Full Text: COPYRIGHT Oxford University Press (UK), 1994. Use of this article is restricted to class use in Spring 1997 of internet course in Conflict Resolution. This article may not otherwise be reproduced or distributed without written permission.

National and ethnic conflict has been a persistent feature of modernity but the last few years have brought seismic changes in the relations between several ethnic communities around the world. The disintegrations of the Soviet Union and Yugoslavia have led to multiple ethnic wars. In South Africa, apartheid, the world's most infamous system of ethnic domination, appears to be on the verge of extinction, while a new system of domination has been established in Fiji. Elsewhere, the Czechs and Slovaks have snapped the hyphen which held their fragile federation together, and India seems poised on the brink of catastrophic communal conflict sparked by the rise of Hindu fundamentalism. In Canada, Cyprus, Israel/Palestine and Northern Ireland the principal protagonists are engaged in negotiations about their political future or in ~negotiations about negotiations'. The times are especially appropriate for examining the ways in which national and ethnic conflicts might be politically regulated. This article develops a taxonomy of the macro-political methods of national and ethic conflict regulation. We also briefly explain the circumstances under which particular strategies for managing such conflict are attempted; and we evaluate the normative merits of the different forms of conflict regulation from a liberal-democratic perspective.

Eight distinct macro-methods of conflict regulation are distinguishable, to wit: (i) genocide; (ii) forced mass population transfers; (iii) partition and/or secession (self-determination); (iv) integration and/ or assimilation; (v) hegemonic control; (vi) arbitration (third-party intervention); (vii) cantonisation/federalism; and (viii) consociationalism. This taxonomy suggests neither discreteness nor exhaustiveness. Often the eight strategies are mixed and targeted at the same ethnic group(s), or alternatively, different strategies are aimed at different ethnic groups within the same state. Oliver Cromwell offered Irish Catholics a choice between genocide and forced mass population transfer: they could go ~To Hell or Connaught'! Stalin used genocide, forced mass population transfers and coercive assimilation to manage Soviet ethnic conflict. Belgium has practised consociationalism to regulate divisions between its ~spiritual families' and federalism to resolve tensions between its linguistic communities.

Our taxonomy is divisible according to the goals of political agents. The first four (genocide, mass population transfers, partition/secession and integration/assimilation) generally aim to eliminate ethnic or national differences, while the last four aim to manage differences (hegemonic control, arbitration, cantonisation/federalism, and consociation). It is not, in our opinion, either plausible or desirable to say which of these two approaches is inherently superior. Of the eliminating-differences strategies, there are moral justifications for partition (or secession) and integration (assimilation) which have been advanced by generations of liberals and socialists. However, there is no obvious moral hierachy which enables people to claim that integration is better than partition or vice versa, unless there is widespread consent among the relevant ethnic communities for one option rather than the other. The merits of partition (or secession) as against integration (or assimilation) must be decided by political argument and pragmatic considerations, such as feasibility and estimates about long-run efficacy. There is nothing moral about genocide or forced mass population transfers, the other difference-eliminating stategies, although ~ethical' arguments have usually accompanied their implementation. Of the managing-differences strategies, only hegemonic control should be morally unacceptable to liberals and democrats (see below). The rest (arbitration, cantonisation/ federalism and consociationalism) are fully compatible with democratic norms. Support for them must, however, be tempered by empirical judgements about their feasibility and long-term efficacy.

Genocide, literally the killing of a race or kind ([gamma][epsilon][nu][omicron][sigma]), is a con term. Article II of the UN Convention on the Prevention and Punishment of the Crime of Genocide defines it as ~acts committed with intent to destroy, in whole or in part, a national, ethnic, racial or religious group as such'; and its sub-clauses include ~(a) killing members of the group; (b) causing serious bodily or mental harm to members of the group; (c) deliberately inflicting on the group conditions of life calculated to bring out its physical destruction in whole or in part; (d) imposng measures intended to prevent births within the group; (e) forcibly transferring children of the group to another group'. Much conceptual and moral controversy surround the definition of genocide but for our purposes genocide is the deliberate mass killing of a group (national, ethnic, linguistic or religious) in a territory controlled by the killers.[1]

Such genocides have been perpetrated throughout history. Native groups throughout the Americas were the victims of genocide at the hands of European colonisers. Genocides were committed by the Nazis and their allies in the 1940s, by the Turks against the Armenians, and within the Communist bloc in Eurasia. Since 1945 there have been genocides perpetrated in Burundi (of Hutu); in Iraq (of the Kurds); in Indonesia (of the Chinese and the indigenous population of East Timor); in Nigeria (of Ibo residents in the North); in Pakistan (of the Bengalis in what became Bangladesh); in Uganda (of the Acholi, the Lango, the Bagandans and the Nilotic tribes) and in Burma (of Muslims in border regions).

Genocides are intended to terminate national and ethnic conflict, and they often succeed in securing the relevant territories for imperial rulers. Yet genocides often fall to achieve their objectives, and always create explosive and historically entrenched bitterness and fear amongst the descendants of victims. After nearly eighty years, the Armenians are still the bitter enemies of the Turks. Serb-Croat relations in what was Yugoslavia are inflamed by memories of war-time genocide during World War II. Serb-Muslim relations in Bosnia have been irreparably damaged for the foreseeable future by the recent atrocities. If Russians and indigenous groups in the Baltic states, Ukraine and Kazakhstan are to coexist peacefully, many skeletons will have to remain buried.

It is possible to identify when and where genocide is likely to be contemplated. State-genocide is more likely to occur when: an empire is being contructed and maintained, where genocide is used as a deliberate policy of land acquisition and mass terrorisation, or to exact revenge upon a rebellious population (e.g. the conquest of the ~New World'); an ethnic community lacks geopolitical resources (its own state or a powerful diaspora); a subordinate ethnic community lacks geopolitical resources (its own state or a powerful diaspora); a subordinate ethnic community is left vulnerable within a disintegrating system of control (e.g. Armenians in Turkey); a given ethnic community (e.g. Jews, Ibos, Armenians, overseas Chinese) possesses economic superiority and cultural identifiability in conditions of industrialisation, but lacks military and political power[2]; and when a totalitarian state is established with a hegemonic ideology which demonises and scapegoats a section of the population (e.g. Jews).

~Frontier genocide', which, by contrast, may not be directly implemented by public officials, is likely to occur when settlers, possessed of technologically superior resources, displace natives from their access to land. It is a concomitant of colonisation and conquest.

A necessary condition for genocide appears to be the presence of an ethnic, racial or religious ideology which sanctions a non-universalist conception of the human species, and makes mass murder easier to accomplish. These belief systems may be more important than technological capacities for managing mass killings, as it is the discipline of the killers, rather than their instruments, which may best account for the scale of genocides. Genocides can be instrumental and ~pre-emptive' as well as being ideological: indigenous peoples were killed by European colonisers on the supposition that their circumstances were those of ~kill or be killed'. The same beliefs seem to have been important in motivating Tutsi genocides of Hutu in Burundi.

FORCED MASS POPULATION TRANSFERS

Forced mass population transfers occur where an ethnic community (or a set of communities) is compelled to leave its home. The Serbian term ~ethnic cleansing' brutally expresses the objective. Forced mass population transfers have been applied to those living in their ancestral homelands, as well as to recently settled immigrants. The target communities can be evicted from the state's territory or transferred internally. They can also be shifted internally under the guise of being ~repatriated', as happened under South Africa's so-called ~homelands' policy'.

Forced mass population transfers must be disinguished from three other types of population movement which appear outwardly similar. They are different from agreed ~population exchanges' which accompany partitions (such as those between Greece and Turkey after the end of World War I. The populations involved in ~agreed exchanges' never consider such moves to be voluntary, but their fate can be distinguished from those unilaterally compelled to move. Second, forced population transfers imply premeditated acts of policy and can be distinguished from involuntary movements of refugees which are by-products of wars and civil wars. Third, forced transfers involve the physical expulsion of an ethnic group and can be distinguised from policies which, by creating an inhospitable environment for dominated groups, result in their disproportionate emigration. The latter policies are associated with hegemonic control which we discuss below. There is also a question of scale involved in mass population transfers, and they should be distinguished from the expulsion of individuals or small groups. However, ~induced' transformations of demographic balances are the functional equivalents of mass population transfers. The deliberate importing of settlers into a territory to dilute the strength of a compact and contiguous community, i.e. settler colonialism, has similar consequences, although it is best seen as a form of hegemonic control.

Planned mass transfers of peoples against their will have been widespread in history. British imperial authorities, and their American counterparts, forced North American natives off their ancestral lands and onto reservations throughout the eighteenth and nineteenth-centuries. Stalin routinely shuffled around national and ethnic groups in the Soviet Union in the 1930s and 1940s. Jews were the victims of forced population transfers under the Nazis (and even earlier under the Romans). Several ethnic groups in eastern Europe, including Germans in Poland, Czechoslovakia and the Baltic region, were forced to relocate in the aftermath of the Second World war. In the 1970s, the Turkish Army expelled Greek Cypriots from northern Cyprus and Idi Amin drove the Asian community from Uganda. In the former Yugoslavia, the republics of Bosnia and Croatia have been scenes of forced population transfers. In western Europe, fascist parties want to expel or ~repatriate' refugees and ~guest-workers'.

As a policy instrument, forced population transfers may be designed to achieve a variety of ends. Like genocides, they are often advocated as integral components of imperial consolidation strategies. They can be designed to punish rebels and set examples ~pour encourager les autres', as with Cromwell's ~transplantation' strategies in Ireland, or they can form part of a defence policy against an external attack, as with Stalin's movement of the Volga Germans during World War II. Forced population transfers which accompany war or civil wars are often designed to establish demographic facts to strengthen territorial claims: consider the expulsion of Greek Cypriots from northern Cyprus in 1974 and the expulsion of Muslims from large chunks of Bosnia in 1992-93. Expulsions of groups may also be motivated by an acquisitive desire for their resources, as in the case of the eviction of Uganda's Asians or North American Indians. Alternatively, immigrant groups and temporary workers may be expelled during economic recessions when there are not enough jobs to go around. South Africa's "homelands' policy", by allowing the government to argue that many of the country's large black majority were citizens of other states, was a crude attempt to deflect international criticism of apartheid.

Forced population transfers may displace but do not always terminate national or ethnic conflict. Current outrages in the northern Caucasus can be traced directly to Stalin's displacements of peoples, some of whom have returned to reclaim their homes. One obstacle to a peace settlement in Cyprus is the demand of expelled Greek-Cypriots to have their land back. ~Ethnic cleansing' in the former Yugoslavia has added to the stockpile of historic grievances which exist there. There are no moral merits to forced mass population transfers. They violate any minimalist conceptions of human rights and any egalitarian political philosophies.

PARTITION & SECESSION (SELF-DETERMINATION)

By contrast with genocide and forced mass population transfers, partition and/or secession can, in principle, respect the right of national and ethnic communities to enjoy ~self determination'. Partitions resolve national and ethnic conflict, if they work, by the principle of divorce. They can be executed in three different ways: by the core of the relevant state, e.g. when the United Kingdom in effect decided how much of Ireland would be permitted to secede between 1920 and 1925; by the agreement of the divorcing parties, e.g. the break-up of Czecho-Slovakia proceeded with the agreement of the Czech and Slovak governments (but without the support of the Hungarians of Slovakia); by external imposition, e.g. the dismemberment of the Ottoman and Austro-Hungarian empires by the victorious allies after World War I.

Secessionists aspire to independent statehood, as with Slovak, Scots, Ukrainian, and Quebecois nationalists; while ~semi-secessionists' aspire to be linked to another state, as with Bosnian Serbs, Northern Irish nationalists, and possibly the Hungarians of Slovakia. Secessions normally involve the breakaway of minorities, but they can involve the bulk of the state's population as with Russia's secession from the Soviet Union in 1991 (the Yelstin option). In the years between 1948 and 1991 only one new state, Bangladesh, was carved out of an existing state, if we exclude the numerous cases of decolonisation of European and US controlled territories in Asia, Africa and Latin America, and the unrecognised (except by Turkey) secession of the Turkish Republic of Northern Cyprus in 1981. However, since the collapse of Yugoslavia and the Soviet Union secession has become a growth industry, the invogue method of national and ethnic conflict resolution. Czechs and Slovaks joined the bandwagon in early 1993; Eritrea will do so soon. There are multiple secessionist movements around the world, in Europe, Canada, the Commonwealth of Independent States, in Africa, and in central and south Asia.

Many secessions can and are justified as ways of escaping oppression and achieving freedom and self-government. However, the key problem with the principle of self-determination taking the form of secession as a means of eliminating national and ethnic conflict is that it begs four questions: who are the people; what is the relevant territorial unit in which they should exercise self-determination; what constitute a majority; does secession produce a domino effect in which minorities within seceding territories will seek self-determination for themselves?

In what were Yugoslavia and the Soviet Union these questions, far from being academic, have led to multiple civil wars. There are many other hard cases in seeking to apply the doctrine of self-determination. In Transylvania there are two major populations (Hungarians and Romanians) mixed together in the same region along with other smaller communities. In Quebec, aboriginal Canadians are unwilling to secede from Canade with the Francophone majority. In the Punjab and Kashmir, Hindus vehemently oppose the very idea of secession. In Slovakia, the significant Hungarian minority fears that the break-up of Czecho-Slovakia will be detrimental to their interests.

The constitution of a majority for self-determination begs the question of a majority in what region? As Ivor Jennings cynically remarked of the principle of self-determination, on the surface it seems reasonable: let the people decide, it is in fact ridiculous because the people cannot decide until somebody decides who are the people[3.] Exercising the principle of self-determination is only straightforward where there is no large or disgruntled minority within the relevant region affected by the proposed secession and when the seceding area includes the great majority of those who wish to leave. Unfortunately, it is difficult to think of instances where these optimum conditions have applied. Norway's secession from Sweden, and Iceland's from Denmark were exemplary cases. So was Slovenia's secession from Yugoslavia. However, the partitions of Ireland and India left significant minorities behind in Northern Ireland and Kashmir; and those who celebrated the exercise of self-determination in Yugoslavia and the Soviet Union have tempered their enthusiasm in the light of the ethnic time-bombs left behind.

With the collapse of the global cold war, there is now much greater room for successful secession and the alteration of borders artificially frozen by the strategic interests of the superpowers, as the reunification of Germany suggests. ~Globalisation' and the increasing power of regional supra-state organisations, may also make some state boundaries less inviolate. However, secession remains an option very likely to produce violence, and problems (initially) as bad as the ones it is intended to solve. Whether or not implementing secession is straightforward, the proposal of any community to secede from a state is likely to encourage key elites in the affected states to behave in chauvinistic and warlike ways. Secessionist movements provoke elites satisfied with the existing state into mobilising ~unionist' movements against traitors. It was ironic to watch American commentators warning the Soviet Union during 1990-91 to allow its republics the right to self-determination. Lincoln's heirs had short memories.

What can be said of a general nature about the circumstances under which secession/partitions are likely to be carried out? Three external phenomena matter most and need to be studied closely: the nature of the inter-state system (is it permissive or restrictive?); the aftermath of wars, which often lead to territorial transfers/partitions, often without any consideration of consent; and the disintegration of empires, although this observation is almost tautological.

The important ~internal' phenomena are diverse. People seek full self determination, in the form of independent statehood, for a variety of reasons. The urge may be motivated by a reaction against ethnic discrimination and humiliation, by the pragmatic expectation that the new nation-state will have greater economic and political freedom, by the desire for power and prestige amongst nationalist elites, or to protect a given culture from extinction. Not much of a very general nature can be successfully claimed about the economic circumstances or motivations of full-scale ethnic secessionist movements. One observer notes that secessions are demanded both by economically advanced groups (e.g. Basques, Catalans, Ibos, Lombards, Sikhs, Tamils) and by economically backward communities (East Bengalis, Karens, Kurds, Slovaks); and that the secessionist communities can be located in either backward or advanced regional economies[4]. He claims, however, that backward communities in backward regions are likely to be early rather than late secessionists. Yet even this generalisation, derived from intimate knowledge of African and South Asia, does not withstand the scrutiny of the immediate past. In both the Soviet Union and Yugoslavia the economically advanced communities in economically advanced regions (in the Baltic states and Slovenia and Croatia) led the secessionist vanguard.

Most importantly, enthusiasm for the principle of self-determination flows from the democratisation of the world. Democratisation means that the people are to rule. The statist declares that the people are all those who are resident in a given state or political unit's boundaries (the civic nationalist); the nationalist that they are the nation (the ethnic nationalist). In a few happy cases--e.g. Iceland--these two answers approximately coincide. In most, however, the two definitions of the people do not coincide. In the general case the definition and championing of the people are up for grabs, and the possibility of partition/secession enters into the fabric of any state where ethnic and civic nationalisms may point to different definitions of the nation.

Once democratisation poses the issue of the definition of the people, a clustered set of issues automatically follows. The most important issues are the definition of citizenship, the possession of the franchise, the state's boundaries and the organisational structure of the state. These issues create incentive for political entrepreneurs to build organised interests out of ethnic cleavages, whether at the foundation of the state or afterwards. Politicians in polyethnic states have multiple inducements to play the national or ethic card, whether it be Randolph Churchill playing the Orange card in the UK in the 1880s or Jean-Marie Le Pen playing the Algerian card in France in the 1980s. It does not seem possible to immunise the democratic process to exclude potentially explosive civic and ethnic issues. They are always there for mobilisation by the oppressed or the opportunist (internal and external) or both. Those who lose out politically under existing state-structures and public policies may always choose to redefine the rules of the game by playing the national or ethnic card in the arena of party politics.

A final reason why national and ethnic questions are potentially explosive and raise the possibility that some people(s) will be tempted to exercise self-determination through secession is simple. National and ethnic questions raise relatively non-tradable issues. Nationality and ethnicity (and their potential building blocks, like language, territorial homelands, and historic cultures) are not easily bargained over. They create zero-sum conflicts and therefore provide ideal materials for political entrepreneurs interested in creating or dividing political constituencies.

Having suggested reasons why democratisation increases the likelihood that political actors will seek self-determination for their community and thereby destabilise existing multi-national or polyethnic states, we must make two qualifications. First, destabilisation is likely to be contained if the relevant state or region exists in a milleu or other liberal democratic states. Thus far, in the twentieth century, liberal democracies have never gone to war against one another. Second, there are some circumstances under which the destabilising effects of democratisation upon multi-national or polyethnic states can be muted, and inhibit the impetus to consider secession. These factors include: internal territorial segregation which permits self-government (~good fences make good neighbours'); demographic dominance (where the large group is sufficiently secure not to fear the minority, or minorites, an generous way); demographic stability (where one or more groups are not outgrowing or ~outfalling' one another); and a history of predemocratic cooperation amongst ethnic political elites which gives the post-authoritarian state a reasonable chance of promoting accommodation.

INTEGRATION/ASSIMILATION

A fourth method of macro-political conflict regulation is built upon the idea of trying to eliminate politically relevant differences within the state by seeking to integrate or assimilate the relevant communities into a new transcendent identity, through ~nation-building'. Whereas civic integration has the more modest object of Creating a common civic, national or patriotic identity and citizenship; assimilation aims eventually to create a common ethnic identity through the merging of differences (the melting pot). Where one community adopts the culture of the ~host' or dominant community, we can speak of assimilation through ~acculturation'.

Integration has been the official aspiration of many groups including the African National Congress in South Africa, unionist integrationists and the integrated education lobby in Northern Ireland, and the democratic left in those European countries striving to cope with immigrant influxes. Though inconceivable a few years ago, integration has been embraced by pragmatists in South Africa's National Party who believe that the ecoomic status quo can be secured and improved through liberal democracy, with its protection of individual rights, better than under apartheid.

Integrationists favour policies which reduce the differences between communities, ensuring that the children of the (potentially rival) communities go to the same schools, socialising them in the same language and conventions, encouraging public and private housing policies which prevent segregation, and seeing that the work-place is integrated by outlawing discrimination. Liberal integrationists promote Bills of Rights with equal rights for individuals, rather than communities. Leftist integrationists stress socio-economic policies with trans-ethnic appeal. Assimilationist policiies go further. They favour the merging of ethnic identities, either into one already established identity (e.g. a French identity) or into a new one (e.g. a Soviet or Yugoslav identity). Integrationists and assimilationists also support ~catch-all' political parties, ones with trans-ethnic appeals. The ultimate proof of successful assimilation 's large-scale intermarriage across the former ethnic boundaries which leads first to their blurring and then to their eradication.[5]

Integration and assimilation are driven by both high-minded and intrumental motives .Liberals and socialists associate ethnic pluralism, what we call strategies for the democratic management of national and ethnic differences, with sectarianism, parochialism, narrow-mindedness and chauvinist bigotry. Liberal assimilationists often reject ~special treatment' for ethnic groups as offensive to the ~merit' principle. Canadian integrationists demanded a Charter of Rights after 1945 to prevent a repeat of the war-time internment of ethnic minorities (Japanese, Italians and Ukrainians). White liberals in the USA funded court cases promoting black integration. Other liberals in North America sincerely advocate the assimilation of aboriginal minorities as the best way to end the atrocious conditions on reservations. European socialists struggled to overcome ethnic differences at the turn of the century, because they regarded them as bourgeois devices to impoverish and disorganise the working class. Today, the European left generally espouses the integration of immigrants because it abhors racism and discrimination.

Go to Part Two