constitutionally separate competencies – although they may also have concurrent powers. Both the federal and the regional governments are empowered to deal directly with their citizens, and the relevant citizens directly elect (at least some components of) the federal and regional governments. In a federation the federal government usually cannot unilaterally alter the horizontal division of powers: constitutional change affecting competencies requires the consent of both levels of government. Therefore federation automatically implies a codified and written constitution, and normally is accompanied at the federal level by a supreme court, charged with umpiring differences between the governmental tiers, and by a bicameral legislature œ in which the federal as opposed to the popular chamber may disproportionately represent the smallest regions. Elazar emphasised the ”covenantal‘ character of federations, i.e. the authority of each government is derived from a constitution and convention rather than from another government. Federations vary in the extent to which they are majoritarian in character, but most constrain the power of federation-wide majorities. They constrain the federal demos, though there is extensive variation in this respect (Stepan 2001: 340-57). The United States, Australia and Brazil allow equal representation to each of their regions in the federal chamber, which means massive over-representation for the smaller ones. Other federations also over-represent less populous units, but not to this extent. Federations differ additionally in the competences granted the federal chamber. Some, such as the US Senate are extremely powerful, and which is arguably more powerful than the House of Representatives because of its special powers over nominations to public office and in treaty-making; others, including those in Canada, India, and Belgium are weak (Watts 1999: 93-4). Constitutional change can be blocked by individual regions in some instances, although normally a veto requires a coalition of regions. A federation is majoritarian to the extent that it lacks consociational practices of executive power-John McGarry is the Canada Research Chair in Nationalism and Democracy in the Dept. of Political Studies, Queen‘s university, Ontario (email Brendan O‘Leary is the Lauder Professor of Political Science at the U. of Pennsylvania (email: A number of organisations have funded our joint and individual research. Both of us thank the United States Institute of Peace. John McGarry thanks the Social Sciences and Humanities Research Council of Canada and the Carnegie Corporation of New York. sharing, proportionality principles of representation and allocation, cultural autonomy and veto-rights; and it is majoritarian to the extent that it lacks consensual institutions or practices œ such as the separation of powers, bills of rights, and courts and monetary institutions insulated from immediate governing majorities. A majoritarian federation concentrates power-resources at the federal level and facilitates executive and legislative dominance either by a popularly endorsed executive president or by a single party premier and cabinet. The federal principle of separate competencies says nothing about how much power each level enjoys. Regions in some federations may enjoy less de facto power than those in decentralized unitary states. The constitutional division of powers (even as interpreted by the courts) is not always an accurate guide to policy-making autonomy and discretion enjoyed by different tiers. Some powers may have fallen into abeyance, or the superior financial and political resources of one level (usually the federal) may allow it to interfere in the other‘s jurisdiction. A better indicator of the degree of autonomy enjoyed by regions may be the proportion of public spending that is under the control of the respective levels (for such measurements see Watts 2001: 29 and Lijphart 1979: 504). A key distinction for our purposes is that federations can be multi-national/multi-ethnic or mono-national in character. In the former, the boundaries of the internal units are usually drawn in such a way that at least some of them are controlled by national or ethnic minorities. In addition, more than one nationality may be explicitly recognized as co-founders and co-owners of the federation. The first such federation was Switzerland, established in its current form in 1848, and the second, Canada, established in 1867. The Indian subcontinent was divided after decolonization into the two multi-ethnic federations of India and Pakistan. Africa has two federations, Nigeria and Ethiopia, while South Africa appears federal in all but name. The communist Soviet Union, Yugoslavia and Czechoslovakia were organized as multi-national federations, and the Russian Republic (RSFSR), one of the constituent units of the Soviet Union, was itself organized along federal lines. These communist federations did not bestow genuine democratic self-government on their minorities, and fell apart in the early 1990s, although Yugoslavia continued as a dyadic federation incorporating Serbia and Montenegro until 2003, when it was transformed into a confederation renamed Serbia and Montenegro that looked likely to dissolve into two independent states. Bosnia became a multi-national federation under the internationally enforced Dayton Agreement of 1995, with one of its units itself being another bi-national federation of Bosniacs and Croats. Belgium has recently evolved into a federation, and both Euro-optimists and pessimists think that the European Union (EU) is moving in the same direction. Multinational federations have been proposed for a significant number of other divided societies, including Afghanistan,2 Burma, China, Cyprus, Georgia, Iraq and Indonesia.3 2 See Multi-ethnic federalism was once suggested for the United States as a way of giving self-government to southern Blacks. The idea of creating a black state in the South was supported by the American Communist Party in the 1930s and various black power organisations in the 1960s. It is no longer discussed, primarily because of black migration into northern cities (Glazer, 1977: 74). National federations may be nationally or ethnically homogeneous (or predominantly so), or they are organized, often consciously, so as not to recognize more than one official nationality — often this happens in such a way that the state‘s national and ethnic minorities are also minorities in each of the constituent units. The intention behind national federalism is nation-building, the elimination of internal national (and perhaps also ethnic) differences. The founding and paradigmatic example of a national federation is the United States. Its model was adopted by the Latin American federations of Mexico, Argentina, Brazil and Venezuela. Germany, Austria, Australia, and the United Arab Emirates are also national federations. American and American-educated intellectuals often propose national federations as a way to deal with ethnic heterogeneity in postcolonial and post-communist societies. Table 1. Examples of National and Multinational Federations National federations Duration Multinational Federations Duration Argentina 1853 -Belgium 1993 -Australia 1901 -Bosnia 1995 -Austria 1920 -Burma 1948 -Brazil 1891 -Cameroon 1961 œ 1972 Germany 1949 -Canada 1867 -Mexico 1917 -Czechoslovakia 1968 œ 1992 United Arab Emirates 1971 -Ethiopia 1992 -United States 1789 -India 1947(50) -Venezuela 1960 -* Malaya 1957 œ 1963 Malaysia 1963 Mali 1960 œ 1960 Nigeria 1960 -Pakistan ** 1947 œ 1971 Russia 1993 South Africa 1996 -Soviet Union 1918 œ 1991 St. Kitts-Nevis 1983 Switzerland 1848 West Indies Federation 1958 œ 1962 Yugoslavia/Serbia and Montenegro 1992 -*** Yugoslavia (Former) 1953 œ 1992 * Venezuela abolished its Senate in 1999. ** Pakistan (before the secession of Bangladesh). *** Yugoslavia (Serbia and Montenegro) was transformed into the confederal union of Serbia and Montenegro in February 2003. Federations can also be distinguished according to their level of democracy. Some, such as Canada, the United States, and Belgium should be seen as maturely democratic; others, such as Malaysia and Nigeria as partially democratic; and others, such as the communist federations of the Soviet Union, Yugoslavia and Czechoslovakia as not democratic. There is an increasingly popular view in the academic literature on federalism that this distinction is unimportant. A number of prominent American academics thus interpret the failings of the communist federations as an indictment of (multi-national) federalism per se (Brubaker 1996; Bunce 1999; Leff 1998; and Roeder 1991). In these accounts, it is structure that counts. One of our arguments, however, is that democracy matters, crucially, as does the type of democratic system. This chapter is primarily concerned with multi-national and multi-ethnic federations because we regard national federations largely as devices associated with integrationist or assimilationist politics œ which are not the subject of this volume. We shall first discuss the debate on the value and feasibility of federations as management devices for ethnic and national differences, then the track-record of multi-national federations in mitigating conflict, and conclude with an analysis of the factors that contribute to their success and failure. The Debate: Nationalism and Federalism in Practical Political Design and Argument There are four important positions on the value of federalism and federation as a method of accommodating national and ethnic minorities, all of which have been articulated by intellectuals, constitutional lawyers and political scientists, and have had an effect on the design of particular states. 1. Jacobin Unitarism: Federalism as State-Destroying. In the French revolutionary tradition, associated with the Jacobins, federalism was part of the counter-revolution, hostile to the necessity of linguistic homogenisation, a road-block in the path of authentic, indivisible, monistic popular sovereignty. Rather than accommodating minorities through self-government, the Jacobins sought cultural assimilation; they were determined to make peasants into Frenchmen; and therefore they were deeply hostile to all forms of accommodation that inhibited this goal, including federalism. The Jacobin response to diversity was a strong unitary state and one French nation. This tradition survives in contemporary France, where it is central to the myth of the French Republic (Hayward 1993).4 Federalism, with its multiple governments, is seen by those in the Jacobin tradition as incompatible with equal citizenship and a sovereign people. This is not just a concern about regional governments creating uneven (”patchwork quilt‘) public policy provisions. Latter-day Jacobins cannot accept the federal principle that allows citizens in regions with small populations to be over-represented at the expense of those in more 4 Article 2 of the Fifth Republic‘s constitution declares that ”France is an indivisible, secular, democratic, and social Republic‘. populous regions, and they have difficulty with the federal idea of a judicial umpire who can over-rule the people‘s elected representatives. Both facts explain the reported French astonishment at George Bush Jr. being elected U.S. president in 2000 with fewer popular votes than his opponent (a result in part of the disproportionality inherent in the Electoral College, and a partial by-product of the US‘s federal system), and the incumbency being effectively decided by the federal Supreme Court. (Ferenczi 2001). Modern Jacobins think that the accommodation of minorities and ethnocentrism go together. If minorities do not want to promote ethnocentrism, the argument goes, why do they seek self-government? They think that political recognition of multiple nations or ethnic communities leads to regressive government and discrimination against minorities dominated by local regional majorities, and institutionalises and reinforces divisions, endangering national/state unity. These views are shared on the left and right. Communists claim that Paris‘s proposals to give self-government to Corsica will undermine ”solidarity between Corsican and French workers, who can only defend their interests by working together‘, and will lead to discriminatory measures against those on the island who are not of Corsican descent.5 The then French interior minister, Pierre Chevenement, resigned over the proposals, protesting that they would lead to an ”island ruled by an underworld that spends three-quarters of its energy settling accounts and internal battles‘.6 While the proposals for Corsica fall short of federation, both Chevenement and French president Jacques Chirac attacked them as leading in that direction: Brittany, Alsace, Savoy, as well as French Basques and Catalonians, would allegedly follow Corsica‘s lead (Ferenczi 2001: 42).7 Ultimately, in this view, federation promotes state break-up, with the attendant risks of ethnic cleansing and Matrioschkadoll secessions emerging as ethnic nationalism takes hold.8 The Jacobin view that unitarism is needed for unity, if not always its support for civic equality and popular sovereignty, is replicated throughout the world. It was the dominant view in Great Britain until recently, particularly among Conservatives. Most ex-colonies in Africa, Asia and the Caribbean have shunned federalism as an obstacle to economic development, political stability and state unity. Post-colonial state-builders‘ antipathy to federalism is now matched amongst the intellectuals and governing elites of Eastern Europe, who regard it as a recipe for disaster, given the Czechoslovakian, Yugoslavian and Soviet experiences. Federalism is their ”f‘ word. The recent emergent principle of international law, stemming from the report of the Badinter Commission on the former Yugoslavia, that permits the disintegration of federations along the lines of their existing regional units, may strengthen the belief that federation should not be considered a desirable form of multi-national or multi-ethnic accommodation (Horowitz 1998; Weller 1992). Several Eastern European states have moved in the opposite direction in recent years, replacing multi-national federations with what Brubaker calls —nationalizing“ 5 ”Partial autonomy for Corsica splits French government‘, World Socialist Web Site, 5 September 2000 ( 6 ”Corsica: The perils of devolution‘, Economist, July 7, 2001, p.49. 7 Chevenement claimed that self-government for Corsics would be as contagious as the ”I love you‘ computer virus, ”Partial autonomy for Corsica splits French government‘, World Socialist Web Site, 5 September 2000 ( states, that is, states that are tightly centralized and controlled by their dominant national community. Ironically, the Jacobin argument that federalism is incompatible with nation-building is shared by ”hard-line‘ nationalists trapped inside states controlled by other nations. They concur that nation and state should be congruent, although they disagree on the appropriate boundaries. This has been the position of Quebec‘s Parti Quebecois, particularly the faction around the ex-Premier Jacques Parizeau, and of Basque Nationalists in Herri Batasuna. It is also the view of the Turkish Cypriot leadership under Rauf Denktash, the Chechens, and, until very recently, the Liberation Tigers of Tamil Eelam. Such hard-liners seek independence as unitary, sovereign and indivisible nationstates, although some are prepared to consider confederation.9 2. Federalism as Nation-Building. Unlike the Jacobins, who see (state) nationalism and federalism as inconsistent, some exponents of federalism think that (state) nationalism and federalism go together. The earliest federalists in the German-speaking Swiss lands, and in what became the Netherlands, the USA, and the second German Reich were ”national federalists‘, i.e. they saw the prime function of federalism as being ”to unite people living in different political units, who nevertheless shared a common language and culture‘ (Forsyth 1989: 4). They maintained that only an autonomous federal government could perform certain necessary functions that confederations or alliances found difficult to perform, especially a unified defence and external relations policy (Riker 1964). They advocated federation as a tool for nation-building and sometimes saw it as a stepping stone towards a more centralised unitary state. The USA is the paradigm case of national federalism. Americans have little difficulty with what Jacobins consider the ”demos-constraining‘ features of federalism: radical autonomy for regions or states (non-centralization); the over-representation of small states in upper chambers, electoral colleges and constitutional amending formulas. In fact, Stepan has argued that the United States competes with Brazil for the title of the world‘s most demos-constraining federation (Stepan 2001: 334). The attractiveness of demos-constraining institutions reflect the historic stress of some American on liberty rather than equality. The American founding myth is of colonies that won independence from empire. Many Americans reject the strong state favoured by French republicans and praise federalism precisely because it diffuses power to multiple points. American exponents of federalism, such as Riker, have argued that the demos-constraining features of American federalism are liberal because they protect individuals from populist majorities (Riker 1964). Americans insisted on a federation for post-war Germany, because they were convinced it would make a resurgence of fascism less likely. The view that federalism is essential to liberty is central to American discourse, in spite of the abysmal track-record of federalism in all of the Latin American federations, as well as in Pakistan, Nigeria and the USSR. Confusingly, hardline minority nationalists sometimes say they support federation when they mean confederation, as in the case of the Turkish Cypriot leader, Rauf Denktash. The Parti Quebecois does not commonly use the term confederation, but offers a synonym, ”sovereignty-association‘ But America‘s makers and their celebrants have taken the position that federalism is antithetical to nation-building if it is multi-national, multi-ethnic, or ”ethnofederal‘. As the United States expanded southwestward from its original largely homogeneous (except for African slaves) thirteen colonies, it was decided that no territory would receive statehood unless minorities were outnumbered by White Anglo-Saxon Protestants (WASPs) (Glazer 1983). Sometimes, the technique employed was to gerrymander state boundaries to ensure that Indians or Hispanics were outnumbered, as in Florida. At other times, as in Hawaii and the southwest, statehood was delayed until the region‘s longstanding residents could be swamped with enough WASP settlers. American authorities were even skeptical of immigrant groups concentrating in particular locations lest this lead to ethnically-based demands for self-government, and grants of public land were denied to ethnic groups in order to promote their dispersal; William Penn dissuaded Welsh immigrants from setting up their own self-governing barony in Pennsylvania (Gordon 1964: 133). In consequence, the US federation shows ”little coincidence between ethnic groups and state boundaries‘ (Glazer 1983: 276). National federalism was part and parcel of American nation-building, aiding the homogenisation of white settlers and immigrants in the famous melting pot of Anglo conformity (Gordon 1964). Celebration of the homogeneity of the founding people is evident in The Federalist Papers (see especially John Jay‘s assumptions in Madison et al 1788/1987: paper II).10 America‘s experience with federalism has informed an interesting argument on how federalism can be used to manage divisions in contemporary ethnically heterogeneous societies. Donald Horowitz (1985: chs. 14 and 15), and Daniel Elazar (Elazar 1994: 168), building on earlier work by S.M. Lipset (1960)11, and indeed, on an important American tradition that goes back to James Madison, suggest that federations can be partly designed to prevent ethnic minorities from becoming local provincial majorities. The strategic thinking here is to weaken potentially competing ethno-nationalisms: federalism‘s territorial merits are said to lie in the fact that it can be used as an instrument to prevent local majoritarianism (which its attendant risks of local tyranny tyrannies of the majority, or of secessionist incentives). The provincial borders of the federated units on this argument, should be designed on ”balance of power‘ principles œ proliferating, where possible, the points of power away from one focal centre, encouraging intra-ethnic conflict, and creating incentives for inter-ethnic co-operation (by designing provinces without majorities), and for alignments based on non-ethnic interests. This logic is interesting, but empirical support for it seems so far confined to the rather uninspiring 10 ” Providence has been pleased to give this one connected country to one united people – a people descended from the same ancestors, speaking the same language, professing the same religion, attached to the same principles of government, very similar in their manners and their customs, and who, by their joint counsels, arms and efforts, fighting side by side throughout a long and bloody war, have nobly established liberty and independence.’ Publius [John Jay], in Madison et al. 1987 [1788], paper II: 91 11 Lipset (1960: 91-92) argues that the main benefit of federalism for divided societies is that it creates cross-cutting cleeavages, but it can only do this if internal federal boundaries and ethnic boundaries intersect. Federalism ”increases the opportunity for multiple sources of cleavage by adding regional interests and values to the others which crosscut the social structure‘. case of post-bellum Nigeria (Suberu 2001:4-6). In most existing federations to re-draw regional borders deliberately to achieve these results would probably require the services of military dictators or one-party states. Historically mobilised ethno-national groups do not take kindly to efforts to disorganise them through the re-drawing of internal political boundaries. Belgium may, however, become an interesting exception to this scepticism: the Brussels region, created in the new federation, is neither Flemish or Wallonian, and perhaps its heterogeneity will stabilise ”inter-national‘ relations in Belgium, because Flanders will not secede without Brussels and there is presently little prospect of Brussels obliging Flanders. American republicans, with a small ”r‘, have shared with Jacobins the view that minority nationalists are backward, a ”revolt against modernity‘(Lipset 1985) or people who ”tend to subordinate all free government to [their] uncompromising position‘ (Elazar 1994: 128-9, 163-4). They think that it is both counterproductive and unnecessary to accommodate minority nationalists. This view may have been strengthened by America‘s own experience, in the Deep South, of southern whites using their control of state governments to oppress blacks. America‘s experience with a disastrous civil war over secession attuned its intellectuals to the centrifugal potential of federalism, particularly when regions are controlled by distinct cultural communities. Eric Nordlinger, one of the first contemporary American political scientists to take an interest in ethnic conflict regulation, rejected the use of federalism as an instrument for accommodating minorities as he feared it would lead to state break-up and the abuse of power by ethnocentric minorities (Nordlinger 1972: 32-33; See also Tarlton 1965). Reflecting these sentiments, a number of American academics have argued that the break-up of the former communist federations and the accompanying chaos, can be traced squarely to ”ethno-federal‘ structures (Brubaker 1996; Bunce 1999; Leff 1998; and Roeder 1991). Rogers Brubaker maintains that the Soviet regime went to ”remarkable lengths, long before glasnost and perestroika, to institutionalize both territorial nationhood and ethnocultural nationality as basic cognitive and social categories. Once political space began to expand under Gorbachev, these categories quickly came to ”structure political perception, inform political rhetoric, and organize political action‘ (Brubaker 1996: 9). The implication is that (at least some of) these divisive identities did not exist before the Soviet Union federated and would not have come into play had it not federated. In Jack Snyder‘s view, ”ethnically based federalisms … create political organizations and media markets that are centered on ethnic differences‘ (our italics). According to him, the decision to establish ethnofederations in the Soviet Union, Czechoslovakia, and Yugoslavia was unnecessary (Snyder 2000: 327, our italics): ”Arguably, ethnofederalism was a strategy of rule actively chosen by its Communist founders not a necessity forced upon them by the irresistible demands of ethnic groups‘( ibid).12 The results of ethnofederalism in his view were straightforward: only the communist federations broke up and ”nationalist violence happened only where 12 Snyder says the decision to create ethno-federations was ”often‘ needless (p.327), but does not specify where it was needless and where it was not. The substance of the argument throughout his book suggests that he thinks it was always needless. …ethnofederal institutions channelled political activity along ethnic lines (USSR and Yugoslavia)‘ (2000: 252, our italics ). 3. Cosmopolitans, Federalism as a stage in Nation-transcendence . A third perspective holds that federalism is capable of dissolving all national allegiances, including minority and majority nationalisms. It comes in two different variants. The first is represented by several nineteenth century anarchic and liberal federalists, notably Joseph Proudhon and Carlo Cattaneo, who were resolutely hostile to nation-state nationalism (Majocchi 1991: 162), and by many twentieth century liberal federalists, notably within the European movement (see e.g. Bosco 1992: Part Three). Such federalists have been, and are, resolutely anti-nationalist, associating both state and minority nationalisms with ethnic exclusiveness, chauvinism, racism, and parochially particularistic sentiments. For them federalism belongs to an entirely different co-operative philosophy, one that offers a nonnationalist logic of legitimacy, and an antidote to nationalism rather than a close relative. This viewpoint was most clearly articulated by Pierre Trudeau – educated by Elie Kedourie at the LSE – before he became Canadian Prime Minister.13 Thinkers like Trudeau regard federalism as the denial of and the solution to nationalism, though occasionally they adopt the view that federalism must be built upon the success of nationalism which it then transcends in Hegelian fashion (Majocchi 1991: 161). In effect they echo Einstein‘s reported remark that nationalism is the measles of mankind. A different perspective was articulated by the Austro-Marxists, Karl Renner and Otto Bauer, who proposed it in the last days of the Habsburg empire (see e.g. Bauer 2000; Hanf 1991; Pfabigan 1991). For them nationalism had to be accommodated en route to a global socialist and communist order. They thought it was feasible to combine national autonomy in federal and consociational formats. Lenin and Stalin pressed their arguments, in an adapted format, into service in the Soviet Union. Federalism was to be used to offer a limited accommodation to minority nationalism, but solely towards the end of building a socialist society. Minorities were to be offered the fiction, but not the fact, of national self-government.14 While this policy was superficially similar to that of multi-national federalists (to be discussed below), Marxist-Leninists were, of course, formal cosmopolitans, committed to a post-nationalist global political order. However, pending the world revolution, they maintained that federal arrangements, ”national in form, socialist in content‘, were the optimal institutional path to global communism. 13 In an article entitled ”Federalism, Nationalism and Reason‘ Trudeau squarely associated federalism and functionalism with reason, nationalism with the emotions (Trudeau 1968 (1965). Trudeau‘s antinationalism was directed at what he considered ”ethnic‘ nationalism, particularly that associated with his home province of Quebec. He had less difficulty with civic nationalism, of the Canadian variety. For a view on Kedourie, see O‘Leary 2002. 14 In the authoritative words of Walker Connor, Lenin‘s second commandment on the management of nationalism was strategically machiavellian: ”Following the assumption of power, terminate the fact œ if not necessarily the fiction œ of a right to secession, and begin the lengthy process of assimilation via the dialectical route of territorial autonomy for all compact national groups‘ (Connor 1984: 38). Multi-National Federalists: Multi-National Maintenance Engineers. Multi-national or multi-ethnic federalists, by contrast, advocate federation ”to unite people who seek the advantages of membership of a common political unit, but differ markedly in descent, language and culture‘ (Forsyth 1989: 4). They seek to express, institutionalise, and protect at least two national or ethnic cultures, on a durable and often on a permanent basis. Any greater union or homogenisation, if envisaged at all, is postponed for the future. They explicitly reject the strongly integrationist and/or assimilationist objectives of national and or post-national federalists, and see these as nation-destroying rather than nation-building. They believe that dual or multiple national loyalties are possible, and indeed desirable. Multi-national federalists represent a third branch of liberalism, distinct from the Jacobin (federalism breeches civic equality), and American varieties (national federalism promotes individual liberty). For multi-national liberals, a proper understanding of liberal individual rights requires respect for the culture of individuals, and this means allowing minorities the power to protect and promote their culture (Stepan 1999: 31-32; Kymlicka 1995). Unlike unitarists and national federalists, multi-national federalists reject the a priori view that minority-controlled governments are more backward or illiberal in their treatment of their own minorities than majority-controlled central or federal governments. Minority nationalisms are as likely to be of the civic variety as dominant nationalisms according to these liberals; indeed Keating argues that contemporary minority nationalisms are strongly modernist, responding to the shift in power from the state to the global marketplace (Kymlicka 1995; Keating 2001). Multi-national federalism has considerable, albeit critical support, among contemporary academics (Hechter 2000, Linz 1997, Keating 2001, Kymlicka 1995, Moore 2001, Stepan 1999, Watts 1999, McGarry and O‘Leary 1993, O‘Leary 2001). Some supporters make quite remarkable claims for federalism. Von Beyme, referring to western democracies, argued in 1985 that ”Canada is the only country in which federalism did not prove capable of solving … ethnic conflict‘ (von Beyme 1985: 121).15 Others are more modest: Kymlicka supports multi-national federalism normatively, while acknowledging it faces considerable difficulties in practice (Kymlicka, 2001). Multi-national federalists have been influential in the development of federations in the former British Empire, notably in Canada, the Caribbean, Nigeria, South Africa, India, Pakistan, and Malaysia. Austro-Marxists and even some Marxist-Leninists were multi-national federalists and have had an enduring impact in the post-communist development of the Russian Federation, Ethiopia, and the rump Yugoslavia. While unitarists have presently been in the ascendancy in eastern Europe, multi-national federalism has become more popular in western Europe, both amongst proponents of the federalisation of the European Union, and amongst power-holders in established states -as the decision to create a federation in Belgium attests (as perhaps do the novel and more decentralised devolutionary, regional and potentially federal institutions of Spain, the United Kingdom, France and Italy suggest). Multi-national federalists are often soft minority nationalists, but they also include state elites who believe that accommodating national minorities holds the key to stability and unity. They include the Quebec Liberal Party, the Basque Nationalist Party 15 The more normal claim is that Canada is the only country, or Canada and Switzerland are the only countries, where federalism has been successful in preventing conflict. (PNV), and the Catalan Convergencia I Unio. The most ambitious multi-national federalists of our day are those who wish to develop the European Union from its currently largely confederal form into an explicit federation, a ”Europe of the nationstates and a Europe of the citizens‘, as the German foreign minister recently urged (Fischer 2000).16 Plainly the multi-nationalists‘ defence of federation as a way of managing nations œ to each nation let a province be given -is not able to accommodate those minorities that are so small in number or dispersed, that they cannot control federal units or provinces. This includes francophones who live outside Quebec, Flemish-speakers in Wallonia, Francophones in Flanders; and small and scattered indigenous peoples in Australia, India and the Americas. Multi-national federalists reject the view that every minority must inevitably seek its nation-state, and maintain that even amongst those that do they may settle for their own region instead. They argue that if the provincial borders of the components of the federation match the boundaries of the relevant national, ethnic, religious or linguistic communities, i.e. if there is a ”federal society‘ congruent with the federating institutions, then federation may be an effective harmonising device. That is precisely because it makes an ethnically heterogeneous political society less heterogeneous through the creation of more homogeneous sub-units. Multi-national federalism thus involves an explicit rejection of the unitarist and national federalist argument that self-rule for minorities necessarily conflicts with the territorial integrity of existing states. It is also a prima facie challenge to the tacit Gellnerian notion that in modern times the equilibrium condition is one sovereign state, one culture (or nation) (Gellner 1983). If we treat broadly the ”political unit‘ in Gellner‘s definition, to encompass regional or provincial units in a federation, then his theory can accommodate such arrangements, but at the significant concession of recognising that federal systems are compatible with dual and possibly multiple nationalities. National minorities within a multi-national federation often argue that they should have powers beyond those enjoyed by the federal units dominated by the national majority: They support asymmetrical federalism, insisting that their distinct status be officially recognised and institutionalised. They may seek to share in powers that are normally the prerogative of the centre or federal government: some minorities seek a role in federal foreign policy, or to be directly represented in international organizations. This may not mean the same as supporting confederation, because the minorities may be content for 16 This model of federalism appears to be limited to the accommodation of the European Union‘s current member-states, and says nothing about minorities within these states. Contrast this with Michael Keating‘s version of Euro-federalism, which anticipates complex and overlapping levels of governance that includes regions (including minority regions), states, and the European Union (Keating 2001: 4). most purposes to remain part of a federation, but they are clearly stretching the limits of traditional federations, and moving in the direction of confederation.17 Multi-national federations may originate from the union of previously self-governing ethnic communities, as happened in the case of Switzerland. However, in other cases, multi-national federalists may engage in deliberate democratic engineering to match certain ascriptive criteria with internal political borders. This occurred at Canada‘s founding, when the province of Canada was divided largely along linguistic lines into Ontario and Quebec. It also happened in post-independence India, but not until Nehru was forced to concede re-organisation of internal state borders along linguistic boundaries (Arora and Verney 1995; Brass 1990). 18 Nigeria has re-organised its internal boundaries on several occasions, to the advantage of certain minorities. Whereas its original tri-partite federation was dominated by the Ibo, Hausa and Yoruba groups, its current thirty-six state structure includes fourteen states that are dominated by other groups (Suberu 2001: 5). Switzerland carved a new canton of Jura (largely French and Catholic) out of the mostly German-speaking canton of Berne in 1979. Weighing the Evidence There is considerable evidence for the French and American republican argument that multi-national federalism has, as Snyder puts it, ”a terrible track record‘ (2000: 327). Multi-national or multi-ethnic federations have either broken down, or have failed to remain democratic, throughout the communist world, and throughout the post-colonial world. The federations of the Soviet Union, Yugoslavia and Czechoslovakia disintegrated during or immediately after their respective democratisations. Indeed, of all the states in the former communist bloc of eastern Europe, it was only federations that irretrievably broke apart, and all of them did.19 Of all these states, the federations experienced the most violent transitions. In the post-colonial world multi-national or multi-ethnic federations failed, or failed to be successfully established, in the Caribbean, notably in the West Indies Federation. Even the miniature federation of St. Kitts-Nevis recently faced the prospect of break-up (Premdas 1998). Multi-national or multi-ethnic 17 For an account that is sympathetic to the claims of national minorities for asymmetrical federation and for an international role, see Keating (2001) who supports multi-national federalism, but he rejects as too simplistic the view that minorities can be accommodated within traditional sovereign states. Instead, Keating argues that, particularly within the European Union, new post-sovereigntist institutional arrangements are taking shape in which national minorities seek to play within several different fora œ the state, the (transborder) region, the European Union, and the world œ simultaneously. 18 Ironically, the redrawing of new boundaries to accommodate minorities is easier if the federal centre has more power. In India, the central government has been able to create new boundaries without the approval of the state governments concerned. In Canada, by contrast, the federal government is unable to alter boundaries without the consent of the affected provinces. It cannot even create a new province out of federal territories without the consent of existing provinces, which is one reason why the new region of Nunavut is a federal ”territory‘ rather than a province. 19 The latest victim may well be Yugoslavia (i.e. the rump Yugoslav federation of Serbia and Montenegro). In 2003, it was restructured into a looser union to be called Serbia and Montenegro. There is provision for Montenegrins to hold a referendum in three years on whether or not they want to stay part of this. federations have failed in sub-Saharan Africa, in Francophone West and Equatorial Africa, in British East Africa (Kenya, Uganda and Tanganikya), and in British Central Africa (Northern and Southern Rhodesia and Nyasaland), or have failed to remain durably democratic (Nigeria and Tanzania). The break-up of the Nigerian federation between 1966 and 1969 was only prevented after a secessionist conflict that caused approximately a million deaths. In the Arab world, only the United Arab Emirates has survived, but it is a national federation and hardly democratic. The Mali and the Ethiopian federations in independent Africa broke up too; while the Cameroons experienced forced unitarism after a federal beginning. In Asia there have been federative failures in Indochina, in Burma, in Pakistan (the secession of Bangladesh), and in the union of Malaya (the secession of Singapore). In short, new multi-national federations appear not to work as conflict-regulating devices – even where they allow a degree of minority self-government. They have broken down, or failed to be durably democratic, throughout Asia, Africa, and the Caribbean. India stands out as the major exception in Asia. It also seems clear that multi-national federations make it easier for groups to secede should they want to do so. Federalism provides the minority with political and bureaucratic resources that it can use to launch a bid for independence. Giving a minority its own unit makes it possible for it to hold referendums on secession, which can be useful for gaining recognition. Multi-national federations implicitly suggests the principle that the accommodated minorities represent ”peoples‘ who might then be entitled to rights of self-determination under international law. It is far more likely, as the Badinter Commission on the former Yugoslavia confirmed, that the international community will recognize a bid for independence from a federal unit than from a group that lacks such a unit. This is why all of the full constituent units of the Soviet Union, Yugoslavia and Czechoslovakia that broke away are now seen as independent states, whereas breakaway regions that were not constituent units, such as Abkhazia, Trans-Dniestria, the Turkish Republic of Northern Cyprus, and Kossovo, are not recognised. To this extent, unitarists and national federalists have a point – although it is a point that multi-national federalists have little difficulty conceding.20 However, this assessment of track record of multi-national federations has to be qualified in five important ways. First, the major federal failures, including the Soviet Union, Yugoslavia, Czechoslovakia, and Nigeria were or have been, to a significant extent, sham or pseudo-federations. In several cases, they were forced together. The constitutional division of powers and the rule of law were often ignored in practice and they were not authentically representative (i.e. democratic). There was, therefore, no possibility of genuine dialogue, never mind cooperation, among the different national communities involved. In sum, these states had weak or no overarching identities to begin with, and no democratic mechanism for developing them. While the United States can be seen as the paradigmatic example of national federalism, the Soviet Union is the most prominent 20 Stepan, who supports multi-national federalism and argues that the US (national) federal model has little relevance for multi-national socieites, concedes that the ”greatest risk‘ posed by federalism is that it can ”offer opportunities for ethnic nationalists to mobilize their resources‘. Stepan 1999: 19 case of pseudo-federalism. Territorially it consisted of those remnants of the Tsarist empire that the Red Army was able to subjugate after the October Revolution, plus those countries (Estonia, Latvia, Lithuania, and Moldova) it conquered as a result of the Ribbentrop-Molotov pact and its victory in the second world war. While its state structure was federated from the beginning, real power lay in the tightly centralized Communist Party (the CPSU), which operated according to the principle of ”democratic centralism‘ (Lieven and McGarry 1993). The Union Republics were therefore not autonomous in any meaningful way. Moreover, their legislatures (the Soviets), although in theory elected by local populations, were in fact rubber-stamp bodies nominated by the CPSU. Key institutions, including the army and police, were controlled by Moscow. No effective judicial review existed to decide on the division of rights and functional spheres between the centre and the republics. Yugoslavia was more decentralised than the Soviet Union or Czechoslovakia, at least after reforms in the late 1960s, but it was no less undemocratic, and was held together by the League of Communists. The colonial federations arose out of colonies that had been arbitrarily consolidated by white imperialists. However, even the decision to federate at independence was made in some cases by the departing metropolitan rather than the colony‘s indigenous elites. Nigeria‘s original three-unit federation, which collapsed in the mid-1960s, was ”bequeathed‘ by the vacating British (Suberu 2001: 4). The Cameroons federation was a construct of British and French colonialists (particularly the latter), who wanted to preserve the dual personality they thought they had created. (Elazar 1987: 240). The Cameroons were converted into a unitary state by military strongmen soon after independence, while Nigeria has been ruled by centralizing military dictators for more than two-thirds of its post-independence history œ and its presidnetail contenders in recent times have all been ex-generals. Even under democratic conditions, Nigeria is so centralised that it has been described as a ”hollow federation‘ and ”a unitary state in federal guise‘ (Suberu and Diamond 2000: 8).21 Corruption and abuse of power are so pervasive that the rule of law can hardly be said to exist (Suberu 2001).22 These communist and post-colonial federations were additionally burdened by economic systems that were incapable of providing a reasonable or growing standard of living for their citizens. In each case, this caused resentment, not least among minorities in relatively enterprising regions of the state who saw their inclusion in the federation as a drag on their enterprise. It was therefore surprising that when the communist planning system became discredited and collapsed in the late 1980s it produced a legitimacy crisis. 21 Nigeria‘s hyper centralism is a function of Abuja‘s control of oil revenues, but it also has a basis in the 1979 and 1999 constitutions. According to Joye and Igweike, under the new constitution (which largely copies the old one), there ”are few, if any…areas in which state governments can act independently of the Federal Government‘ (ited in Suberu and Diamond 2000: 15). The existence of such separate competencies, as we have pointed out, is an essential hallmark of federalism. 22 Unitarists often claim that decentralization leads to corruption and inefficiency, but contemporary Nigeria demonstrates that corruption and centralization can go hand in hand. Supporters of anticorrpuption reforms in Nigeria argue that this requires ”power and resources [to be] shifted downward, to levels of authority that are closer to the people and more visible‘. Diamond 2001: xviii. Second, the case against multi-national federalism would be stronger, if it could be shown, as critics claim, that it was unnecessary to accommodate national minorities, and that there were democratic civic nationalist (unitarist or national federalist) alternatives that would have worked better if not much better. Once this test is probed the critics‘ position looks less credible. The decision to create both the Soviet and Yugoslav federations was taken in the midst of bitter civil wars and external invasions, when parts of both states had seceded (Connor 1984, 198: Woodward 1995: 30). The decision was regarded as essential for restoring unity and luring breakaway regions back into the state, and was taken in both cases by socialist internationalists, neither of whom was ideologically committed to multi-national federalism. Before he assumed power, Lenin had expressed his vehement opposition to federalism and his clear preference for unitary structures.23 Tito, before taking power, appeared to be a conventional Leninist. If federalism was unnecessary, we must conclude that both Lenin and Tito were extraordinarily incompetent from their own perspectives. The thesis that communist multi-national federalism ”created‘ divisions cannot explain easily why strong ethnic identities exist among groups that were not accommodated through federal institutions, such as the Chechens or Crimean Tatars.24 Similarly, while some have argued that Nigeria‘s divisions at the time of independence reflected British divide and rule strategies, few think that the state could have been (or could be) held together without some form of decentralised or federal structure.25 When an Ibo leader, General Ironsi, tried to convert Nigeria into a centralist state in 1966 it led to his downfall. Even though the Nigerian federation witnessed a failed and bloody bid for secession in Biafra (196770), the victors were careful to retain ethno-federal structures, albeit reformed, with new internal boundaries. One reason to doubt the feasibility of civic nationalism, French or American, as an alternative to multi-national federation is that it has not been particularly successful when it has been applied, under more propitious circumstances, in multi-national states. Turkey still faces a large dissident Kurdish minority despite eight decades of ”Kemalist‘ civic nationalism. British civic nationalism within a tightly centralized union at the 23 In 1913, before he had responsibility for governing the Soviet Union, Lenin made clear his contempt for federalism and his preference for unitarism: ”We are in principle against federation. It weakens the economic connection and is inappropriate for a unified state. Do you want to separate? we say. Then go to the devil and cut yourself off altogether …You don‘t want to separate? Then, please, don‘t decide for me, don‘t believe you have the —right“ to federation‘ (italics and grammatical errors in original) (Connor 1984: 217). As Connor notes, Lenin dropped his opposition to federalism upon assuming power in order to ensure those nations that had seceded ”that reunion would not result in political subservience‘ (ibid: 218). 24 There is an explanation for this implicit in the arguments of critics of multi-nationalism federalism. It is that the decision to accommodate some national groups led those excluded to mobilise. We endorse this argument, but we think the way to deal with it would have been to accommodate the excluded identiy groups, not to refuse to accommodate any of them. 25 ”As the rivalries among these three groups [Yoruba, Ibo, Hausa] crystallized into bitter political struggles during the late colonial period…it became increasingly clear to all interested observers that only by some form of highly decentralized political arrangements could the main groups be accommodated within a single country‘ (Suberu 2001: 20). centre of a global empire could not prevent the break-away of Ireland in 1921 (McGarry 2001).26 Irish nationalists mobilized successfully without the advantages of their own self-governing institutions. They were able to establish democratic legitimacy without the need of a referendum, by winning the overwhelming majority of Ireland‘s seats in every election between 1885 and 1918. Britain‘s civic and unitary state proved incapable of preventing a nationalist rebellion in Northern Ireland from the late 1960s, or of preventing the resurgence of Scottish and Welsh nationalism. Even the home of Jacobinism that was able to turn peasants into Frenchmen in the nineteenth-century has been unable to erode Corsican nationalism in the late twentieth century. The failure of unitarist or national federalist forms of civic nationalism may explain why all western multi-national democracies, including the United Kingdom, Spain, Belgium, France, and Denmark are now more disposed towards decentralized autonomy regimes if not fullblown multi-national federalism. Third, if one accepts that federalism was necessary in the failed federations, the focus of blame for the violence accompanying their break-up can be shifted from multi-national federation per se. To some extent one can argue that secession — and violence —followed from attempts by certain groups to centralize these federations, i.e. to move away from the spirit of multi-national federalism. Yugoslavia‘s break-up, including the de-facto breakaway of Kossovo, followed successive Serbian-dominated moves against the autonomy of Yugoslavia‘s republics.27 The Soviet Union broke up after an abortive right-wing coup aimed at repudiating Gorbachev‘s decentralizing initiatives. Violence was also caused by the centre‘s unwillingness to permit secession, i.e. one can argue that federal constitutions with procedural and negotiable secession rules might have avoided violence better).28 There was no violence in Czechoslovakia because mutual secession was agreed. In the territory of the former Soviet Union, the worst violence was (is) in Chechenya, a region that did not enjoy the status of a ”union republic‘ within the Soviet Union. Had it done so, it would likely have seceded with the other republics, and with as little violence as most of them. In many cases, one might argue that post-communist violence resulted from the absence of ethnofederalism, i.e. from the lack of congruency between constituent unit and ethnic boundaries. In the case of Yugoslavia, Slovenia‘s secession was relatively peaceful, because it was homogeneous. The ”velvet divorce‘ in Czechoslovakia was facilitated because there were few Czechs in Slovakia and few Slovaks in the Czech lands.29 War started in Croatia in 1991 largely because Croatia had a significant Serb population that wanted to stay united with Yugoslavia, and spread to 26 Supporters of civic nationalism might respond that British (or other forms of) civic nationalism were not neutral between the United Kingdom‘s diverse peoples, and that a more genuinely inclusive version of civic nationalism might have worked. This is indeed part of the weakness of civic nationalism. It often reflects the values and interests of the state‘s dominant national community. 27 The Kossovo rebellion of 1997 was a response, albeit delayed, to the Milosevic‘s regime‘s removal of Kossovo‘s autonomy in 1989 (Hechter 2000: 76-77). 28 They might also have avoided secession. 29 Interestingly, Czechoslovakia is absent from Snyder‘s account of the relationship between ethnofederalism and violence. Bosnia because it had Croats and Serbs who also wanted to stay linked to their respective ethnic kin. These groups were aided and abetted by Serbia and Croatia, respectively. Bosnia, the most multi-ethnic republic, was perhaps destined to be the most violent.30 In 2001, violent conflict broke out in Macedonia, whose significant Albanian minority resented the dominance of Slavs. War between Armenia and Azerbaijan was largely fought over the inclusion of an Armenian ethnic enclave (Nagorno-Karabakh) in the latter. In Georgia, two conflicts broke out, between Georgians and South Ossetians cut off by Georgia‘s secession from their kin in North Ossetia (within Russia), and between Georgians and Abkhazians – who baulked at being included in what they saw as a Georgian state. The only other violence was in the Trans-Dniestrian region of Moldova, where Ukrainians and Russians resented their inclusion in Moldova. Just as communist federal break-up was fuelled by centralizing measures, the same could be said of the violence that arose in the newly-independent, still heterogeneous, but unitary, republics. The wars in Croatia, Macedonia, the South Ossetian and Abkhazian regions of Georgia and Trans-Dniestria were all influenced by the majoritarian policies of the states‘ dominant groups. In Croatia, a minority rebellion broke out after the newly independent Croatian regime adopted a flag that resembled that of the war-time Croatian Ustashe regime that had commited genocide against the Serbs, and after it moved to disarm its Serbian policemen (Hayden 1992). Seen in this way, these conflicts were similar to those in the Kurdish region of Turkey or Iraq, or the Basque region of Spain under Franco, i.e. they were reactions to centralization. It seems unreasonable simply to attribute them to multi-national federation per se.31 Fourth, while it is true that only federations broke apart in communist eastern Europe, this glosses over the more basic fact that the states that broke apart were also the most nationally diverse states œ which explains why they were federations. 32 In the case of the Soviet Union, Russians had a bare majority of the total population (51 per cent), while in Czechoslovakia and Yugoslavia, the largest groups had 63 and 39 per cent respectively. In none of the communist unitary states, did the total minority population constitute more than 17 per cent. The largest single minority group was the Turks of Bulgaria, with roughly 8 per cent of the population. It makes at least as much sense to argue that the instability of the communist federations resulted from their ethno-national diversity as their ethnofederal structures. In other research O‘Leary has shown that national federations that are durably democratic and majoritarian have a staatsvolk, a dominant people (O‘Leary, 2001). While lacking a staatsvolk does not guarantee political instability in a federation it makes it more likely. The United States, built 30 It is useful to remember this when considering the Horowitz/American argument that cross-cutting republican and ethnic boundaries have conflict-reducing effects. 31 Just as violence cannot be fairly attributed to ethnofederalism, nor can ethnic unmixing (i.e. the large movements of population that accompanied the break-up of Yugoslavia and the Soviet Union) œ see Brubaker 1996. Unmixing was also a result of the lack of congruency between ethnic groups and federal boundaries, and the nationalizing policies of the successor states (i.e., their unwillingness to consider ethnofederal principles). 32 As Watts claims: ”it is not so much because they are federations that countries have been difficult to govern but that it is because they were difficult to govern in the first place that they adopted federation as a form of government‘, Watts 1999: 110. around an historically dominant nationality of WASPs, proved more stable than Nigeria œ which lacks a clearly dominant people. The same comparison helps suggest why the Russian Federation is more stable and secession-proof, thus far, than the Soviet Union. Russians have a majority of 81.5 per cent in the Russian Federation; they had only 51 per cent in the UUSR. The unitary states of eastern Europe may have held together, in other words, not because they were not federations, but because each of them has dominant communities able to hold their states together if they had wanted to (See Table 8.2). Conversely, it is not at all certain that if, counterfactually, Yugoslavia had been a unitary state when it democratized that it would have stayed together.33 Ireland was able to secede from the much less diverse but unitary United Kingdom after the first universal male suffrage elections were held in 1918. Table 2 The largest community‘s proportion of the population in the communist states of Eastern Europe Communist Federations Largest Community Percentage of Population Yugoslavia Serbs 38.9 Soviet Union Russians 51 Czecho-Slovakia Czechs 63 Communist Unitary States Bulgaria Bulgarians 83 Albania Albanians 95 Romania Romanians 89.5 Hungary Hungarians 89.9 Poland Poles 97.6 Sources: Sources: CIA World Factbook 2001, (information for Bulgaria, Albania, Romania, Hungary, Poland); Library of Congress Country Studies ,; Library of Congress Country Studies; Lastly, it is simply wrong to claim, as Snyder and others do, that ethnofederalism is unworkable. Two of the worlds‘ oldest states, Switzerland and Canada, are ethnofederations. They have lasted from 1848 and 1867, respectively, and both demonstrate that the accommodation of ethnic minorities through ethnofederalism is consistent with prosperity and the promotion of basic individual rights. The world‘s largest democracy, India, is also an ethnofederal state, and the most successful large-scale post-colonial democracy. 34 Belgium, while of more recent vintage, has adopted successful ethnofederalist structures, while Russia, if Chechenya is left aside, appears to be on its way towards establishing itself as a democratic ethnofederation. Within each of these states, there is plenty of evidence, including polling data and the positions of their political parties, that minorities are content with less than a sovereign state (Keating 2002: 7-9). Together, these qualifications question the assumption that multi-national federalism is bound to fail. Our next task is to inquire into the conditions that make success more or less likely. Explaining success and failure The five conditions that facilitate, but do not guarantee successful multi-national federations are implicit in the preceding discussion. Here we spell them out. (i) The presence of a Staatsvolk As Table 8.1 shows, national federations are more stable than multi-national federations. The latter are more likely to fail or break-up. The reason is straightforward: national federations are generally nationally homogeneous, or virtually so. However, O‘Leary‘s data shows that the relative stability of multi-national federations is also related to the demographic preponderance of their largest national communities, to whether or not these constitute staatsvolk. A Staatsvolk can feel secure œ and live with the concessions attached to multi-national federation, and ceteris paribus, has the demographic strength and resources to resist secessionism by minority nationalities. Multi-national federations without a staatsvolk are more likely to be to be unstable, face secessionism or break-up, because minorities are more likely to think they can prevail (O‘Leary 2001). Russia‘s future cannot be extrapolated from the experience of the Soviet Union, because Russians are far more dominant within the former than they were within the latter. The same argument implies that Nigeria and a future European federation will, ceteris paribus, be relatively unstable, as neither possesses a Staatsvolk. What must be considered in our ”ceteris paribus‘ clause? We hypothesize as follows:(a) Multi-national federations without a Staatsvolk, if they are to survive as democratic entities, must develop consociational practicses that protect the interests of all the encompassed national and ethnic communities with the capacity to break-away. (b) The existence of a Staatsvolk, or the existence of consociational practices, will not themselves assure the stabilization of a multi-national democratic federation, though they will separately or cojointly increase its survival prospects. (c) Other conducive external and internal political, economic and social relationships may decide the fate of a multi-national federastion. The character of multinational power-sharing, whether a national minority has backing from a powerful neighbouring state and whether its region is on the border of the federation will 34 India‘s success is explained away by Snyder as a result of the unwillingness of its civic central authorities to recognise ethnicity (2000: 287-96). This is nonsense. It is odd that a federation, the internal boundaries of which are constructed along ethno-linguistic lines, can be seen in this way. Where there is some truth is that Indian governments have refused to recognize religiosity as the basis of provincial formation. assuredly matter, as will the democratic and legal character of the federation, its mode of formation, and its prosperity. (ii) The federation‘s national communities should not only have self-government, there should also be consociational government at the centre. When federalism is defended as a method of conflict regulation, the emphasis, as we have seen, is usually on how it can provide minorities with guaranteed powers of territorial self-government. Sometimes it is also argued that a virtue of federalism is that it avoids the ”winner takes all‘ outcome associated with Westminster-type regimes: a group that is excluded at the centre may be able to console itself with regional power. 35 However, federalism is about ”shared rule‘ as well as ”self-rule‘, and national minorities are likely to want a federal government that represents them, that is inclusive, and indeed, we would say, consociational. National and ethnic minorities excluded from the federal government will have a reduced stake in the federation and the federal government will be less inclined to promote their interests. It is not surprising, then, that all of the durably democratic multi-national federations have practised consociational forms of democracy within the federal government. Such arrangements involve four features: cross-community executive power-sharing, proportional representation of groups throughout the state sector (including the police and judiciary), ethnic autonomy in culture (especially in religion or language), and formal or informal minority-veto rights (Lijphart 1977; O’Leary 2003 Ms.). Consocational practices within the federal government are relatively undisputed in the case of Canada, Switzerland, and Belgium, (see Noel 1993; Steiner 1989; Hooghe 1993), and Lijphart has recently claimed that had effective consociational traits during its most stable period under Nehru (Lijphart 1996; Adeney 2002). Since Congress‘s decline, India has been governed by a broad multi-party coalition representing its diversity. Even if one does not count India as consociational in respect of having cross-community executive power sharing in New Delhi, it has usually had descriptively diverse representation of religious, ethnic and linguistic groups in the cabinet and civil service. We can see the salience of consociational organisation in the federal government in the case of many of the failed federations, where centrifugal pressures were often exacerbated by un-representative federal governments. In Pakistan, before the secession of Bangladesh, a crucial federal agency, the Army, was dominated by the west (Nasr 2001). This was also a serious problem in Yugoslavia, where the army, one of the most important federal institutions (absorbing two-thirds of the federal budget), was dominated by Serb officers, many of them from Serbian minorities who shared Milosevic‘s vision of a recentralized state. The Yugoslav Federal Council, the most important political institution, and one based on (non-democratic) consociational principles, was subject from the late 1980s to an undisguised take-over by Serbian politicians. After having suspended the autonomy of Kossovo and Voivodina, the Serbia-Montenegrin alliance gained control of four of the Federal Council‘s seven seats, plunging the federation into crisis. The Soviet Union broke up after an abortive take-over of the central government by conservatives opposed to decentralization. The episode undermined Gorbachev‘s 35 ”Federalism reduces conflict by allowing those political forces excluded from power at the top the opportunity to exercise regional power‘. Hanf 1991: 43. attempt to re-organise the federation in ways that would have given the republics more self-government and better representation in Moscow. The breakdown of the Nigerian federation in 1966-67, which included anti-Ibo violence in the northern Hausa region and the bloody Biafran war of secession, arose after a coup which led to the centre being dominated by Ibo officers and a counter-coup in which these officers were overthrown (Suberu 2001). 36 Much of Nigeria‘s post-1970 conflict, including sectarian warfare between Muslims and Christians and the rise of violent separatism in the oil-rich Delta area has also been traced to the lack of inclusiveness at the level of the centre (Suberu and Diamond 2000: 6-7, 13).37 The breakdown of the West Indies federation was linked to Jamaicans‘ lack of representation and influence at the centre, and in the case of the federation of Nyasaland, Northern and Southern Rhodesia, it was black Africans who were under-represented (Watts 1999: 111). This suggests that it will not be sufficient for the Nigerian, Ethiopian and Pakistani federations or any prospective Iraqi federation to practise democracy. Past evidence suggests they will need adopt and maintain consociational governance at the federal 36 The Igbo coup led by Major-General Aguiyi-Ironsi in January 1966 was followed by a ”Unification Decree‘ which moved Nigeria towards a unitary state. The regional and federal public services were combined, to the considerable advantage of the better-educated southerners and the disadvantage of the Hausa.. This, and the loss of many Northern military leaders in the January coup, set off anti-Igbo violence in the North, and contributed to Ironsi‘s assassination in July ( Suberu 2001: 31). 37 In a country that is equally divided between Muslim and Christian and between North and South, all four military governments in the 1984-99 era were headed by northern Muslims. After General Abacha‘s rise to power in 1993, not only the head of state was a Muslim, but also the Chief of Defense staff, Inspector-General of Police, Minister of Internal Affairs, National Security Adviser, and Chief Justice. This sectionalism, in Suberu and Diamond‘s words, provoked ”much alarm, alienation, and even paranoia‘ (Suberu and Diamond 2000: 13). It is not surprising that Abacha‘s northern and Muslim-dominated government was able to trample on the rights of the minorities in the oil-rich Niger Delta, and to execute their leaders, including the novelist Ken Saro-Wiwa. centre.38 It also suggests that calls to have a fully-fledged European federation, with the classic bicameral arrangements of the USA, to address the so-called democratic deficit in the European Union will fail unless such calls are accompanied by strong commitments to consociational devices. Consociational governance would imply strong mechanisms to ensure the inclusive and effective representation of all the nationalities of the European Union in its core executive institutions, proportionate representation of its nationalities in its public bureaucracies and legal institutions, national autonomy in all cultural matters deemed of profound cultural significance (e.g. language, religion, education), and last, but not least, national vetoes to protect national communities from being out-voted through majoritarian rules. In short, many of the current consociational and confederal features of the EU which some federalists want to weaken or temper in their pursuit of formal federation are in fact required to ensure the EU‘s prospects as a multi-national democratic federation. The EU‘s greatest current danger stems from its ardent majoritarian federalists. This argument about the importance of accommodation through consociational devices is different from that put forward by Linz and Stepan (Linz and Stepan 1992). They put their faith in the ability of federation-wide political parties to win support from all groups, to balance majority and minority concerns, and to build what Linz calls ”bundestreue‘, an overarching loyalty to the state (Linz 1997). In their view, the key reason for the disintegration of the Yugoslav and Soviet federations was that the first democratic elections were held in the republics rather than the state (whereas in post-Franco Spain it was the other way around). In Yugoslavia this sequencing gave divisive republican elites the resources and space to promote break-up, and obstructed the 38 This is particularly important where, as in Nigeria, the lion‘s share of power is allocated to the centre (Diamond 2001: xvi). An obstacle to the practice of consociationalism in Nigeria is its presidential system of government. An executive presidency is held by one person. It is an integrative institution that does not lend itself as well to broad-based representative government as parliamentary (cabinet-style) executives (Watts 1999: 88). However, even presidential executives can informally establish inclusive cabinets (or formally include representative vice-presidents). For details on recent conventions that Nigeria has developed to (imperfectly) ensure representative government at the centre, see Suberu and Diamond 2000: 26-30. As federal regions are also usually ethnically heterogeneous, it is also helpful to have consociational practices at the intra-regional level also. This not only addresses the criticism that giving self-government to national minorities will lead to an abuse of their powers against local minorities, it also promotes good inter-regional and regional-centre relations. Particularly when a regional minority is part of a state majority, abuse of its position by the regional majority can have serious and negative effects on centreregional and inter-regional relations One of the weakness of Canadian federalism is that while Canada‘s francophone minority has been reasonably accommodated at the federal level through (partly informal) consociational practices, there has been no commensurate accommodation of minorities at the provincial level, except, arguably, in New Brunswick. Measures taken by Quebec against its anglophone minority in the late 1980s helped to produce an anti-Quebec backlash in the rest of the country that undermined constitutional negotiations that were aimed at accommodating Quebec. One result was that, by 1990, support for separation in Quebec had soared to its highest levels ever. The Canadian federation may have been saved by the fact that separatists were not in power at the time and were unable to capitalise on the divisions by calling a referendum on separation. When they were able to call such a referendum, five years later, when the atmosphere was more stable, they still managed to win 49.4 per cent of the vote. organization of federation-wide parties with an interest in holding the state together. Had federal elections been held first, federation-wide parties would have been able to act as unifying forces. This reasoning is, however, questionable. State-wide parties may well be likely to do better in state-wide elections than in regional elections, but there is no guarantee, or even likelihood, that they will do well at any level in societies with noticeable national divisions. In the United Kingdom‘s first democratic elections, in the mid-1880s, the overwhelming majority of Irish seats were won by Irish nationalist parties.39 The fact that they were elected in state-wide elections, as opposed to regional elections, does not appear to have coloured their view of the UK, or their ability to secede from it, and they won despite the presence of competitors from state-wide parties. Czechoslovakia‘s first democratic elections, which involved concurrent state-wide and regional elections, produced no state-wide parties at the state-wide (federal) elections other than the discredited communists, who won 23 of the country‘s 150 seats. Even they subsequently divided into Czech and Slovak factions. All of the other parties that won seats were based on either the Czech, Slovak, or Hungarian populations (Leff 1998: 98). Perhaps this political party fragmentation into ethnonational blocs was due, as Leff claims, to the simultaneity of elections at both levels (ibid. 97), i.e. to the fact that the federal election was not held in advance. But how then, are we to explain the first democratic election returns in the unitary states of Eastern Europe, where there were no regional elections? In these cases, party support still broke down almost exactly along ethnonational lines (See Table 8.3.), with little evidence of integrative vote-pooling activities by either party elites or voters. These results are difficult to square with Linz and Stepan‘s assumption that Yugoslav state-wide elections would have produced strong Yugoslav state-wide parties, unless one is to assume that Yugoslavia was a good deal less divided than its neighbours. Given that it was the only state in Eastern Europe whose major communities had persons who had butchered each other within living memory (1941-45), this assumption is implausible. This comparative evidence suggests that state-wide elections in Yugoslavia would have resulted in elections that reflected its national divisions. Hoping for state-wide parties to hold Yugoslavia together was probably wishful thinking. Stability would have required successful bargaining among the different minority nationalist parties on a new consociational and confederal constitution. Such bargaining as there was on this agenda did not succeed. Table 3. Support for minority political parties in the first post-communist elections in the unitary states of Eastern Europe State/year of election Minority/proportion of state‘s population Support for Minority Party as a proportion of 39 From 1885, elections in the UK were based on a universal make franchise, and from 1918, a universal franchise. 23 votes cast Bulgaria/1990 Turks/8.5 6.040 Romania/1990 Magyars/7.8 7.2 Poland/199141 Germans/1 1.2 Sources: Centre for the Study of Public Policy: Mass Behaviour in New Democracies, at:; the University of Essex’s Project on Political Transformation and the Electoral Process in Post-Communist Europe, at; Elections in Central and Eastern Europe at (iii) Authentic (democratic) multi-national federations are more likely to be successful than pseudo-(undemocratic) federations. An authentic multi-national federation is democratic. It allows the representatives of its respective national communities to engage in dialogue and open bargaining about their interests, grievances and aspirations. Such democratic dialogue is a pre-requisite for the development of co-operative practices. Democratic multi-national federalism may help to preclude the systematic transgression of individual and group rights It can prevents minority (secessionist) elites from exaggerating support for their preferences (Linz 1997). An authentic multi-national federation is also based on the rule of law, law that recognizes national, ethnic, or communal rights, a constitutional division of powers, and legal powers that approach those of impartial umpires. There is not yet an example of an established democratic multi-national federation failing (though the number of cases is small), although there are, as we have seen, numerous examples of democratizing federations that have not worked. The evidence, limited as it is, suggests that we should not automatically assume that Canada, Switzerland, Spain, India or Belgium will go the way of the flawed communist or post-colonial federations. (ii) ”‘Voluntary‘ or ”Holding together‘ multi-national federations are more likely to endure under democratic conditions than those that are coercively constructed after modern social mobilizations. Stepan distinguishes between three types of multi-national federation: a. those that voluntarily come together from distinct polities/colonies, like the Swiss and Canadian federations; b. those that are created from unitary states in an attempt to ”hold‘ the polity together, such as Belgium, and, one might argue, India; and c. those that are forced together (or ”put‘ together) by a dominant group, such as the Soviet Union (Stepan 2001: **).42 40 The 1990 election was to elect a constitutional assembly. The first parliamentary elections took place in 1991. The Turkish minority party, the Movement for Human Rights and Freedoms, won 7.5 per cent on this occasion. 41 These were the first parliamentary elections. Presidential elections were held in 1990. 42 Canada‘s birth was in fact a hybrid of ”coming together‘ and ”holding together‘ processes: on the one hand, it involved the joining together of a number of previously separate British North American colonies; on the other hand, it involved the division of the unitary colony of the ”Two Canadas‘ into the separate federal units of Ontario and Quebec. Federations that are consensually established as a result of elite bargaining, whether of the holding or voluntary variety, are more likely to be considered as legitimate by their citizens and more likely to survive than those that result from coercion. A foundational act of co-operation is also more likely than one of coercion to promote traditions of accommodation. Canada‘s success is owed in part to the fact that it originated in 1867 from a compact between Anglophone and Francophone elites led by John A. MacDonald and George Etienne Cartier. The Swiss federation was also the result of different groups agreeing to ”con-federate‘. While the Spanish and Belgian federations emerged from unitary states, they too were based on agreement between representative elites. India, which stands out as one of the few post-colonial federal success-stories, is also one of the few where indigenous elites took the decision to federate by themselves œ albeit reluctantly, and albeit after prior British tutelage (Adeney 2002). Most of the failed federations, on the other hand. were put together without the consent of minority leaders.43 This does not augur well for Bosnia-Hercegovina, which exists as a federation because of the internationally-imposed Dayton Accords. (v) Prosperous multi-national federations (or states) are more likely to endure than those that are not. Walker Connor has correctly counseled us against exaggerating the importance of materialism when questions of national identity are at stake. Prosperity should not be considered a sufficient or even a necessary condition (as the example of India shows) for holding a multi-national federation together (Connor, 1994: 145-64). Nonetheless, ceteris paribus, prosperity œ and distributive fairness œ may matter. The plight of the communist federations and post-colonial federations was plainly exacerbated by their inability to provide materially for their citizens and by the discrediting of communist central planning. In the Ukraine and the Baltic republics, even Russians voted for the break-up of the USSR. In both Yugoslavia and the Soviet Union, the catalyst for break-up was necessary economic reforms, and the charge was led in both cases by those republics (Slovenia and Croatia in the case of Yugoslavia, the Baltic republics in the case of the Soviet Union) which had the most to gain materially from going it alone. And it needs no belabouring of the obvious to observe that distributive fiscal and expenditure issues are the meat and drink of political controversy in those federations which do not use equitable formulae for fiscal equalization. Conclusion We have attempted to offer a more balanced and nuanced assessment of the value and durability of multi-national federations than that put forward by critics of ”ethnofederalism‘, without falling victim to the blandishments of the most aredent 43 The importance of voluntary origins for the legitimacy and stability of states, whether federations or unitary states, is often recognized in the rival historiographies of federalists/unionists and separatists, with the former arguing that the federation/union arose voluntarily while the latter argue it was imposed. Thus in Canada, separatists point to the conquest of 1759 as Canada‘s starting point, and/or argue that the confederation agreement of 1867 was not ”really‘ voluntary, as Francophone elites did not have a serious alternative of separating. Federalists, on the other hand, point to the key role that Francophone elites had in shaping the federal agreement. Similar debates take place between unionists in Britain and Scottish separatists. federalists. Democratic federalism did not cause the break-up of the communist states, as these were not authentic democratic (or economically efficient) federation. Not all multinational federations have failed. There are also a small number of remarkable success stories. We have tried to identify conditions that are conducive to the success of multinational federations. It is important that they be democratic and respect the rule of law. It helps if they are prosperous. It helps if they came together voluntarily. If federations develop from a unitary state, our arguments suggest that early and generous responses to expressed demands for minority self-government will work better than delayed and grudging responses. The demographic composition of the federation matters: a federation with a dominant ethno-national community is likely to be more stable that one that does not. Lastly federalism is usually not enough: Consociational practices, particularly at the level of the federal government, are very important to the success of multi-national federalism if there is no Staatsvolk. References Adeney, K. (2002) ”Constitutional Centring: Nation Formation and Consociational Federalism in India and Pakistan‘, Journal of Commonwealth & Comparative Politics 40, 3: 8-33. Agranoff, R. ed. 1999. Accommodating Diversity: Asymmetry in Federal States. Baden-Baden: Nomos. Anderson, B. 1998. The Spectre of Comparisons: Nationalism, Southeast Asia and the World. London: Verso. Arora, B., and D. V. Verney. 1995. Multiple Identities in a Single State: Indian Federalism in Comparative Perspective. New Delhi: Konark Publishers PVT Ltd. Bauer, O. 2000. The question of nationalities and social democracy (Translation of Nationalitätenfrage und die Sozialdemokratie by J. O‘Donnell). Minneapolis: University of Minnesota Press. Bosco, A. ed. 1992. The Federal Idea: the History of Federalism since 1945. London: Lothian Foundation Press. Brass, P. 1990. The Politics of India Since Independence. New Delhi: Cambridge University Press. Brubaker, R. 1996. Nationalism Reframed. Cambridge: Cambridge University Press. Bunce, V. 1999. Subversive Institutions: The Design and the Destruction of Socialism and the State. Cambridge: Cambridge University Press. Buchanan, A. 1990. Secession. The Morality of Political Divorce from fort Sumter to Lithuania and Quebec. Boulder. Westview. Coakley, J. 1994. ”Approaches to the Resolution of Ethnic Conflict: The Strategy of Nonterritorial Autonomy‘, International Political Science Review, Vol. 15, No. 3: 297-314. Connor, W. 1984. The National Question in Marxist-Leninist Theory and Strategy. Princeton N.J.: Princeton University Press. de Certaus, M., J. Dominique, and J. Revel. 1975. Une Politique de la Langue. La Révolution Française et les patois: L’eenquête de Grégoire. Paris: Gallimard. Diamond, L. 2001. ”Foreword‘ in Suberu, R. 2001. Federalism and Ethnic Conflict in Nigeria. Washington D.C.: United States Institute of Peace). Elazar, D. 1994. Federalism and the Way to Peace. Kingston: Queen‘s Institute of Intergovernmental Relations. —— 1987. Exploring Federalism. Tuscaloosa: University of Alabama. Ferenczi, T. 2001. ”The Republic, a French myth‘ (unpublished paper). Fischer, J. 2000. —Apologies to the UK, but ‘Federal’ is the Only Way.“ The Independent May 16: 4. Forsyth, M., ed. 1989. Federalism and Nationalism. Leicester: Leicester University Press. Gellner, E. 1983. Nations and Nationalism. Ithaca, NY: Cornell University Press. Glazer, N. 1983. ”Federalism and Ethnicity: The American Solution.‘ In N. Glazer ed., Ethnic Dilemmas, 1964-82. Cambridge, Mass: Harvard University Press: 274-92. Gordon, M. 1964. Assimilation in American Life. New York: Oxford University Press. Hanf, T. 1991. ”Reducing Conflict Through Cultural Autonomy: Karl Renner’s Contribution.‘ in U. Ra’anan, M. Mesner, K. Armes and K. Martin eds., .State and Nation in Multi-Ethnic Societies: The Breakup of Multi-National States. Manchester: Manchester University Press: 33-52. Hayden 1992. ”Constitutional Nationalism in the formerly Yugoslav Republics‘, Slavic Review, 51: 654-673. Hayward, J. 1993. After the French Revolution: Six Critics of Democracy and Nationalism. New York, New York University Press. Hechter. M. 2000. Containing Nationalism. Oxford: Oxford University Press. Hooghe, L. 1993. ”Belgium: From Regionalism to Federalism‘, in J. Coakley ed., The Territorial Management of Ethnic Conflict. London. Frank Cass: 44-68. Horowitz, D. 1998. —Self-Determination: Politics, Philosophy and Law.“ in M. Moore ed., National Self-Determination and Secession. Oxford: Oxford University Press: 181-214. ——1985. Ethnic Groups in Conflict. Berkeley: University of California Press. Keating, M. 1996. Nations Against the State: the new politics of nationalism in Quebec, Catalonia and Scotland. London: Macmillan Press. ——2001. Plurinational Democracy: Stateless Nations in a Post-Sovereignty Era. Oxford: Oxford University Press. King, P. 1982. Federalism and Federation. London: Croom Helm. ——2002. ”Plurinational Democracy in a Post-Sovereign Order‘ (unpublished paper). Kymlicka, W. 2001. Politics in the Vernacular. Oxford: Oxford University Press. —— 2000. —Federalism and secession: At Home and Abroad“, The Canadian Journal of Law and Jurisprudence Vol. XIII, No. 2, 2000, pp. 210-11. —— 1995. Multicultural Citizenship. Oxford. Oxford University Press. Leff, C. S. 1998. The Czech and Slovak Republics: Nation Versus State. Boulder, CO: Westview. Leonardy, U. 2000. ”Treaty_Making Powers and Foreign Relations of Federated Entities‘ in B. Coppieters, D. Darchiashvili and N. Akaba (eds), Federal Practice: Exploring Alternatives for Georgia and Abkhazia. Brussels: VUB University Press: 151-68. Lieven D. and J. McGarry. 1993. ”Ethnic Conflict in the Soviet Union and its Successor States‘, in J. McGarry and B. O‘Leary (eds), The Politics of Ethnic Conflict Regulation. London: Routledge: 62-83. Lijphart, A. 1996. ”The Puzzle of Indian Democracy: A Consociational Interpretation‘, American Political Science Review 65: 682-693. ——ed. 1981. Conflict and Coexistence in Belgium: The Dynamics of a Culturally Divided Society. Berkeley: Institute of International Studies University Of California. ——1979. ”Consociation and Federation: Conceptual and Political Links‘, Canadian Journal of Political Science, XII, 3, 1979: 499-515. ——1977. Democracy in Plural Societies: A Comparative Exploration. New Haven, London, Yale University Press. Linz, J. 1997. ”Democracy, Multinationalsm and Federalism‘ Working Paper, 1997/103. ——and A. Stepan, ‘Political Identities and Electoral Sequences: Spain, The Soviet Union, and Yugoslavia’, Daedalus, 121, no, 2, Spring 1992: 123-39 Lipset, S.M. 1960. Political Man: The Social Bases of Politics.Garden City, New York: Doubleday —— 1985. —The Revolt against Modernity“, in S.M. Lipset, Consensus and Conflict: Essays in Political Sociology. New Brunswick: Transaction: 253-93 Lukic R. and A Lynch. 1996. Europe from the Balkans to the Urals: The Disintegration of Yugoslavia and the Soviet Union. Oxford: SIPRI, Oxford University Press. Madison, J., A. Hamilton, and J. Jay. 1788/1987. The Federalist Papers, edited and with an introduction by Isaac Kramnick. Harmondsworth: Penguin. Majocchi, L. V. 1991. ”Nationalism and Federalism in 19th Century Europe.‘ In A. Bosco ed., The Federal Idea: The History of Federalism from Enlightenment to 1945. London: Lothian Press: 155-65. McGarry, J. 2001. ”Northern Ireland and the Shortcomings of Civic Nationalism‘, in J. McGarry ed., Northern Ireland and the Divided World. Oxford: Oxford University Press: 109-36. McGarry, J. and B. O‘Leary. 1995. Explaining Northern Ireland: Broken Images. Oxford: Blackwell. McGarry, J. and B. O‘Leary. 1993. ”The Macro-Political Regulation of Ethnic Conflict‘, in J. McGarry and B. O’Leary eds., The Politics of Ethnic Conflict Regulation. London: Routledge: 1-40. Moore, M. 2001. The Ethics of Nationalism. Oxford: Oxford: Oxford University Press. Nasr, V. 2001. ”The Negotiable State: Borders and Power-Struggles in Pakistan‘ in B. O’Leary, I. S. Lustick and T. Callaghy (eds.), Right-Sizing the State: The Politics of Moving Borders. Oxford, Oxford University Press: 168-200. Noel, S. 1993. ”Canadian Responses to Ethnic Conflict: Consociationalism, Federalism and Control.‘ In J. McGarry and B. O‘Leary eds., The Politics of Ethnic Conflict-Regulation: Case Studies of Protracted Ethnic Conflicts. London: Routledge: 4161. Nordlinger, E. 1972. Conflict Regulation in Divided Societies. Harvard: Center for International Affairs. O‘Leary, B. 2001. ”An Iron Law of Nationalism and Federation? A (neo-Diceyian) theory of the necessity of a Federal Staatsvolk, and of Consociational Rescue‘, Nations and Nationalism 7 (3): 273-96. ——2002. ”In Praise of Empires Past.‘ New Left Review 2nd series(18): 106-30. ——2003 (manuscript in progress). Consociation. Pfabigan, A. 1991. ”The Political Feasibility of the Austro-Marxist proposal for the Solution of the Nationality Problem of the Danubian Monarchy.‘ in U. Ra’anan, M. Mesner, K. Armes and K. Martin eds., State and Nation in Multi-Ethnic Societies: The Breakup of Multi-National States. Manchester: Manchester University Press: 53-63. Premdas, R. R. 1998. Secession and Self-Determination in the Caribbean: Nevis and Tobago. St. Augustine, Trinidad, The University of the West Indies. Riker, W. H. 1964. Federalism: Origin, Operation, Significance. Boston. Roeder, P. 1991. ”Soviet Federalism and Ethnic Mobilization‘, World Politics, 43, January: 196-232. Snyder, J. 2000. From Voting to Violence: Democratization and Nationalist Conflict. New York. Norton. Steiner, J. 1989. ”Power-Sharing: Another Swiss Export Product ?‘ in J. Montville ed., Conflict and Peacemaking in Multiethnic Societies. Lexington, Mass.: Lexington Books: 107-114. Stepan, A. 2001. Arguing Comparative Politics Oxford: Oxford University Press. —— 1999. ”Federalism and Democracy: Beyond the U.S. Model‘, in Journal of Democracy, 10, 4, October: 19-34 Suberu, R. 2001. Federalism and Ethnic Conflict in Nigeria. Washington D.C.: United States Institute of Peace. ——and L. Diamond, 2000, ”Institutional Design, Ethnic Conflict Management, and Democracy in Nigeria.‘ In A. Reynolds ed., The Architecture of Democracy: Institutional Design, Conflict Management and Democracy. Oxford: Oxford University Press, 2002: 400-428. Tarlton, C. 1965. ”Symmetry and asymmetry as elements of federalism: A theoretical speculation‘, Journal of Politics, 27, 4: 861-74. Trudeau, P. 1968 (1965). ”Federalism, Nationalism and Reason.‘ in P. Trudeau ed., Federalism and the French Canadians. Toronto: University of Toronto Press: 182-203. Von Beyme, K. 1985. Political Parties in Western Democracies Aldershot: Gower. Watts, R. 1998. ”Federalism, Federal Political Systems, and Federations.‘ Annual Review of Political Science 1: 117-37. ——1987. ”Federalism.‘ in The Blackwell Encyclopaedia of Political Institutions, edited by Vernon Bogdanor. Oxford: Basil Blackwell: 228-30. —— 2001. ”Models of federal power-sharing‘, International Social Science Journal, March: 23-32 . Weller, M. 1992. ”The International Response to the Dissolution of the Socialist Federal Republic of Yugoslavia.‘ American Journal of International Law 86: 569-607. Woodward, S. 1995. Balkan Tragedy. New York: Brookings Institution.