Nationalism and Multiculturalism

Chandran Kukathas. Handbook of Political Theory. Editor: Gerald F Gaus & Chandran Kukathas. Sage Publication. 2004.

Diversity and Contemporary Political Theory

If any issue dominates contemporary political theory, it is how to deal with cultural diversity and the claims—moral, legal, and political—made in the name of ethnic, religious, linguistic, or national allegiance (Kymlicka, 2001: 17). Today, governments are confronted by demands from cultural minorities for recognition, protection, preferential treatment, and political autonomy within the boundaries of the state. Equally, international society and its political institutions, as well as states themselves, have had to deal with demands from various peoples for political recognition as independent nations, and for national self-determination. The turbulent politics of the contemporary world may account in part for this development: the collapse of communist Eastern Europe led to an upsurge of nationalist demands from peoples aspiring to statehood; the challenges to the legitimacy of rulers in such places as Kashmir, Burma, East Timor, and Bougainville have fed demands for national independence as well as attempts at secession; the emergence of an indigenous peoples’ movement gave further encouragement to aboriginal groups calling for affirmative action, or compensation for past injustice, as well as rights of self-government; and the mass migrations of peoples, fleeing war or simply seeking better opportunities in new countries, have seen the emergence of substantial cultural minorities in states unprepared for the problems this could bring.

Political theory had, until recent times, said relatively little about these matters (exceptions include Plamenatz, 1960; Van Dyke, 1977; 1982; 1985). But the issues raised by cultural diversity and nationalist claims could not be ignored for long, since they posed a challenge to the prevailing political theories—and to liberal and democratic theory in particular. Indeed, the challenge of multiculturalism and nationalism has provoked a re-examination of a great number of issues in political theory, from the role of the state, the limits of toleration and the rights of women, to the proper scope of public education and the nature of citizenship. It has brought about a reconsideration of the basis of political order.

This chapter surveys the literature of nationalism and multiculturalism as it has grown and developed over the past 15 years. Its aim, however, is not simply to summarize that body of writing but to draw attention to the problems that have confronted contemporary political theory—and liberal theory in particular—as it has struggled to embrace diversity. How can the many live as one? That is an old question in political theory, and the theorists of nationalism and multiculturalism have, in different ways, tried to offer an answer.

What is Multiculturalism?

The term ‘multiculturalism’ predates its use in political theory, but not by very long. Although nationalism is an old concept which has been much discussed in the past century (Kedourie, 1967; Minogue, 1967), multiculturalism did not appear until the 1960s and 1970s, when it was used to describe a new public policy, first in Canada and then in Australia. In both of these cases, this development marked an explicit movement by federal governments away from policies of assimilation of ethnic minorities, and immigrants in particular, toward policies of acceptance and integration of diverse cultures (Lopez, 2000: 2-3). The term did not enter the American (or British) lexicon until the 1980s (Glazer, 1997: 8). When it did enter American debates, however, it did so in the first instance, in discussions about public education. ‘Multiculturalism,’ according to Nathan Glazer, ‘is just the latest in [a] sequence of terms describing how American society, particularly American education, should respond to diversity’ (1997: 8).

Multiculturalism, then, is a term that describes one particular way of responding to ethnic diversity. ‘It is a position that rejects assimilation and the “melting pot” image as an imposition of the dominant culture, and instead prefers such metaphors as the “salad bowl” or the “glorious mosaic”, in which each ethnic and racial element in the population maintains its distinctiveness’ (1997: 10). Yet in reality there is no single multiculturalist position but rather a range of views of what multiculturalism requires. For some, multiculturalism requires moderate changes to social and political institutions to enable cultural minorities to preserve their languages and their distinctive customs or practices. For others, however, multiculturalism requires much greater social transformation to turn modern society into one in which racism has been eliminated and ‘difference’ is nurtured rather than repudiated, or simply tolerated.

But if multiculturalism is a way of embracing diversity, this still leaves open the question of how diversity is to be embraced. If a multicultural society is one in which different religions, cultures, languages, and peoples can coexist without some being subordinated to others, or to a single, dominant group, how can this be achieved, and what principles would describe such a society? This issue arises because even if there is diversity, there must surely be some kind of unity for a society to exist. Unless we aspire to a borderless world, in which people could move freely unimpeded by national (and other) boundaries, even a multicultural society would have to settle on some basic institutions, decide what it would accept as official languages, and define itself as a nation, membership of which it controls by determining who (and how many) may join it. The real question, in other words, is what does multiculturalism mean in practice?

This question, however, was not addressed systematically until the 1990s when political theorists began to consider what might be the principled basis of a multicultural society. It was only then that the case for multiculturalism began to receive any kind of sustained defence—and criticism.

Multiculturalism Defended

Kymlicka’s Theory

The first systematic theory of multiculturalism was developed by Will Kymlicka in two major works: Liberalism, Community and Culture (1989) and Multicultural Citizenship (1995a). This field of inquiry has also been shaped by Kymlicka’s other writings (Kymlicka, 2000) and edited collections (Kymlicka, 1995b; Kymlicka and Shapiro, 1997; Kymlicka and Norman, 2000b). Kymlicka’s work was born out of a dissatisfaction with the political theory of post-war liberalism which, in his view, had wrongly assumed that the problem of national minorities could be resolved by ensuring the provision of basic individual rights. Just as religious minorities were protected by the separation of church from state, and the entrenching of freedom of religion, so would ethnic identity be protected by freedom to express in private life those cultural attachments that were no business of the state. The state would neither oppose nor nurture the freedom people enjoyed to express their attachments to their particular cultures, but respond with what Glazer called ‘benign neglect’ (1975: 25; 1983: 124; Kymlicka, 1995a: 3). But benign neglect, in Kymlicka’s view, was not a plausible option. The question is, why?

According to Kymlicka, minority rights could not simply be subsumed under human rights because ‘human rights standards are simply unable to resolve some of the most important and controversial questions relating to cultural minorities’ (1995a: 4). These included questions about which languages should be recognized in the parliaments, bureaucracies and courts; whether any ethnic or national groups should have publicly funded education in their mother tongue; whether internal boundaries should be drawn so that cultural minorities form majorities in local regions; whether traditional homelands of indigenous peoples should be reserved for their benefit; and what degree of cultural integration might be required of immigrants seeking citizenship (1995a: 4-5). Traditional human rights doctrines, Kymlicka suggests, simply give us no guidance on these questions. And unless they are supplemented with a theory of minority rights, human rights theory will not enable us to address some of the most pressing issues confronting us in places like Eastern Europe, where disputes over local autonomy, language, and naturalization threaten to leave those regions mired in violent conflict. Kymlicka’s ambition, therefore, has been to develop a liberal theory of minority rights that explains ‘how minority rights coexist with human rights, and how minority rights are limited by principles of individual liberty, democracy, and social justice’ (1995a: 6).

The theory ultimately advanced by Kymlicka distinguished three kinds of minority or group-differentiated rights that were to be accorded to ethnic and national groups: self-government rights, polyethnic rights, and special representation rights. Self-government rights require the delegation of powers to national minorities, such as indigenous peoples, but these rights would not be available to other cultural minorities who had immigrated into the country. The latter would be eligible for polyethnic rights, which guarantee financial support and legal protection for practices peculiar to some ethnic or religious groups. Both indigenous peoples and immigrant minorities might also be eligible for special representation rights which guarantee places for minority representatives on state bodies or institutions. Central to Kymlicka’s account of group-differentiated rights is a distinction between two kinds of minorities: national minorities and ethnic minorities. National minorities are peoples whose previously self-governing, territorially concentrated cultures have been incorporated into a larger state. Examples include ‘American Indians’, Puerto Ricans, Chicanos, and native Hawaiians in the United States; the Quebecois and various aboriginal communities in Canada; and the Australian Aborigines. Ethnic minorities, however, are peoples who have immigrated to a new society and do not wish to govern themselves, but nonetheless wish to hold on to their ethnic identities and traditions. A modern state may be ‘multicultural’ in one (or both) of two senses: either because it is ‘multinational’, since its members belong to different nations; or because it is ‘polyethnic’, since its members emigrated from different nations (1995a: 18). In Kymlicka’s theory, these two kinds of groups have very different legitimate claims to make, and understanding this should make clear that national minorities need not fear that policies of multiculturalism would reduce them to the status of migrants, just as other citizens need not fear that multiculturalism implied that immigrants had a legitimate claim to self-government.

At the heart of Kymlicka’s theory of multiculturalism is a form of nationalism—or liberal nationalism, to be precise. It is his contention that the liberal tradition has a history of recognizing group-differentiated rights. This is most evident, in his view, in the fact that most liberal theorists accept that the world is made up of separate states. These states are normally assumed to have the right to decide who may enter their jurisdictions to visit, reside, or acquire citizenship. Kymlicka’s view is that ‘the orthodox liberal view about the right of states to determine who has citizenship rests on the same principles which justify group-differentiated citizenship within states, and that accepting the former leads logically to the latter’ (1995a: 124). That is to say, citizenship or state membership is itself a group-differentiated notion, and liberalism is a view that recognizes the rights of individuals as members of states. It therefore makes perfect sense for liberals to be willing to recognize groups within states, for groups, like states, exist to protect people’s cultural membership. What liberals defend is individual freedom. Yet this is ‘not primarily the freedom to move beyond one’s language and history, but rather the freedom to move around within one’s societal culture, to distance oneself from particular cultural roles, to choose which features of the culture are most worth developing, and which are without value’ (1995a: 90-1). National cultures are ‘societal cultures’, and the modern world is divided into such groupings. They provide their members with meaningful ways of life across the range of human activities—from the economic to the educational and religious. ‘These cultures tend to be territorially concentrated, and based on a shared language’ (1995a: 76). These are ‘societal’ cultures because they comprise not just shared memories or values but also common institutions and practices. A ‘societal culture’ is embodied in schools, in the media, in the economy, and in government.

For Kymlicka, national minorities are, typically, groups with societal cultures—albeit cultures that have struggled against conquest, colonization, and forced assimilation. Immigrants, however, have no societal culture (though tthey may have left their own societal cultures). Societal cultures tend to be national cultures, and nations are almost invariably societal cultures (1995a: 80). In the modern world, cultures which are not societal cultures are not likely to prosper, given the pressures towards the creation of a single common culture in each country. His theory of group-differentiated rights accordingly focuses on enabling national minorities to sustain their societal cultures, while protecting immigrants with polyethnic rights that would ‘help ethnic groups and religious minorities express their cultural particularity and pride without it hampering their success in the economic and political institutions of the dominant society’ (1995a: 31). Nonetheless, both kinds of group-differentiated rights have something in common: they afford groups protection against the impact of external decisions without granting the group any right to make internal restrictions on their members as it sees fit. His argument is that ‘liberals can and should endorse certain external protections, where they promote fairness between groups, but should reject internal restrictions which limit the right of group members to question and revise traditional authorities and practices’ (1995a: 37). What group-differentiated rights are granted, then, depends on whether the particular multinational, polyethnic, or special representation rights in question provide ‘external protections’, or enforce ‘internal restrictions’.

Problems of Liberal Multiculturalism

Kymlicka’s defence of group-differentiated rights immediately raised a range of questions and problems, and the literature on multiculturalism over the past decade has tackled many of them. The first issue to be addressed was the question of whether groups could properly be the bearers of rights. To some it was plain that they could not: only individuals could have rights (Narveson, 1991; Hartney, 1991). According to one view, groups were fictitious entities—and fictitious entities could not be rights bearers (Graf, 1994: 194). Yet in spite of such reservations, political theory has in recent years (with the rise of multiculturalism) become much more sympathetic to the idea of group rights.

Even before multiculturalism acquired its current prominence, however, some philosophers had already advanced accounts of group rights. Joseph Raz (1986: 207-8), for example, in his influential account of rights leaves space for collective rights. Larry May (1987: 180), while remaining cautious about the extent to which groups should be recognized as rights holders, argued that moral theorists needed to examine more closely the actions and interests of social groups as possible bearers of rights and responsibilities. And Frances Svensson (1979) had earlier suggested that group rights were needed to do justice to the claims of native peoples. Nonetheless, theorists (or critics) of multiculturalism did not always mean the same thing when they invoked group rights or ‘cultural’ rights. The most helpful elucidation of the different kinds of rights claims made on behalf of cultural groups was offered by Jacob Levy (1997: 24-5), who distinguished eight categories of rights. These include exemption rights (exempting groups from laws that burden their cultural practices), assistance rights (to do those things the majority can do unassisted), self-government rights, rights to impose external rules (say, restricting non-members’s rights to buy property or restricting their right to use their own language), rights to enforce internal rules (even if they violate other rights), rights of recognition of the group’s legal code, rights of representation in government, and rights to symbolic claims to acknowledge the worth, status, or existence of the group (1997: 25).

The consensus of opinion is that it is quite possible for groups to have rights, or for rights to be accorded both to groups and to individuals on the basis of identity. A group may hold a right as an independently recognized entity; and individuals may hold particular rights because they are members of particular collectivities. Nonetheless, this issue has remained controversial because of the implications of granting rights on the basis of group membership. As Peter Jones put it, ‘Group rights are often articulated as demands for group freedom, but they are also feared as vehicles for group oppression’ (1999: 354). Thus Raz’s view of group rights, though widely accepted (Brett, 1991; Freeman, 1995; Margalit and Halbertal, 1994), has been criticized for being too capacious in as much as it identifies groups as no more than collectivities of individuals who share nothing more enduring than an interest in a matter (Réaume, 1988; 1994; Jones, 1999: 359). Yet even if we draw a distinction between collective rights and corporate rights (Jones, 1999) there remain other concerns about the need for, or wisdom of, granting rights to groups. For one thing, it is not clear that the identity of the group is readily established, since groups are not only changeable but also often composed of other, smaller groups. To recognize groups is often to reify or entrench formations that might otherwise be temporary, and also to empower those who have authority to speak for the group (Kukathas, 1992a; 1992b; 2003a). Equally, the way in which boundaries distinguishing groups are drawn may be contentious because people may not be happy about being excluded, or included, within a designated collective (Offe, 1998).

Further issues arise, however, once we begin to consider the content of group rights. The demands of some groups for rights in the form of exemptions, for example, have generated a substantial debate about the implications of such special rights. This debate becomes especially vigorous, however, when particular issues become salient: religion, education, and children. While most liberal defenders of multiculturalism have been ready to grant cultural minorities the right to live by their own beliefs, children and education have raised special problems. For many, the limits of multiculturalism are set by the need to protect the interests of children, which override even the rights of parents or communities to inculcate their own religious beliefs. There is considerable tension here within liberal theories of minority rights in particular, as liberals have sought both to respect cultural minorities and to demand of those minorities that they abide by certain liberal strictures.

Kymlicka (1995a :163), for example, recognizes the dilemma liberals face here, but suggests that in the end children need to be educated so that they choose for themselves the paths they will take. Others, however, have been more insistent that the education of children is without doubt of fundamental importance, both from the perspective of the individual child and from the perspective of the liberal state—and should take priority over religious or other claims. Education is education for citizenship. Stephen Macedo, for example, argues that those who embrace multiculturalism should not forget that ‘liberal citizens do not come into existence naturally’, that diversity ‘must be constituted for liberal democratic purposes’. Children, in his view, must be educated so that they become liberal citizens (Macedo, 1995a: 68; and also Macedo, 1995b; 2000). A similar view is advanced by Amy Gutmann, even as she is at pains to emphasize the importance of a multicultural education and the dangers of a schooling that ignores the diversity of traditions found in a society. Education must, in the end, be education for democratic citizenship, even if not only for citizenship (Gutmann, 1996). That it will also tend to assimilate minorities, and work toward the transformation of religious communities, cannot be denied, and so must be accepted (Walzer, 1995: 29). As Eamonn Callan observes, ‘schools must somehow honour both the interest in identity formation that rightly belongs to parents and the interest we all share as members of a civic community’ (2000: 66; see also Callan, 1997).

The tension between the claims of state and religious community when the treatment and education of children are at issue has surfaced on numerous occasions in legal cases, which have in turn generated considerable debate in political theory. In 1972, in the case of Wisconsin v. Yoder, the United States Supreme Court decided in favour of Old Order Amish parents who wished to withdraw their children from Wisconsin state schools after eighth grade, two years earlier than statutory requirements for compulsory education permitted. A number of theorists have argued that this was a poor decision, either because it neglects the interests of the child, though this should not be exaggerated since only two additional years of education were at issue (Feinberg, 1980; Gutmann, 1980), or because it fails to recognize the importance of education for citizenship (Arneson and Shapiro, 1996). Others, however, have argued that the liberal state should resist usurping parental authority in order to impose its opinion on what is the best way of life for the child (Burt, 1996: 432).

An equally significant amount of ink has been spilt discussing the somewhat different case of Mozert v. Hawkins, which involved an (unsuccessful) attempt by Christian fundamentalist parents to win an exemption from Tennessee state regulations requiring public school children to be taught from textbooks that exposed them to a variety of values. Once again, liberal theorists have been divided on the limits of parental authority and the scope for exemptions based on religious or cultural beliefs (compare Callan, 1997 and Tomasi, 2001). One of the dilemmas posed by such circumstances is whether to bear the costs of granting exemptions or the costs of refusing them. If the cost of granting parents exemptions is that their children will not be exposed to a diversity of views (which, presumably, would make them better citizens), the cost of denying parents exemptions might be that more parents decide to home school their children, thereby cutting them off even more seriously from the democratic mainstream (Reich, 2002).

The conflict between state and religion or culture surfaced in a different, though no less controversial, form in France in 1989 in the so-called ‘head-scarves affair’. In this instance, a problem arose because three North African immigrant women in a French public secondary school chose to wear their headscarves in class, in a gesture that was interpreted as a challenge to the national policy of secularism in schools. The headscarves were regarded as a form of (Muslim) religious dress, and when the French education minister insisted that the pupils be readmitted to class many objected that this amounted to buckling before the power of religious fanaticism. As Bhikhu Parekh notes, this issue ‘went to the heart of the French conceptions of citizenship and national identity and divided the country’ (2000: 250). But it also divided political theorists (Galeotti, 1993; 1994; Moruzzi, 1994a; 1994b). In this, as with other controversies surrounding the matter of dress, the problem is that dress is not unambiguously a private matter. It is complex enough when turbaned Sikhs seek exemption from laws mandating the use of motorcycle helmets. In the headscarves case, however, the problem was deepened by the French educational system and its philosophical principle, laïcité, which demands state neutrality towards ‘all kinds of religious practices, institutionalized through a vigilant removal of sectarian religious symbols, signs, icons, and items of clothing from official public spheres’ (Benhabbib, 2002: 95-6). How this was to be squared with other public commitments to freedom of religion and liberty of conscience, as well as personal liberty, became entirely obscure.

On this, as on many other occasions, the liberal contention that individuals should be left free to live by their own lights in matters that are of private and not public concern does not help to resolve things. Even the matter of what one eats has a public dimension since there are laws governing the treatment of animals, and in particular the slaughtering of animals for human consumption. Religious demands for kosher or halal meat go against laws providing for the humane slaughter of animals in Europe. And to the extent that religious and cultural groups can gain exemptions to allow ritual slaughter or killing for sport, multiculturalism turns out not only to be bad for animals but problematic for political theory (Casal, 2003). What is to be regarded as a public issue and what as private itself becomes a matter of political and philosophical disagreement. This is even more evidently the case when disputes centre on state symbols, the official status of languages, and the timing of holidays.

One of the reasons such issues become problematic is that diversity is sometimes sustained by continuing immigration which, in many countries, has led to the growth of significant ethnic or cultural minorities within a host society. Immigration might thus further complicate the cultural landscape when a society is already composed of a settler population and an indigenous minority, and perhaps a significant minority population among the settler society. Canada is an obvious example, as a polity dominated by the descendants of English-speaking settlers, with a substantial French minority, various indigenous peoples and a significant immigrant community of people from across the world. But most of Europe and North America is now marked by a similar diversity, accentuated by immigration by peoples from the Third World. While Kymlicka’s philosophical response to this has been a theory of group-differentiated citizenship, with specific rights for immigrant and indigenous minorities, others have responded with calls for a slowing or halting of immigration from culturally different people (Brimelow, 1995) or restricting the granting of citizenship to those who have more completely assimilated into the ways of their new society (Pickus, 1998). For some, the nation-state is indeed the expression of a specific ethnocultural group, and to try to create a multicultural state is therefore a mistake (Auster, 1992).

Conservative reservations about immigration notwithstanding, immigration policy is unlikely to change substantially enough to alter the fact that migrants will continue to add to cultural and ethnic diversity, particularly in the developed West. Immigration will therefore continue to shape multicultural policy, and so multicultural theory (Kukathas, 2003c). At this point, the strains in multicultural policy also start to become evident in political theory, and in liberal political theory in particular. In part, this is because immigration is itself problematic from a liberal point of view and political theorists are divided on the question of how free people should be to move from country to country. While some favour open borders (Dowty, 1987; Carens, 1987; 1992; 2000; Goodin, 1988; 1992), others are less sure of the wisdom of letting the liberal state throw open its society to all-comers, particularly if that might threaten to undermine the liberal state (Buchanan, 1995).

Unsurprisingly, then, much of the debate about multiculturalism has been a debate about citizenship. The question is, how can it be possible to admit a diversity of people into a society, and allow (or even encourage) them to retain their own cultural traditions or customs, and still preserve a polity governed by, and respectful of the rights of, citizens united by a common allegiance? The dilemma is that the more robust the conception of citizenship, the less accommodating must the polity be of cultural diversity, to the extent that it cannot tolerate cultural traditions that do not value citizenship. The greater the diversity it wishes to admit, the weaker must be the demands of citizenship the polity imposes upon its members (Kukathas, 1993; 2003d: 72-5). Or citizenship may have to be rethought completely (Kymlicka and Norman, 1994; 2000a).

Alternative Theories

While Kymlicka’s work has dominated the landscape of the political theory of multiculturalism, this is not for any lack of other contributors. On the contrary, a number of other theorists have offered their own accounts of multiculturalism, some agreeing with Kymlicka’s general standpoint while disagreeing on particular questions, others offering entirely independent theories of multiculturalism, or rejecting Kymlicka’s ideas altogether.

Among the most important of these alternative theories is that offered by Charles Taylor, in his political writings generally but, more particularly, in his influential essay ‘The politics of recognition’ (1994). Taylor rejects as inadequate what might be called the liberal theory of multiculturalism, for liberalism, in his view, is incapable of giving culture the recognition it requires. Liberalism offers to recognize individuals as the bearers of rights and the possessors of dignity as equal citizens, regarding each person as essentially the same. But what many cultural groups want is recognition not of their sameness, but of their distinctness. Out of such desires, according to Taylor, grew a philosophical alternative to liberalism: the politics of difference. This view is sceptical about the pretensions of liberalism to offer neutral or difference-blind principles that are more than simply reflections of the standards of the dominant culture. Taylor thus rejects the efforts of Kymlicka to develop a liberalism that might accommodate difference by granting individuals differential rights to enable them to pursue their particular cultural ends. For him, the problem with this solution is that it works only ‘for existing people who find themselves trapped within a culture under pressure, and can flourish within it or not at all. But it does not justify measures designed to ensure survival through indefinite future generations’ (1994: 62). It cannot, for example, justify the collective goals of the Québecois, whose aim is the long-term survival of the French-speaking community in Canada.

A number of other theorists have developed arguments about how cultural diversity might be accommodated by giving greater recognition to ‘difference’ rather than extending the scope or range of liberal rights (Baumeister, 2000). James Tully’s Strange Multiplicity (1995), for example, offers a reconstruction of modern constitutionalism that is able to accommodate a greater variety of cultural traditions, and adapt elements from some of them to enhance the quality of liberal constitutional arrangements. In many of these cases, defenders of the politics of difference present an approach to cultural diversity which not only criticizes liberal individualism but also advocates a greater emphasis on the extension of democratic processes to give greater scope to the participation of cultural minorities in the shaping and governing of the polity (see Young, 1990; 2000; Phillips, 1995; Devaux, 2000; Williams, 1998; Tully, 2003).

Yet divisions exist not only between liberal defenders of multiculturalism and their critics but also among liberal theorists themselves. Two major interrelated issues have shaped debate among them: the extent to which diversity ought to be tolerated by liberals when minorities turn out to be illiberal in character, and the principled basis of liberal acceptance of cultural diversity. For some, the limits of liberal toleration are clear: toleration is not extended to illiberal minorities. For Kymlicka, for example, liberalism endorses group-differentiated rights which provide for external protection for groups, but does not permit ‘internal restrictions’: groups may not curb the basic civil rights of their members. Indeed, for Kymlicka (1989; 1995a) what liberalism protects, above all, is the individual’s capacity for autonomous choice; culture is important because it is the context within which individuals learn how to choose, but its value diminishes when it ceases to enable individuals to choose their lives for themselves. A number of other liberal theorists concur with Kymlicka in this matter, arguing that liberalism protects autonomy, and that cultures that do not value or promote autonomy are less deserving of toleration or, at best, should be tolerated on pragmatic rather than principled grounds (Fitzmaurice, 1993; Levey, 1997; Gill 2001).

Other liberals, however, are less enamoured of autonomy. Some, like Jeff Spinner-Halev, consider autonomy to be valuable, but are critical of those who over emphasize its importance or define autonomy so strictly that many ways of living do not qualify (Spinner-Halev, 2000: 62-7; Spinner, 1994). Others, however, have been more critical still of autonomy, suggesting that toleration or respect for diversity are much more important considerations for liberals (Galston, 1995; Kukathas, 1992a; 1999; 2003a; for an analysis of this liberal divide see Levy, 2003). Kukathas (1997; 2001; 2003b), in particular, has argued vigorously that toleration is so important a liberal virtue that a liberal order will tolerate a diversity of cultures even if some of them are highly illiberal. What a good society protects is freedom of association, not autonomy. And for as long as individuals are free to exit the arrangements or communities or groups within which they find themselves, that order is legitimate—even if it might be one in which many groups or communities are highly illiberal in as much as they are themselves intolerant of diversity. This view, however, gives no particular rights to groups as such, and denies them the external protections advocated by Kymlicka and others; though it also denies outside authorities any right to intervene to lift internal restrictions imposed by such communities upon their members.

This issue of the treatment of minorities within minorities has itself become the subject of considerable debate, many arguing that respect for minorities cannot become grounds for accepting the mistreatment of internal minorities (Green, 1994). Levy, in particular, has offered a treatment of this matter which is sensitive to the claims of minorities seeking to live by their own cultural traditions but nevertheless robust in its rejection of claims to perpetuate practices that are cruel or hateful. The Multiculturalism of Fear, borrowing from Judith Shklar’s reading of the liberal tradition emphasizing the significance of Montesquieu, insists that a political theory of multiculturalism must be ‘centrally concerned neither with preserving and celebrating ethnic identities nor with overcoming them’ but with ‘mitigating the recurrent dangers such as state violence toward cultural minorities, interethnic warfare, and intra-communal attacks on those who try to alter or leave their cultural communities’ (Levy, 2000: 12-13). This theory tries to steer a course between condemning cultural identification and insisting that all minorities become good Millian liberals, and condemning liberalism for failing to be sufficiently hospitable to diversity.

Multiculturalism under Fire

Not all theories, however, have been entirely sympathetic to multiculturalism, particularly when the price of cultural diversity has looked like being acceptance of illiberal or tyrannical practice. According to Stanley Fish (1998: 73-5), no one could genuinely advocate multiculturalism because that would require tolerating all cultures, including those determined to stamp out tolerance. Most multiculturalists are thus not ‘real’ but ‘boutique’ multiculturalists, prepared to tolerate difference for as long as differences are trivial. But regardless of whether Fish’s argument is sound—and it is at least debatable, in light of the long history of debate over the question of the toleration of the intolerant (Heyd, 1996)—some theorists have concluded that when multiculturalism comes into conflict with more important fundamental values, multiculturalism will have to yield. Two critiques of multiculturalism are particularly worthy of note, the first coming from the liberal egalitarian perspective and the second from a feminist point of view. A third, coming from an aboriginal perspective, is perhaps also worth considering.

The Liberal Egalitarian Critique

The most comprehensive liberal egalitarian critique of multiculturalism has been offered by Brian Barry in his book Culture and Equality (2001). According to Barry, multiculturalism is inconsistent with liberalism and a respect for liberal values and should therefore be rejected. Attempts to show that it is consistent with liberalism are, in his view, implausible. In this regard, he rejects what William Galston has termed ‘Reformation liberalism’. Unlike ‘Enlightenment liberalism’, which emphasizes the importance of individual autonomy, ‘Reformation liberalism’, Galston maintains, values diversity and sees the importance of ‘differences among individuals and groups over such matters as the nature of the good life, sources of moral authority, reason versus faith, and the like’ (1995: 521). Barry rejects this distinction, but is especially critical nonetheless of those who are members of the diversity-promoting liberalism camp.

Barry rejects three major arguments advanced in support of Reformation liberalism. The first is that liberal theory values respect for persons and this implies respect for the cultures to which individuals belong. To this Barry replies that illiberal cultures often violate the requirement of equal respect and to that extent they do not deserve respect (2001: 128). The second argument is that liberalism values diversity because it increases the range of options available to individuals. To this Barry responds that liberals prize individuality rather than diversity (2001: 129). The third argument is that liberalism attaches great importance to the public/private distinction, and so should be committed to nonintervention in the private realm. To this Barry replies that liberalism has historically challenged the sanctity of parental and paternal authority, and sought to protect individuals from the groups to which they belong. This does not mean that, for Barry, liberalism requires every group to conform to liberal principles. Individuals must be free to associate in any way they like (consistent with the law protecting the interests of those outside the association). But there are two important conditions: all participants in the association should be sane adults, and their participation should be voluntary (2001: 148). Groups may then do as they please, provided those who do not like the way a group’s affairs are run are able to exit without facing excessive costs (2001: 150).

Barry’s view imposes serious constraints, then, on the operation of groups. In the end, what it tolerates is only what Fish calls ‘boutique multiculturalism’. It requires that illiberal practices not be condoned, that parents be required to send their children to school, and that generally the state ensures that children are appropriately educated and not made the victims of creationists and religious zealots—even if they are their parents. Equally, multiculturalism provides no warrant, in Barry’s view, for allowing religious groups to be exempt from the legal requirements for the humane treatment of animals. Jewish and Muslim demands for kosher or halal meat are indefensible, and diversity provides no warrant for making an exception (2001: 40-6). In the end, Barry’s view amounts to a reassertion of liberal egalitarianism as a doctrine that is simply incompatible with multiculturalism. (For criticisms of Barry see the papers in Kelly, 2002; for another defence of liberal egalitarianism see Kernohan, 1998.)

The Feminist Critique

If multiculturalism is inconsistent with liberal egalitarianism, it is perhaps even more at odds with contemporary feminism. One of the most important objections to multiculturalism is that, in seeking exemptions or special rights for cultural groups or religious communities and organizations, it in effect seeks protection for groups whose practices are sexist and highly disadvantageous—if not altogether harmful—to women. This view has been put most forcefully by Susan Okin (1998; 1999a; 1999b; 2002), who has taken issue with almost all of the most prominent defenders of multiculturalism, and found their commitment to women’s rights and interests wanting.

Multiculturalism is in tension with feminism because the two ideas represent political visions that stand some way apart. As Katha Pollit puts it, ‘In its demand for equality for women, feminism sets itself in opposition to virtually every culture on earth… multiculturalism demands respect for all cultural traditions, while feminism interrogates and challenges all cultural traditions’ (1999: 27). Feminist critics of multiculturalism thus not only ask why groups which do not accord women equal opportunity, or even equal dignity, should be given special rights or protections, but also why the liberal state fails to intervene in such cultural communities to ensure that women are not denied education, forced into marriage or made the victims of bodily mutilation. Why should a cultural group be entitled to try to live by its ways if these ways violate the individual rights of their members? ‘Why shouldn’st the liberal state, instead, make it clear to members of such groups, preferably by education but where necessary by punishment, that such practices are not to be tolerated?’ (Okin, 1998: 676). Thus when writers such as Margalit and Halbertal (1994) defend public funding of religious education for ultra-orthodox Jews on the basis of the right to culture, feminists like Okin (1999b: 131) ask how this can be defensible when the corollary of this practice is an education for girls that is oriented towards facilitating the religious life of boys.

There is no doubt that feminism and multiculturalism come into conflict, for precisely the reasons that Okin has identified. But the fact of this conflict does not establish whether one philosophical stance or the other ought to prevail (Kukathas, 2001). Some writers, however, have tried to argue that multicultural accommodation need not be incompatible with feminist concerns. The most notable contribution to this position has come from Ayelet Shachar, who argues that it is a mistake to think of multiculturalism simply in terms of the granting of ‘external protections’ to cultural groups. Since individuals are typically members of many groups, the question is how to ‘allocate jurisdiction to identity groups in certain legal arenas while simultaneously respecting group members’ rights as citizens’ (Shachar, 2001: 27-8). The fact that individuals are members of multiple groups holds out the hope that power might be divided among a number of jurisdictions, enabling women both to secure protection against the power of particular groups and to retain the capacity to participate in the cultural traditions they cherish.

Other writers have also sought ways to reach some solution to the tension between feminism and multiculturalism. Some have concluded that some form of differentiated citizenship will need to be developed if the claims of women and the claims of culture are to be mediated (Benhabib, 2002: 82-104). Others have suggested that a dialogic solution, forswearing the appeal to individual rights or procedural justice, offers a better prospect of reaching an accommodation of cultural values and women’s interests (Eisenberg, 2003).

The Aboriginal Critique

Generally, multiculturalism is assumed to speak not only for the interests of immigrant cultural minorities but also for the aboriginal peoples who are minorities in modern states. Canada, Australia, New Zealand, and the United States, no less than Fiji, Malaysia, Indonesia, India and most of South and Central America, are home to peoples whose ancestry may be traced back to premodern times, and their interests are sometimes thought to be addressed by the development of the institutions of a multicultural society. Yet for many indigenous peoples multiculturalism is less than welcome, for its implication is the further marginalization of their communities and culture in a modern state more attuned to the needs of migrants than to those of aborigines.

The recognition of this issue has shaped the development of Kymlicka’s theory, which is particularly aware of the distinctive concerns of indigenous peoples. His model of group-differentiated rights deliberately makes space for national minorities, as distinct from polyethnic groups. Whether or not Kymlicka’s theory is defensible, however, aboriginal groups around the world have pressed the case for the rights of indigenous minorities. (For a sceptical assessment of the notion of indigenous rights see Mulgan, 1989a. Mulgan, 1989b also suggests that, in the case of New Zealand, the land is occupied by two indigenous peoples: the Maori and Pakeha, or descendants of white settlers.) Moreover, many indigenous groups have insisted that, unlike immigrant peoples, what they need is not only recognition of their independent status but also rectification for past injustice.

Extended treatments of the problem of incorporating aboriginal peoples into modern liberal democratic society, in a way that respects the integrity of aboriginal traditions, have been offered by Tully (1995) and, more recently, Ivison (2002). Both suggest that a viable liberal order requires the establishment of a constitutional modus vivendi that incorporates recognition of aboriginal custom and culture. However, as Ivison argues, mere incorporation of indigenous law may not be enough given that circumstances vary and both society and indigenous societies are themselves changing (2002: 141-62).

The problem of rectification for past injustice, however, remains a serious difficulty, particularly when the effluxion of time has made the matter of ascribing to present generations responsibility for past injustice a difficult one, morally, legally, and politically. Jeremy Waldron (1992), for one, has suggested that public policy should focus on future welfare rather than past injustice if the aim is to do justice to the concerns of aboriginal people (see also Sher, 1981; Goodin, 2001). Though others have offered theories of rectification that might do justice to the demands of aboriginal peoples (Kukathas, 2003a; Hill, 2002), it seems unlikely that those demands will ever be met philosophically, let alone politically. In this regard, the distinctive position of aboriginal peoples may not survive the advance of multiculturalism, even if aboriginal peoples remain reluctant to be content with the status of one minority among many.


The emergence of multiculturalism over the past three decades has been coterminous with the re-emergence of nationalism, both as a political phenomenon and as a topic for historical and philosophical investigation. Multiculturalism as a public policy has been, at least in part, a response to nationalist demands of a sort—at least to the extent that cultural groups have begun to demand some form of recognition of their distinctive identity, even if they have not always demanded rights of self-government or independence. Unsurprisingly, then, concerns about culture have prompted a rethinking of theories of nationalism, no less than have concerns about nationality helped to shape theories of multiculturalism. (The literature on nationalism is vast, but the focus here is on normative theories of nationalism.)

Defining and Defending Nationalism

Nationalism, according to Margaret Moore, is ‘a normative argument that confers moral value on national membership, and on the past and future existence of the nation, and identifies the nation with a particular homeland or part of the globe’ (2002: 5). In her account, nations are moral communities characterized by bonds of solidarity and mutual trust, and the attachment people feel to such communities is reason enough to recognize national identity. This very recent account of nationalism takes issue with a number of prominent theories such as Ernest Gellner’s, which argued famously that ‘nationalism is primarily a political principle, which holds that the political and national unit should be congruent’ (1983: 1). The problem with this view is that it implies that every nationalist movement seeks independence and political separation. Yet there are many groups which are nationalist in character but do not demand statehood, and would be content with greater freedom from external control within the existing state (Moore, 2001: 4).

In Moore’s view, nations are moral communities marked by bonds of solidarity and mutual trust. Thus they are not grounded in culture, for national identity should not be confounded with a common culture. While nationalists seek to preserve political communities, this does not mean that they seek to preserve their cultures. In this regard, Moore’s account is at odds with the arguments of liberal nationalists such as Kymlicka, Margalit, Raz, and Yael Tamir, who see nationality as grounded in culture (Kymlicka, 1995a; Raz, 1994; Margalit and Raz, 1990; Tamir, 1993). It has perhaps more in common with Goodin’s (1997) suggestion that group attachment is best explained in Bayesian terms, as conventions arising out of an unwillingness of people to expend scarce resources to question the prejudices and presuppositions they grow up with inside their own groups (for a similar analysis see Kukathas, 2002).

The definition, and also the sources, of nationalism are much disputed, some seeing it as the product of modernity and others as its cause. (See the differing historical accounts of Gellner, 1983; Greenfeld, 1992; and Anderson, 1993.) Similarly, the question of the justifiability of nationalism has been much argued about among political theorists. Among liberal theorists in particular, nationalism is viewed with suspicion, since its emphasis on community and belonging puts it at odds with liberal commitments to individual rights and to freedom and equality as universal values. Often, they are inclined to give it only a qualified endorsement (see McMahan, 1997; Hurka, 1997; Lichtenberg, 1997). Increasingly, however, liberal theorists (though not only liberal theorists) have begun to look more sympathetically at nationalist aspirations (Tamir, 1993; Kymlicka, 1995a; Kymlicka, 2001: 203-89). This has led to a reconsideration of the claims of nationality in two respects. First, there is the claim for national self-determination, often associated with demands for independence or secession. And second, there is the claim for the importance of the principle of nationality for the coherence of the state and the pursuit of liberal values in particular. Both kinds of arguments in defence of nationality reveal important conflicts of value with which political theory—and liberal theory in particular continues to grapple.

National Self-Determination and Secession

National self-determination has re-emerged as an important issue in part because the 1990s saw the break-up of an Eastern European empire as well as the rise of secessionist movements around the world, from Kashmir to East Timor. But the question is an old one, not only in the politics of the twentieth century (which saw the redrawing of the maps of Africa, the Middle East and South East Asia to accommodate nationalist demands for independence), but also in liberal thought. In the nineteenth century, nationalism was allied with liberalism as the principle of nationality was invoked as a principle of freedom—and against alien rule. The liberalism of Mazzini, for example, advocated the unification of Italy as a national republic from which French, Austrian and Papal power was expelled. And John Stuart Mill saw a common nationality as a prerequisite for (liberal) representative government.

In this light, national self-determination might seem unproblematic, as an ideal both liberals and non-liberals alike might readily accept: liberals because they favour self-determination, and non-liberals because they favour national community. Yet matters are not so straightforward. In the first instance, what is always, and inescapably, controversial is the issue of who is the self that is entitled to self-determination. Even if people within a boundary are entitled to govern themselves, how is the boundary to be drawn: who is to be included and who is to be excluded (Barry, 1991; 2001: 137)?

Theorists such as Raz and Margalit (1990) look to resolve the problem by tying group membership to culture, suggesting that ‘encompassing groups’ have a number of characteristics that give them a unity which enables them to mount claims to selfhood and therefore self-determination. Central to such groups is a common culture, but no less important is the fact that people within them recognize each other as members and regard their membership as important for their own self-identification. It is also important to recognize, however, that the right of self-determination can be enjoyed only by a group that is a majority in a territory (1990: 441). What Raz and Margalit reject, as an undesirable illusion, is the individualist principle of consent: ‘It is undesirable since… the more important human groupings need to be based on shared history, and on criteria of nonvoluntaristic (or at least not wholly contractarian) membership to have the value they have’ (1990: 456).

Yet it is difficult to see how consent can fail to play a significant role in any account of self-determination if self-determination is to mean something more than the determination of the lives of some by the will of others. And many other theories of self-determination give a substantial role to consent as central to any account of political legitimacy. Among the most sustained defences of the importance of consent is that offered in the writings of Harry Beran, particularly in his defence of the right of secession as central to the legitimacy of the liberal state (Beran, 1984; 1987; but see also Green, 1988; and Simmons, 2001).

Secession has attracted considerable attention from political theorists since Beran revived the issue, not least because of its pertinence whenever the question of nationality is raised. The most influential work is Allen Buchanan’s Secession: The Morality of Political Divorce (1991), which explicitly rejected consent as a sufficient condition for the justification of secession, and suggested that, while groups could have the right to secede, this was very much a limited right. In the literature that has developed in the debates that followed Buchanan’s original contribution, three main categories of theories of secession have emerged: just-cause theories (Buchanan, 1991; 1997; Norman, 1998), choice theories (Philpott, 1995; 1998), and nationalist theories (Raz and Margalit, 1990; Nielsen, 1998). Of the three, only just-cause theories have come close to developing justifications with any prospect of being codified in a way that might influence or shape secessionist politics (Norman, 1998). However, the reality of political power and its operation in the world makes one suspect that no theory of secession is likely to provide the basis for a workable, constitutionally guaranteed, right of secession (Norman, 2003: 609).

Nationalism and Multiculturalism

Clearly national sentiment sometimes leads to calls for some consideration to be given to the case for secession. On other occasions, however, it pushes in the other direction. While the principle of nationality is sympathetic to the interests or claims of groups, and particularly to their claims to the protection of their identity, by definition it must also be wary of group claims that might undermine a national identity.

This issue arises in particular when nationalism runs up against the question of multiculturalism. To the extent that multiculturalism advocates the accommodation of a plurality of identities holding to divergent values within a polity, it is inconsistent with any form of nationalism. And yet, in another sense, multiculturalism is the theory of nationalism par excellence, at least if one takes Moore’s view that nationalism implies not separatism, but only a measure of independence. To reconcile multiculturalism, or indeed any form of pluralism, with nationalism has been an important concern for a number of theorists who wish to hold on to the principle of nationality without jettisoning cultural diversity and the toleration of difference.

Of particular significance here is the work of David Miller (1995; 2000; 2001), who considers that nations have good reason for wanting to be self-determining, but also thinks that we can recognize the claims of nationality without suppressing other sources of identity, such as ethnicity. A major reason why nationality is important for Miller is that it is a precondition of the pursuit of social justice, which cannot plausibly be pursued globally (Miller, 1999). The pursuit of social justice requires, in particular, that a measure of social solidarity is necessary if citizens are to go along with institutions which perform a redistributive function. Indeed, as others have argued, the workings of legal and political institutions may depend to a significant degree on a substantial willingness on the part of the population of a state to view themselves as members of the same group, who owe something to each other in a way which they do not owe to outsiders (Patten, 2001). For this reason, it may be necessary for the state to take an interest in the fostering of a sense of citizenship and belonging.

This, however, brings us back to the problems with which our discussion began. The desire of each to be recognized as different and distinctive gives rise to a demand for a politics of multiculturalism one that recognizes and tolerates, or even encourages and honours, diversity. Yet the politics of diversity in turn may give rise to a demand for political separation, and the emergence of communities in which diversity has no place. How the many can live as one remains a salient question in political theory.