You are on page 1of 24


Copyright 2003 SAGE Publications (London, Thousand Oaks, CA and New Delhi)
Vol 3(3): 393416 [1468-7968(200309)3:3;393416;034984]

Ways of belonging
Ethnonational minorities and models of differentiated
University of Stirling, Scotland

ABSTRACT Recent campaigns for recognition by ethnic religious and national

minorities have highlighted the limitations of the dominant unitary conception of
the nation state. While the model of differentiated citizenship advocated by Taylor
and Carens constitutes an innovative response to these challenges, the difficulties
that surround this model underline the dilemma that confronts many modern states.
Although a commitment to equal citizenship implies that the state should acknowledge the different histories, needs and goals of the various communities that constitute it, a multicultural society must at the same time develop a genuinely shared
culture. These difficult questions surrounding political stability and the equality of
citizens cannot be easily resolved. However, one potentially promising avenue is
indicated by Parekhs emphasis upon intercultural dialogue. In cases concerned with
the mediation of individual rights and certain intergroup and intragroup power
relations, such dialogue can be fostered via imaginative institutional mechanisms
such as Shachars model of transformative accommodation.

KEYWORDS cultural rights joint governance multiculturalism

The campaigns by many national minorities for greater regional autonomy
to establish distinct legal and political frameworks and calls for cultural
recognition by ethnic and religious minorities have highlighted the limitations of the dominant model of the modern nation state as an administratively centralized and politically and culturally homogeneous unit. The
emphasis that such groups place upon collective identity and their



subsequent demands for group rights question the widespread assumption

that the nation state should constitute the only significant locus of political
belonging and that citizenship should be seen primarily in terms of equal,
universal individual rights. For Charles Taylor (1993) and Joseph Carens
(2000), these challenges highlight the need for an alternative conception of
the modern state that can accommodate the aspirations of diverse cultural,
ethnic and national groups. In place of the preoccupation with homogeneity that characterizes dominant models of the nation state including the
liberal ideal of civic nationalism Taylor and Carens propose distinctive
models of liberal democratic citizenship that seek to recognize multiple
forms of belonging to the political community and overlapping identities
and citizenships. In developing these models, Taylor and Carens draw upon
recent Canadian struggles for recognition that have been characterized by
a complex and at times crosscutting set of demands, including campaigns
by the Qubcois and indigenous peoples. What characterizes Taylors and
Carenss models of differentiated citizenship is a commitment to what
Taylor terms deep diversity; an arrangement that would grant collective
group rights to many minorities, including rights to self-government for
territorially-based minorities. On such a model, individual citizens are
incorporated into the state not universally, but consociationally; that is, as
members of their cultural or national group. Thus, members of different
minority cultural and national groups belong to the state in different ways.
Hence membership of the relevant cultural or national group becomes the
primary locus of political identity.
While attempts to develop models of the state that recognize the
multiple dimensions of citizenship are to be welcomed, these conceptions
of citizenship remain problematic. Not only are the liberal credentials of
this approach open to question, but the emphasis that Taylor and Carens
place upon the recognition of existing identities does not provide an
adequate model for the struggles for recognition by social groups such as
women. Finally, if national or cultural group membership is to be the
primary focus of political identity, it may be difficult to generate the genuine
sense of cohesion necessary for political stability. These difficulties suggest
that, if models of differentiated citizenship are to succeed, they must answer
complex questions surrounding political stability and the equality of
citizens. Here one potentially promising avenue is indicated by Parekhs
(2000) emphasis upon open-minded, morally serious intercultural dialogue
that probes the nature of existing cultural practices and seeks to promote
mutual adaptation and compromise. In cases concerned with the mediation
of individual rights and certain intergroup and intragroup power relations,
such dialogue can be fostered via imaginative institutional mechanisms such
as transformative accommodation (Shachar, 2001), which combines the
recognition of group rights with a typically liberal concern for individual
wellbeing, freedom and equality.




The dominant model of the nation state as an administratively centralized
and politically and culturally homogeneous unit has its roots in the late 15th
and early 16th centuries and reflects an historically and culturally distinctive vision of the political order that has a strong liberal orientation
(Parekh, 2000). Whereas premodern polities tended to be composed of
communities like castes, clans, tribes and ethnic groups, the modern state
is defined as an association of individuals. Individuals are not linked to the
state through membership of a particular cultural community, but stand as
singular citizens in the same direct relationship to the state. Thus, citizens
relate to the state in an identical manner and enjoy equal status. This
preoccupation with the relationship between state and individual citizen is
probably most apparent in the liberal contract tradition. For contract
theorists, the state is the product of a process of deliberation among undifferentiated individuals, and no account is taken of the customs, traditions
and institutions that may constitute a particular people prior to the social
contract. On this model, to be a citizen is to transcend ones ethnic,
religious and other particularities, and to think and act as a member of a
political community (Parekh, 2000: 181). The aim is to constitute one
uniform political association. Hence, even liberal states hostile to the idea
of ethnic nationalism have promoted forms of civic nationalism that stress
the importance of unity of government and seek to engender loyalty to
the key values, principles and procedures enshrined in the constitution
(Canovan, 1996; Gilbert, 1998). For civic nationalists, these constitutional
values and principles override the more particular attachments of citizens.1
This drive for homogeneity is reinforced through a uniform legal system. A
single set of constitutional principles guarantees identical rights and obligations to all citizens and, in the case of federal states, ensures that the
various component units are granted roughly the same rights and powers.
In contrast, the constitutions of premodern polities were multiform, an
assemblage of overlapping legal and political jurisdictions based upon a
variety of local customs (Tully, 1995). Finally, whereas earlier polities recognized multiple and often crosscutting ethnic, religious and territorial identities, the modern nation state attaches unprecedented importance to
territory. Thus, the modern state is typically perceived as a self-contained
unit within which the state acts as the sole collective spokesman (Parekh,
2000: 181). Indeed, since the 19th century, citizenship has come to be
identified with membership of a modern nation state (Bellamy, 2000).
Yet this vision of the nation state has important limitations. Whereas the
dominant model of the nation state seeks to build a culturally and politically homogeneous unit, most modern states are multi-ethnic and multinational societies whose constituent communities entertain different views




of its nature, powers and goals (Parekh, 2000: 184). This diversity has given
rise to ever more vocal demands on the part of an increasing number of
minority groups for the recognition and support of their particular identity
and culture. Thus, numerous national minorities have sought to protect
their distinct identity and culture by gaining legal protection and support
for their national language. For instance, the Catalans in Spain, the
Qubcois in Canada and, to some degree, the Welsh in Britain have
attempted to enshrine in law the use of their national language in schools,
employment, the media and public life in general. Such campaigns have
frequently been associated with wider demands for greater regional
autonomy and, in the case of Qubec, secession. Another distinctive set of
demands stems from the aspirations of indigenous people in New Zealand,
Australia, Canada and the US to gain greater control over the resources
and policies that affect their way of life. Here, the focus has tended to be
on land rights,2 yet a different set of concerns has been voiced by religious
minorities, who have expressed fears about the impact of educational
provisions upon their ability to raise their children in accordance with their
own religious beliefs.3
These demands for recognition highlight the political significance of
those interests individuals possess by virtue of their membership of a
particular religious, ethnic or national group. Rather than transcend their
particular identities, those campaigning for recognition have stressed the
impact of ethnicity, religion and nationality upon the identity of individual
group members and have pointed to the links between the wellbeing of
individual members and the wellbeing of the group as a whole. Consequently, many groups campaigning for recognition seek group-specific
cultural rights designed to protect and perpetuate their culture and identity.
Such group-differentiated rights can take the form of either personal or
corporate cultural rights. While personal cultural rights are exercised by
group members individually, corporate cultural rights are exercised by the
group collectively. For example, the campaign by Muslim girls in France for
the right to wear headscarves in the classroom constitutes a demand for
personal cultural rights. In contrast, campaigns by national minorities for
greater self-government and demands by indigenous people for special
hunting, fishing and land rights are calls for corporate cultural rights.
Although, in practice, many nation states already grant at least some
cultural rights to religious ethnic and national minorities,4 on the dominant
model of the modern nation state such demands for group-differentiated
rights are problematic on at least two counts: first, both personal and
corporate cultural rights question the assumption that citizenship should be
seen primarily in terms of identical, universal individual rights and hence
challenge the conventional meaning of equal citizenship; and, second,
corporate cultural rights contest the claim that the state should constitute



the only significant locus of political belonging and thus raise important
questions about the manner in which citizens are to be incorporated into
the state.


In the light of these difficulties, many theorists have expressed serious
doubts about the adequacy of the dominant unitary model of the nation
state. Drawing upon recent Canadian struggles for recognition, Taylor and
Carens attempt to address these concerns by developing an alternative
model of liberal democratic citizenship that recognizes multiple forms of
belonging to the political community and accommodates overlapping identities and citizenships. Thus, both writers reject the difference-blind conception of equality that characterizes the unitary model in favour of the idea
of equitable treatment, which recognizes that the standard schedule of
liberal rights might apply differently in different cultural contexts. Furthermore, both want to enable citizens to be incorporated into the state not just
universally, but also consociationally; that is, as members of their cultural
or national group.
The emphasis that Taylor and Carens place upon the need for equitable
treatment reflects widespread concerns regarding difference-blind conceptions of equality. Although every liberal democratic regime must uphold
certain principles, such as freedom of speech, freedom of religion, majority
rule, and so on, these principles can be implemented in a variety of ways
and every actual liberal regime remains thickly embedded in particular
contexts. After all, every state must make decisions about which official
language to use, which public holidays to observe and how to draw internal
boundaries. Since these decisions typically reflect the norms and sensitivities of the majority culture, they tend to place members of ethnic,
religious and cultural minorities at a disadvantage. Not only may minorities
have to spend resources on cultural goods that the majority obtains for free
(Kymlicka, 1995), but the traits, characteristics, behaviour and options
associated with the majority culture are allowed to shape the public sphere
and are thus perceived as normal. In contrast, those of the minority are
singled out as different and remain socially invisible, erased or despised
(Galeotti, 2002). This may lead to a lack of confidence and self-esteem,
which may undermine the capacity of minorities to take advantage of the
opportunities society has to offer and may make it difficult for members of
minority groups to become functioning social agents and full citizens. In the
light of these difficulties, a number of liberal writers including Habermas
(1993), Kymlicka (1995) and Raz (1994) have argued that, if members of




minorities are not to be discriminated against, they must be granted groupspecific cultural rights. The aim of such rights is to ensure that the autonomy
of all citizens is protected equally.5
What distinguishes the models of differentiated citizenship advocated by
Taylor and Carens is that, in addition to personal cultural rights designed
to protect the members of minority groups from discrimination, these
writers advocate corporate cultural rights that seek to guarantee the longterm survival of a particular culture and identity. Both maintain that, since
individual identity is constructed within a particular cultural context, a
secure identity can only be established if the survival of the cultural
community in which it was formed is ensured. A failure to safeguard the
particular identity of existing cultural communities constitutes a threat to
the identity of their members and, hence, is incompatible with the typically
liberal commitment to equal respect. Hence, Taylor concludes that a liberal
society can legitimately adopt measures designed to protect and promote
the culture and way of life of a particular community provided that such
measures respect the fundamental liberal rights of all citizens and grant
equal citizenship to all members of society. Here Taylor draws a careful
distinction between fundamental rights, such as rights to life, liberty, due
process, freedom of religion, free speech, and so on, and the wide range of
possible privileges and immunities that may be granted by the state. While
the liberal state must protect the fundamental rights of all citizens equally,
privileges and immunities can be revoked or restricted for reasons of public
policy (Taylor, 1992: 59). Indeed, the liberal state should at times weigh
the importance of certain forms of universal treatment against the importance of cultural survival, and opt sometimes in favour of the latter (1992:
61). Thus, for instance, on this account, the government of Qubec is
entitled to safeguard the continuation of French Canadian language and
culture. Hence, laws stipulating that commercial signs be in French and
requiring that companies with more than 50 employees use French, as well
as the compulsory French language education of children of Frenchspeaking parents and immigrants, do not, according to Taylor, violate the
fundamental rights of either immigrants or the English-speaking community in Qubec; they merely restrict certain privileges and immunities.
Thus, a government may effectively safeguard the survival of a nations
culture without violating the rights of those citizens who do not subscribe
to this particular vision of the good. This view is echoed by Carens. While
the state cannot legitimately restrict the right of citizens to communicate in
their chosen language, it does not have to be neutral as far as the positive
support of languages is concerned. As long as the Qubec state permits the
English-speaking community and immigrants to communicate in their
chosen language, it can legitimately insist upon the use of French in public
life. Not only is French central to the culture and way of life of the majority
of Qubcois, but French-speaking Canadians constitute a potentially



vulnerable minority in overwhelmingly English-speaking North America;

the Qubcois are keenly aware of the rapid and substantial decline of the
use of French among francophone emigrants from Qubec. Prior to the
introduction of the language laws, this vulnerability was reinforced by the
tendency of immigrants to Qubec to learn English rather than French.
However, while Carens believes that the Qubcois are entitled to take
steps to secure the long-term survival of their distinctive cultural community, he stresses that Qubcois culture should be defined in an open
way, with an almost exclusive emphasis on the French language as a shared
cultural commitment and without privileging the culture of the descendants
of the settlers of Qubec. This ensures that, while immigrants will have to
adopt the French language, they can nonetheless feel a genuine sense of
belonging to the community.
As Taylors and Carenss discussion of Qubec indicates, these writers
endorse a model of citizenship that grants cultural and national minorities
the right to develop their own distinct legal and political frameworks,
designed to protect and perpetuate their cultural identity. However, Taylor
and Carens do not merely challenge the preoccupation with political and
legal homogeneity that characterizes the dominant model of the nation
state, they also raise important questions about the manner in which
citizens are incorporated into the state and identify with it. While the
unitary model expects citizens primary attachment to be to the state,
citizens frequently identify more strongly with the national and cultural
communities that make up the state than with the state itself. Thus, francophones living in Qubec identify more strongly as Qubcois than as
Canadian (Carens, 2000). Similarly, many Scots in Britain see themselves
as Scottish first and British second, and something analogous holds for
many aboriginal communities (Taylor, 1993). According to Taylor and
Carens, liberal democratic states should not only seek to accommodate, but
should embrace such deep diversity as a positive basis for unity and a
common identity (Carens, 2000: 172). First level diversity acknowledges
that there are differences in culture, outlook and background, but insists
that all citizens stand in an identical relationship to the state. Such first level
diversity is typified by approaches such as the Canadian mosaic model of
pluralism, which, rather than seeking to assimilate different ethnic groups,
allows for a degree of diversity in terms of language and culture.6 In
contrast, deep or second level diversity allows members of different
cultural and national groups to belong to the state in different ways. Thus,
reflecting upon the diverse loyalties and senses of identification that characterize Canadian politics, Taylor (1993: 183) concludes:
To build a country for everyone, Canada would have to allow for second-level
or deep diversity, in which a plurality of ways of belonging would also be
acknowledged and accepted. Someone of, say, Italian extraction in Toronto or
Ukrainian extraction in Edmonton might feel Canadian as a bearer of




individual rights in a multicultural mosaic. His or her belonging would not pass
through some other community, although the ethnic identity might be
important to him or her in various ways. But this person might nevertheless
accept that a Qubcois or a Cree or a Dn might belong in a very different
way, that these persons were Canadian through being members of their national
communities. Reciprocally, the Qubcois, Cree or Dn would accept the
perfect legitimacy of the mosaic identity.

Just like asymmetrical federalism, deep diversity requires that the institutional arrangements, including the legal rights and duties of citizens,
should reflect the psychological realities of differentiated political identities (Carens, 2000: 170; see also Fossum, 2001). Thus, the state should play
a stronger role in parts of the federation where the primary attachment is
to the state and a lesser role in areas where the primary attachment is to a
local, national or cultural identity.7 However, deep diversity goes further
than asymmetrical federalism in the sense that it affirms, and does not
merely yield to, the multiplicity and variety of political identities (Carens,
2000: 172). Deep diversity therefore expresses a particular political culture
rather than just a set of political arrangements.
This model of deep diversity has particularly far-reaching implications
for liberal democracies as far as the accommodation of non-liberal minorities, such as aboriginal peoples, is concerned. These are illustrated by
Carenss discussion of the demands by aboriginal peoples in Canada for
greater control over their lives. According to Carens, if a case can be made
for a right to self-government for the Qubcois, then the same must apply
to aboriginal peoples; their political identities are clearly much more
sharply distinct from Canadian citizenship than those of any other people
who live in Canada (2000: 172). For Carens, this degree of difference has
far-reaching political consequences. Whatever the differences between
French-speaking and English-speaking Canadians, both cultures are
broadly liberal and recognize the standard language of liberal rights and
freedoms. However, where difference runs deeper, as in the case of aboriginal peoples, the language of liberal rights and freedoms may be perceived
by some minority communities as alien. Thus, for instance, while some
Canadian aboriginal peoples, such as the Mtis, have embraced the
Canadian Charter of Rights and Freedoms introduced in 1982, other aboriginal groups have argued that the Charter should not apply to them.
Opponents of the Charter maintain that the typically liberal language of
individual rights and freedoms is alien to their culture and tradition, which
emphasizes responsibilities and the wellbeing of the community. While
Carens does not wish to simply endorse the view that the Charter should
not apply to aboriginal governments, he does recognize such a view as
potentially defensible and plausible. For Carens, principles of justice, like
all other principles, require mediation, instantiation, embeddedness in
some concrete social context. Hence, in the face of deep cultural



differences within a state, it may be appropriate and even necessary to try

to construct alternative forms of mediation, forms more concurrent with
non-dominant cultures (2000: 190). Carens is therefore, in principle,
sympathetic to the idea that aboriginal governments should be free to draw
up their own Charter or that the Charter should only apply to aboriginal
governments as long as those governments have the same override powers
as the Canadian federal and provincial governments (2000: 192). In
Carenss opinion, the latter is particularly difficult to object to. After all,
fundamental rights and freedoms can be protected in a variety of ways and
what will work best in a particular political context depends partly on the
character of the culture of the people whose rights and freedoms are being
protected (2000: 192).


The demands for recognition and the multiple and crosscutting loyalties
identified by Taylor and Carens are by no means unique to Canada. As
noted earlier, many liberal democracies have recently been confronted not
only with demands for cultural rights on the part of ethnic and religious
minorities, but have also witnessed challenges to the existing constitutional
arrangements that govern the relationship between the state and its
national and cultural communities. The model of differentiated citizenship
advocated by Taylor and Carens clearly constitutes an innovative response
to these challenges and avoids the difficulties associated with formal
equality and a unitary model of citizenship. Yet, despite these strengths, this
model of differentiated citizenship remains problematic on at least three
counts: first, collective rights that seek to guarantee the long-term survival
of a particular language or culture may ultimately remain at odds with the
liberal ideals of equality, freedom and self-determination; second, the
emphasis that these writers place upon the recognition of existing identities
does not provide an adequate model for the struggles for recognition by
social groups such as women; and, third, if national or cultural group
membership is to be the primary focus of political identity, it is difficult to
see what common good the state will secure.
Although Taylor and Carens stress that their model of differentiated
citizenship is compatible with liberalism, the liberal credentials of this
approach are open to question. In the eyes of liberal critics, an emphasis
upon ethnic identity and corporate cultural rights designed to secure
cultural survival is liable to generate a separatist mentality that is inconsistent with the ideals of freedom, equality and the ongoing, co-operative
and experimental search for truth and well-being (Rockerfeller, 1992: 92).
Yet, for many liberals, these ideals are definitive of a liberal society. Carens




acknowledges this point when he observes that the unitary model of citizenship does offer an ideal of equality that provides grounds for criticising
policies and practices that exclude or marginalise distinct groups of citizens
(2000: 178). Not only is the closure with respect to outsiders, which tends
to accompany the consolidation of cultural identity, liable to impair the
sense of solidarity that underpins the liberal commitment to equal dignity
and universal rights, it may also threaten to undermine the freedom and
autonomy of group members. As Habermas (1993: 131) notes, policies such
as the Qubec language laws, which seek to guarantee the long-term
survival of a particular cultural community, extend well beyond what could
be justified as just providing . . . a facility to already existing people.
Although the primary justification for these laws is to ensure equal opportunity for francophones against the economic and political pressure of the
anglophone majority in Canada, these laws also place restrictions upon
members of the francophone community to use English (Kymlicka, 1995:
205). For example, laws that stipulate that commercial signs must be in
French and specify that children of French-speaking parents and immigrants must attend French language schools restrict the availability of
English language-based education and limit the use of English in certain
spheres of life.8 Clearly, all states require a common language or two or
three official languages, and public schooling is typically only available in
the dominant language(s). Thus, while a liberal state cannot legitimately
restrict the right of immigrants to communicate in their first language, it can
expect immigrants and their children to learn the dominant language so as
to enable them to participate in economic, social, cultural and political life.
However, given that both English and French are official languages in
Canada and that Qubec is already obligated to provide publicly-funded
English language-based schools for the children of anglophone parents, a
typically liberal regard for individual freedom and autonomy arguably
entails that where the state is already committed to funding English
language-based schools, the children of French-speaking parents and immigrants should be free to choose in which official language they wish to be
educated.9 After all, the children of anglophone parents living in Qubec
are free to attend French-speaking schools. While a commitment to individual freedom and autonomy entails that the state should protect the
francophone community from the external economic and political pressure
of the majority anglophone community by, for example, adopting measures
that ensure francophones can speak French in the workplace, such a
commitment is not readily compatible with measures that constrain the
ability of francophones to freely choose to speak English. Yet, if cultural
survival is to be guaranteed, such restrictions are vital. To ensure long-term
survival, a culture must systematically socialize future citizens into the
language, values and norms definitive of that culture. This is most easily
achieved if the choices of future citizens are restricted. However, such



attempts to fix the identity of future generations are not readily compatible
with the ideals of individual freedom and self-determination upon which
liberalism is premised. As Ripstein (1995: 333) observes, while keeping
traditions alive is one of the most abiding of human aspirations, it is not
clear that this is the sort of aspiration that liberals should be prepared to
use the state to protect.
Here, liberals are liable to be particularly fearful of the desire of nonliberal societies to preserve their traditional character. For example, while
the long-term survival of certain religiously-based cultures may well require
state intervention and protection to ensure that children continue to share
the faith of their parents, liberals would surely not wish to use the powers
of the state to enforce religious faith.10 Taylor seeks to address such fears
by distinguishing between fundamental rights and privileges and immunities. However, fundamental liberal rights will always require interpretation. In a truly multicultural society, should the right to freedom of
religion be interpreted primarily in terms of the right of individuals to
autonomously choose faith commitments or should it merely imply tolerance of the faith of others? Taylors distinction cannot resolve these questions.
Similar difficulties surround Carenss insistence on the need to define
culture in an open-ended manner so as to guarantee equal citizenship to
immigrants and ensure that they can feel a genuine sense of belonging to
the community. Thus, in the case of Qubec, Carens maintains that while
the provincial government can legitimately insist upon the adoption of the
French language as a shared cultural commitment, it should not privilege
the culture of the descendants of the French settlers of Qubec. However,
in practice, it may be difficult to distinguish between preserving the French
language and upholding Qubcois culture. After all, groups frequently
seek to maintain their traditional language precisely because it plays an
important role in perpetuating the wider culture of the group.11 Although,
from a liberal perspective, immigrants can legitimately be asked to accept
the constitutional principles anchored in the wider political culture, the
liberal principle of impartiality implies that they cannot be expected to take
on board the particular culture that prevails in a given country; on the
contrary, there has to be room for mutual adaptation. Thus, Habermas
(1995), for example, distinguishes between the civic sense of a nation based
upon the common civic identity of a group of citizens and the particular
ethical discourses of subcultures. Although the ethical discourses of specific
subcultures provide the basis for and feed into the collective identity of a
particular nation, for Habermas the wider political culture must be decoupled from the majority culture. While all liberal states inevitably reflect
the concerns of the social and cultural groups within their boundaries, the
arrival of new ethnic groups should, in the long run, lead to changes in the
political identity of the state. This gives rise to an uncomfortable dilemma.




If, in the case of Qubec, the common cultural commitment could be

successfully limited to the French language, it would be difficult to see what
would remain of the Qubec demand for distinctiveness and why Qubec
should be entitled to restrict the freedom of citizens. If, on the other hand,
a strong case can be made for a genuinely distinct French Canadian culture
tied to a specific history and particular cultural practices, it is difficult to see
how, on liberal grounds, this identity can be tied in exclusively with the
Qubec state.
A preoccupation with cultural identity may not only restrict the freedom
of future generations and immigrants, but may also prove oppressive vis-avis existing group members. Even if individual rights are respected, a
politics of cultural survival may lead to pressure being placed upon the
individuals to define themselves primarily in terms of their cultural
membership rather than a whole host of alternative criteria that may shape
their identity. After all, our cultural identity constitutes only one influence
upon our conception of the good life. In modern societies, we all belong to
many distinct and sometimes opposing groups. Thus, our occupation, social
status and choice of neighbourhood all contribute to our sense of self and
conception of the good life. In the face of such a range of possible sources
of identity, a preoccupation with cultural membership may lead to an undue
emphasis upon this aspect of the self.12 Such pressures upon individual
members of a community become particularly troubling once the contested
nature of cultural identity is recognized. While Taylor and Carens are sensitive to the diversity of differently constituted political societies, their vision
of particular cultures remains quite homogeneous. For Taylor, the members
of a cultural community share meanings and values and are collectively
committed to the promotion of their particular conception of the good.
Such an emphasis upon shared meanings and goals, however, appears to
underestimate the extent to which cultural identity will always remain
internally contested. As Houston (1998) observes, part of the difficulty here
stems from the manner in which Taylor tends to conflate the construction
of individual identity and cultural identity. While the construction of individual identity presumes a constituting self, which generates a sense of unity
by organizing diverse influences and roles into one narrative, cultural
identity lacks such a unifying centre. In the construction of cultural identity,
there is no one nodal point. Hence, there simply is no neutral referee to
choose a coherent logic between multiple competing narratives (Houston,
1998: 238). The construction of cultural identity is therefore an inherently
political process characterized by powerful intracultural struggles to determine whose description is to count as authoritative. What is to be regarded
as shared inheritance is therefore always liable to remain to some degree
These worries regarding the contested nature of identity also highlight a
second problem inherent in the idea of differentiated citizenship, as



advocated by Taylor and Carens. The emphasis that Taylor and Carens
place upon the recognition of particular distinctive cultural identities is
clearly well placed to capture the character of modern nationalist politics.
However, critics have questioned whether such an appeal to the recognition
of existing identities can provide an adequate model for the struggles for
recognition of social groups such as women. As critics such as Nicholson
(1996) and Wolff (1992) have been quick to point out, in the case of women,
it is precisely the recognition of their particular identity that is problematic.
Historically, women have only too readily been recognized and defined in
terms of their gender. Consequently, in the case of women, the demands for
recognition do not primarily reflect a desire to have a particular way of life
recognized, but stem from an analysis of the ways in which oppression was
manifesting itself in terms of forms of descriptions and evaluations (Nicholson, 1996: 6). These concerns about the nature of recognition are particularly troubling with regard to the recognition of cultural and religious groups
who do not endorse the liberal commitment to gender equality. According
to Carens, a commitment to a genuinely differentiated citizenship requires
that liberal societies tolerate many of the gender differences inherent in the
internal cultures of religious, ethnic or national groups. While Taylor and
Carens link the recognition of cultural identities to the liberal ideal of equal
respect, the comparatively low esteem in which women are held in such
cultures and the restrictions that cultural expectations place upon womens
choices and the kind of lives they should lead may well undermine womens
self respect.13 While Carens believes that the commitment to equality
inherent in a liberal public culture subverts patriarchal values, the worry
remains that traditional communities may be able to resist such pressures
only too well. Thus, cultural rights that seek to support and sustain existing
cultural identities and practices may ultimately threaten rather than
promote the kind of recognition that Wolff and Nicholson regard as central
to womens identity politics. Some of the tensions here are illustrated by the
debate among aboriginal women in Canada regarding the Canadian Charter
of Rights and Freedoms.14 Whereas some Canadian aboriginal women have
sought to have their voices heard within aboriginal governments, others
have argued that the Canadian Charter should be applied directly to aboriginal governments so as to protect women from a traditional male-dominated leadership insensitive to their rights and interests. While traditional
communities are clearly capable of adaptation and change, the emphasis
that Taylor and Carens place upon the recognition of existing identities fails
to promote intragroup dialogue regarding current norms and values. From
a liberal perspective, the worry remains that, in the absence of mechanisms
that allow traditionally marginalized groups such as women to challenge
existing norms, attempts to mediate typically liberal rights and freedoms in
the light of the values and practices of traditional communities will undermine key liberal commitments such as gender equality.




Not only does the model of recognition advocated by Taylor and Carens
fail to capture the aspirations of social groups such as women, but it is also
not ideally placed to reflect some of the concerns of indigenous peoples,
many of whom do not live in territorially-defined communities. While, for
Taylor, modern nationalist politics, with its emphasis upon the recognition
of ones particular distinct cultural identity, provides a model for identity
politics in general, this preoccupation with the demands of national minorities gives rise to a conception of recognition that emphasizes separatism and
territorially-based self-government. However, indigenous peoples do not
tend to see the right to self-government in the way that national minorities
typically do; namely, as the right to secede. Thus, for instance, in the context
of discussions surrounding the UN Draft Declaration on the Rights of
Indigenous Peoples, indigenous delegates have stressed that selfdetermination does not need to imply separation or isolation, but entails
control over ones own destiny and living in accordance with ones own
values (Feldman, 2001). Rather than separate entitlements, this may entail
more effective participation in society as a whole.
In addition to these concerns regarding the liberal credentials of the
model of citizenship advanced by Taylor and Carens, it is questionable
whether a differentiated citizenship, which allows citizens to be incorporated consociationally, will be able to generate bonds strong enough to
sustain the state. Both Taylor and Carens maintain that, in the face of
genuine cultural diversity, a differentiated citizenship is more likely to
promote civic integration than the conventional unitary model. Thus, for
Taylor (1993: 183), deep diversity is the only formula on which a united
federal Canada can be rebuilt. Yet, as Horton (1998) notes, if the primary
identity of citizens is as members of their national community, it is difficult
to see what common good the federation will secure. Although an alliance
can promote law and order, security and common provisions and may well
be of instrumental benefit to the various national communities that constitute it, Taylor himself recognizes that securing common goods in this sense
is not sufficient to make them a single people (Horton, 1998: 170). While,
in good times, it may not matter if the allegiance to the state is based on
purely instrumental considerations, in times of crisis . . . citizens of this sort
tend to prefer exit over loyalty or voice (Canovan, 1996: 86). Even though
a modern state can afford a purely instrumental allegiance on the part of
some citizens, it may struggle to sustain itself should such an attitude
become widespread. Consumer loyalty does not provide a strong enough
basis for political power. The preoccupation with Qubec leads Taylor and
Carens to underestimate the difficulties in maintaining a genuinely differentiated citizenship. While English-speaking Canada and Qubec differ
with regard to certain collective goals, both share a broad range of cultural
commitments, including important liberal values such as the recognition of
fundamental individual rights and freedoms.15 Yet, in deeply diverse



societies where such a sense of a genuine common purpose over and above
the purely instrumental is absent, a federation that recognizes deep diversity will be difficult to maintain in the long run.


While the conception of differentiated citizenship advocated by Taylor and
Carens recognizes multiple forms of belonging and can accommodate overlapping identities and citizenships, the difficulties that surround this model
highlight the dilemma that confronts many modern states. The modern
state can only secure legitimacy if all citizens are shown equal respect and
are able to participate in the political process on equal terms. If this is to
be achieved, the different histories, needs and goals of the various
communities that constitute the state must be acknowledged. Yet, at the
same time, a multicultural society must develop a genuinely shared culture.
If such a culture is to provide a basis for civic integration and mutuality, it
must be based upon more than purely instrumental considerations. While
the collective rights advanced by Taylor and Carens are well placed to
accommodate the demands for recognition by many national and cultural
minorities, such rights risk reinforcing differences between communities
and thus may undermine attempts to develop a shared culture and an overarching national identity. Hence, attempts to secure the long-term survival
of a language or culture via collective rights not only remain problematic
from a liberal perspective, but also threaten to give rise to fragmentation
and political instability. These worries are reinforced by the multifaceted
nature of demands for recognition. Not all such demands can be easily
reconciled. The tensions between the aspirations of women for greater
gender equality and the desires of traditional communities to perpetuate
their way of life pose a complex dilemma to those seeking a differentiated
These difficult questions surrounding political stability and the equality
of citizens cannot be easily resolved. However, one potentially promising
avenue is indicated by Parekhs (2000) emphasis upon intercultural
dialogue. While, in a multicultural society, the recognition of diversity may
entail granting some minorities collective cultural rights and may require
institutional arrangements that are sensitive to the differentiated political
identities that constitute the state, such a differentiated citizenship must be
accompanied by an open-minded, morally serious dialogue between the
majority and the minority. Such dialogue must search for common ground
and aim at mutual adaptation. In this search, neither majority nor minority
can expect all their existing cultural practices to remain unchanged. Where




the current practices of the majority and minority conflict, intercultural

dialogue must probe the nature and importance of existing cultural practices and must encourage reflection and debate not only between communities, but also within the various communities themselves. Thus,
communities must consider whether contentious cultural practices are
essential to their way of life or whether they could safely be abandoned.
For example, on Parekhs account, many of the most contentious cultural
practices regarding the treatment of women, such as female circumcision,
are arguably not essential to the way of life and values of the cultures that
practice them and therefore could be abandoned without threatening the
survival of these cultures. To promote such serious intercultural dialogue,
states will have to develop institutional mechanisms that facilitate exchange
between minority and majority and actively encourage reflection upon
existing practices. Although the model of deep diversity advocated by
Taylor and Carens recognizes cultural rights and divides jurisdiction
between several actors, it is not particularly well placed to promote
dialogue and mutual adaptation. Since deep diversity divides jurisdiction
into separate, but quite self-contained units, it allows groups to accumulate
authority over all aspects of members behaviour relating to its area of jurisdiction. This may well encourage groups to exacerbate their differences
rather than seek mutual adaptation.
The design of institutions that can promote intercultural dialogue clearly
poses a complex problem and, given the diversity of demands for recognition, will require a range of institutional responses.16 However, at least
some questions regarding the mediation of individual rights and certain
intergroup and intragroup power relations can potentially be addressed
through innovative models of joint governance such as transformative
accommodation, as developed by Ayelet Shachar (2001). As Shachar
notes, many contested social areas, like family law or criminal justice, are
internally divisible into distinct, yet interdependent, submatters or functions. On the model of transformative accommodation, both the minority
group and the state are given authority over part of the submatter, but
neither is allowed to control all aspects of the contested social area. For
example, in the case of marriage, the minority group may manage the
demarcatory function, which controls changes in marital status and entitlement to community membership, whereas the state regulates the distributive function, which includes the rights and obligations of marriage partners
and economic and custodial matters. Similarly, in criminal law, the state
may be responsible for establishing whether a person is guilty whereas the
minority community determines sentencing via mechanisms such as
sentencing or healing circles.17 To take account of the power asymmetry
between the state and the minority group, the presumptions in the negotiations regarding the initial allocation of areas of authority should be in
favour of the group. Thus, the group may, for example, be given the



opportunity of setting the agenda for negotiations or it may be given

priority in the allocation of the submatter it regards as the most crucial.18
Since a legal dispute in a particular social area can only be fully resolved if
these submatters are addressed together, neither party has complete
control over its members. Therefore, in contrast to the model of deep diversity advocated by Taylor and Carens, transformative accommodation
allows standards from a range of sources to govern the matters arising in
a social area simultaneously (Shachar, 2001: 1256). Such an arrangement
promotes dialogue and cooperation between the state and the minority
group and encourages both parties to reflect upon their values and to be
responsive to the needs of their members. Furthermore, it relieves the
pressure upon individuals to define themselves primarily in terms of their
cultural identity.
These features are reinforced via the final characteristic of transformative accommodation: the establishment of clearly delineated choice options.
At predefined reversal points, individuals are to be given the choice
whether to remain within the jurisdictional authority of the original powerholder. On this model, individuals are justified in opting out if the current
powerholder in a particular submatter systematically fails to address their
concerns. Such reversal options allow individuals to bring pressure to bear
on the groups that represent them and thus create a strong incentive for
powerholders to address in-group problems. For instance, if the rules of
divorce systematically discriminate against women by granting husbands
the right to unilaterally divorce their wives or by insisting that, in order to
obtain a divorce, a wife must gain the consent of her husband, women may
choose to opt out of the groups jurisdiction in this particular matter.19
Similarly, members of the community who are discriminated against in the
allocation of benefits administered by the community, such as health care
or education, may appeal to the state for redress. Since such an opt out or
partial exit only applies to a particular issue, vulnerable group members do
not have to give up their overall group membership in order to seek redress.
At the same time, groups have a strong incentive not to risk alienating their
members. Consequently, groups may decide to reinterpret the existing rule
to accommodate the concerns of disaffected members. Transformative
accommodation therefore offers a more flexible approach to the allocation
of authority than Taylors and Carenss model of deep diversity. Such a
flexible approach helps to ensure that the public recognition of differences
does not lead to the ossification of cultures, discourages elites within groups
from insisting upon group-differentiated rights, when these are no longer
desired by ordinary group members, and recognizes the potential tension
between the aspirations of women for greater gender equality and the
desires of traditional communities to perpetuate their way of life. Transformative accommodation therefore seeks to transform both intergroup
relations between minority and majority and intragroup relations between




powerful and traditionally marginalized sections of the group by promoting a sense of cooperation and dialogue.
The manner in which transformative accommodation combines group
rights and the protection of individual liberties makes a potentially promising model for exploring some of the difficult questions surrounding
political stability and the equality of citizens in culturally diverse societies.
However, while this model is well placed to accommodate some of the
concerns expressed by potentially vulnerable group members such as
women and certain demands by indigenous peoples for greater control
over the resources and policies that affect their way of life, it is not suitable
in all areas. For example, it is not well suited to disputes concerning
children, since, as minors, children are not able to exercise the choice
option. Given that many disputes in culturally diverse societies centre
upon the upbringing, treatment and education of children, this is potentially an important limitation. More significantly still, transformative
accommodation is not well suited to territorially-defined demands for
recognition such as the claims of national minorities like the Qubcois.
However, given the complexity of the demands associated with claims for
recognition, no one institutional mechanism is likely to provide a suitable
model for all cases.

The model of differentiated citizenship advocated by Taylor and Carens
represents an innovative response to the challenges posed by calls for
cultural recognition on the part of ethnic and religious minorities and
campaigns by national minorities for greater regional autonomy. However,
although this approach addresses many of the limitations of the dominant
unitary model of the nation state, the difficulties inherent in the idea of
deep diversity suggest that, if models of differentiated citizenship are to
succeed, they must address complex questions surrounding political
stability and the equality of citizens. While these questions can potentially
be addressed through open-minded, morally serious intercultural dialogue,
such dialogue requires institutional mechanisms that facilitate exchanges
between minority and majority and encourage reflection upon existing
practices. Some questions relating to the mediation of individual rights and
certain intergroup and intragroup power relations can be addressed via
innovative models of joint governance such as transformative accommodation, yet even such imaginative responses cannot provide a solution in all
contested cases. In particular, such models cannot address territoriallybased claims for recognition such as the demands of national minorities.
Many modern states will therefore have to continue to face the dilemma of



how to combine the search for a shared culture with a recognition of the
different histories, goals and needs of the communities that constitute them.

1 Typical examples of civic nationalism include the Verfassungspatriotismus,
which was adopted in Germany to refer to a loyalty to the liberal democratic
constitution of the post-war Federal Republic (Canovan, 1996: 87), and the
republican patriotism associated with American nation-building. Such civic
nationalism may in practice prove more homogenizing than ethnic variants.
While the belief that members of the nation naturally belong together may give
rise to some quite relaxed views regarding allegiance, civic nationalists will be
anxious to educate citizens into respect for the constitution and to instill loyalty
and respect for the key values and principles enshrined in the constitution.
2 For example, Maori in New Zealand have invoked the Treaty of Waitangi to
establish claims to ancestral lands while, in Australia, the 1992 High Court
judgement in the Mabo case recognized the principle of indigenous land rights.
Similar legal challenges have been mounted in Canada (see the Delgamukw
case in British Columbia). For a discussion of these cases, see Durie (1995) and
Wilson and Yeatman (1995). See also May (2001: Ch. 8).
3 See, for example, demands by Muslim girls in France to wear headscarves (or
the hijab) in the classroom (Galeotti, 1994; Moruzzi, 1994); the exemption from
mandatory high school attendance of Amish children in the US (Galstone,
1995); and attempts by born again Christians in Tennessee to withdraw their
children from a school reading programme, the contents of which they regard
as inhospitable to their beliefs (Macedo, 1995).
4 Thus, in the US, the Supreme Court has upheld the right of the Amish to keep
their children out of local high schools (Wisconsin v. Yoder, 1972), while New
Zealand recognizes claims by the Maori to ancestral land.
5 While Habermas (1993) supports rights to cultural membership, he stresses that
this does not imply that we should seek to preserve cultures in the same manner
in which we may attempt to preserve endangered species. Cultural rights are
legitimate only when exercised as individual liberties. While the constitutional
state should ensure that the reproduction of particular lifeworlds remains
possible, it is up to the members of a particular society or cultural group to
decide whether they want to preserve their cultural heritage. For Habermas,
cultures and traditions only deserve protection in as far as they promote the
wellbeing and freedom of individuals. Individual members of a cultural group
must therefore be free to question existing cultural practices, traditions and
identities. Thus, individual autonomy provides the rationale for, and sets the
limits to, rights to cultural recognition. For Habermas, cultural rights are
valuable because they provide citizens with meaningful cultural choices and
ensure that individuals are not discriminated against for the lifestyle they
choose. Habermass approach echoes Kymlickas distinction between internal
and external protections. While the former deals with the relationship between
the group and its own members, the latter involves the claims of a group against
the larger society. Since, according to Kymlicka, liberals value cultural




membership because it enables individuals effectively to exercise autonomy,

groups may not seek to place internal restrictions on their members. Such
restrictions would undermine individual freedom and are therefore not
compatible with liberal principles.
The Canadian mosaic is frequently contrasted with the US ideal of a melting
pot. While the American model of a melting pot seeks to assimilate or integrate
immigrants into the wider US culture, the mosaic approach reflects a conception of Canadian society as multilingual, multinational and multicultural (see
Bader, 1997).
It is important to ensure that the constitutional arrangements reflect the actual
preferences of ordinary citizens rather than the desires of the political elites of
given groups. As Margaret Moore (2001: 118) notes, provincial elites in Englishspeaking Canada have been engaged in elite-driven province-building, which
is out of step with the desires and aspirations of most (English-speaking)
Canadians, who identify with the central government. While provincial leaders
in Alberta, British Columbia, Ontario and elsewhere have demanded the same
rights as Qubec, opinion polls indicate that English-speaking Canadians would
prefer a stronger role for the central Canadian government.
Indeed, it is precisely these measures that have proved particularly controversial. In 1984, the Canadian Supreme Court ruled that, in addition to the
provisions for English-language education in the original bill, children whose
parents had received elementary instruction in English anywhere in Canada,
and the siblings of a child who had received, or was receiving, elementary or
secondary instruction in English in Canada, could also attend English-language
schools in Quebec (May, 2001: 229). Furthermore, in Ford v. Quebec (1988),
the Supreme Court ruled that the commercial signage restriction was discriminatory (May, 2001).
Supporters of the Qubec language laws, such as May (2001), contend that
these laws are not as exclusionary as they first appear. May argues that access
to English is not being denied, since English remains the dominant language in
Canada, which ensures that francophones are invariably bilingual in English.
Furthermore, advocates of these laws stress that the only way that individual
bilingualism can be maintained and fostered is by, counterintuitively, setting
strict limits on the extent of institutional bilingualism (2001: 231). Finally, May
(2003) notes that to compel children to learn a minority language constitutes
only one of many constraints placed upon childrens education. However, these
considerations arguably do not provide a conclusive case against allowing
children of French-speaking parents or immigrants to choose to attend Englishspeaking schools. If we accept that the dominance of English in Canada ensures
that francophones are bilingual in French, then, by the same token, the
dominance of French in Qubec will ensure that even children who attend
English-speaking schools will develop a good competence in French. Yet, even
if such a policy would in the long run undermine bilingualism, the right of
members to exit the group must be respected on liberal grounds. Once the state
has taken steps to protect the minority from the external pressures of
the majority, members of the minority must be free to choose to leave their
group. To attempt to enshrine bilingualism by fixing the identity of future generations is not readily compatible with the ideals of individual freedom and









self-determination upon which liberalism is premised. Finally, while children

will inevitably be subject to a whole range of constraints upon their educational
choices, liberals will be concerned to keep such constraints to a minimum.
See, for example, the cases discussed by Galstone (1995) and Macedo (1995).
The traditionally associated language reflects and conveys its culture more
felicitously and succinctly than other languages, shapes the customary thinking
of the group and symbolically represents the ethnic or national group (May,
2001: 133). For a more detailed discussion of the link between culture and
language, see May (2001: 1325).
According to Rorty (1994), in the context of the American multiculturalism
debate, members of the Jewish American and African American communities
have been pressurized by their communities to define themselves primarily in
terms of their cultural identity. These communities tend to expect their
members actively to participate in promoting specific policies associated with
Jewish or black interests and to vote along ethnic lines. This is not to deny that
members of minorities are also subject to external pressures to define themselves in terms of their cultural or ethnic group. Undoubtedly, given the salience
of race in the US, African Americans are subject to considerable external
pressures to define themselves as black. From a liberal perspective, both
internal and external pressures constitute an infringement of internal liberty.
A number of feminist writers have expressed concern about the degree to which
recognition of cultural and religious diversity is compatible with a commitment
to gender equality. See, for example, Nussbaum (1999) and Okin (1999). In
response to such concerns, advocates of cultural rights, such as Kymlicka (1999),
have stressed that, although the internal restrictions that most cultures place
upon women are problematic, feminism and multiculturalism nonetheless share
important common concerns. Both draw attention to the structure of societal
institutions, both are critical of traditional liberal notions of citizenship and
both argue that genuine equality requires group-specific rights.
See Carens (2000) and Tully (1995). While this example highlights some of the
potential tensions between the aspirations of women for greater gender
equality and the desire of communities to perpetuate their way of life, this is,
of course, not to suggest that all aboriginal groups are illiberal towards women.
Similarly, the English and the Scots in the UK and the Catalans and Castilians
in Spain continue to share important cultural commitments. This is, of course,
not to suggest that a common national identity in itself is sufficient to ensure
that the public institutions of a state function effectively. To do this, a state will
also have to secure law and order and a range of instrumental goods.
Shane ONeill (this issue), for example, explores the idea of a binational constitution as a form of joint sovereignty. Another avenue is indicated by the international indigenous peoples movement, which has sought to challenge existing
political boundaries through direct access to and participation within the UN
(see Feldman, 2001; Thornberry, 2002). These efforts have given rise to much
greater sensitivity to the concerns of indigenous peoples. The UN, for example,
is currently considering a Draft Declaration on the Rights of Indigenous
Peoples, which recognizes the right to self-determination and demands that
states observe a range of collective rights, respect indigenous autonomy and
customary law and institutions, protect the people from genocide and




ethnocide, abstain from removing them from their lands and territories, respect
their traditions and indigenous knowledge, educate them in their own
languages, restore and protect the environment, respect indigenous citizenship
and allow international adjudication of treaties and agreements between states
and the indigenous. (Thornberry, 2002: 521)
As Thornberry notes, while international human rights texts recognize the
link between persons and community, they eschew corporate conceptions of
rights; that is, rights held by the group as such against outsiders and its own
members. Consequently, these rights fall short of the notion of deep diversity
proposed by Taylor and Carens.
17 To fall under the jurisdiction of a sentencing circle, the offender must consent
to the procedure and must have deep roots in the community. As Shachar notes,
North American aboriginal communities historically relied upon a range of
measures to maintain order, including ostracism, censure and ridicule. Given
that criminal statistics in Canada, Australia and New Zealand suggest that aboriginal peoples are overrepresented in charges laid, court appearances and rates
of incarceration, aboriginal peoples clearly have an interest in the operation and
administration of the criminal justice system.
18 Clearly, intercultural dialogue will only promote dialogue and mutual adaptation if majorities are willing to make meaningful accommodations to minorities. Given the current disparities in power between minority and majority
groups, it is frequently difficult to ensure such genuine dialogue. Feldman
(2001), for example, argues that state responses to attempts by indigenous
peoples to challenge the dominant construction of the nation state exemplify
an anti-dialogic stance. In the light of these difficulties, the initial presumptions
in negotiations should favour minorities.
19 For example, under the Muslim form of divorce known as the talaq, a husband
may unilaterally divorce his wife without recourse to any court or extraneous
authority while, under Orthodox Jewish law, a woman can only gain a divorce
if her husband gives her a get (or consent to divorce) while a man can obtain a
divorce without his wifes consent.

Bader, V. (1997) The Cultural Conditions of Transnational Citizenship on the
Interpretation of Political and Ethnic Cultures, Political Theory 25(6): 771813.
Bellamy, R. (2000) Citizenship beyond the National State: The Case for Europe,
in N. OSullivan (ed.) Political Theory in Transition, pp. 91102. London:
Canovan, M. (1996) Nationhood and Political Theory. Cheltenham: Edward Elgar.
Carens, J. (1997) Two Conceptions of Fairness: A Response to Veit Bader, Political
Theory 25(6): 81420.
Carens, J. (2000) Culture, Citizenship and Community. Oxford: Oxford University
Durie, E. (1995) Justice, Biculturalism and the Politics of Law, in M. Wilson and
A. Yeatman (eds) Justice and Identity, pp. 3344. Wellington: Bridget Williams



Feldman, A. (2001) Transforming Peoples and Subverting States, Ethnicities 1(2):

Fossum, J. (2001) Deep Diversity versus Constitutional Patriotism: Taylor,
Habermas and the Canadian Constitutional Crisis, Ethnicities 1(2): 179206.
Galeotti, A. (1994) A Problem with Theory: A Rejoinder to Moruzzi, Political
Theory 22(4): 6737.
Galeotti, A. (2002) Toleration as Recognition. Cambridge: Cambridge University
Galstone, W. (1995) Two Concepts of Liberalism, Ethics 105: 51634.
Gilbert, P. (1998) The Philosophy of Nationalism. Boulder, CO: Westview Press.
Habermas, J. (1993) Struggles for Recognition in Constitutional States, European
Journal of Philosophy 1(2): 12855.
Habermas, J. (1995) Multiculturalism and the Liberal State, Stanford Law Review
47(5): 84953.
Horton, J. (1998) Charles Taylor: Selfhood, Community and Democracy, in
A. Carter and G. Stokes (eds) Liberal Democracy and its Critics, pp. 15574.
Cambridge: Polity Press.
Houston, C. (1998) Alternative Modernities, Islamism, and Secularism on Charles
Taylor, Critique of Anthropology 18(2): 23440.
Kymlicka, W. (1995) Multicultural Citizenship. Oxford: Clarendon Press.
Kymlicka, W. (1999) Liberal Complacencies, in S.M. Okin Is Multiculturalism Bad
for Women?, pp. 314. Princeton, NJ: Princeton University Press.
Macedo, S. (1995) Liberal Civil Education and Religious Fundamentalism: The
Case of God v. John Rawls, Ethics 105: 46896.
May, S. (2001) Language and Minority Rights: Ethnicity, Nationalism and the Politics
of Language. Harlow: Longman.
May, S. (2003) Misconceiving Minority Language Rights: Implications for Liberal
Political Theory, in W. Kymlicka and A. Patten (eds) Language Rights and
Political Theory, pp. 12352. Oxford: Oxford University Press.
Moore, M. (2001) The Ethics of Nationalism. Oxford: Oxford University Press.
Moruzzi, N.M. (1994) A Problem with Headscarves, Political Theory 22(4): 65372.
Nicholson, L. (1996) To Be or Not to Be: Charles Taylor and the Politics of Recognition, Constellations 3(1): 116.
Nussbaum, M. (1999) Sex and Social Justice. Oxford: Oxford University Press.
Okin, S.M. (1999) Is Multiculturalism Bad for Women? Princeton, NJ: Princeton
University Press.
Parekh, B. (2000) Rethinking Multiculturalism: Cultural Diversity and Political
Theory. London: Macmillan.
Raz, J. (1994) Multiculturalism: A Liberal Perspective, Dissent (Winter): 6779.
Ripstein, A. (1995) Recognition and Cultural Membership, Dialogue 34(2):
Rockerfeller, S. (1992) Comment, in C. Taylor and A. Gutmann (eds) Multiculturalism and the Politics of Recognition, pp. 8798. Princeton, NJ: Princeton
University Press.
Rorty, A. (1994) The Hidden Politics of Cultural Identification, Political Theory
22(1): 15266.
Shachar, A. (2001) Multicultural Jurisdictions: Cultural Differences and Womens
Rights. Cambridge: Cambridge University Press.




Taylor, C. (1992) The Politics of Recognition, in C. Taylor and A. Gutmann (eds)

Multiculturalism and the Politics of Recognition, pp. 2573. Princeton, NJ:
Princeton University Press.
Taylor, C. (1993) Reconciling the Solitudes. Montreal and Kingston: McGill-Queens
University Press.
Thornberry, P. (2002) Minority and Indigenous Rights at the End of History,
Ethnicities 2(4): 51537.
Tully, J. (1995) Strange Multiplicity. Cambridge: Cambridge University Press.
Wilson, M. and A. Yeatman, eds (1995) Justice and Identity. Wellington: Bridget
Williams Books.
Wolff, S. (1992) Comment, in C. Taylor and A. Gutmann (eds) Multiculturalism
and the Politics of Recognition, pp. 7585. Princeton, NJ: Princeton University

ANDREA BAUMEISTER is a senior lecturer in the department of

politics at the University of Stirling and author of Liberalism and the
Politics of Difference (Edinburgh University Press, 2000). Her current
research focuses on the dilemma that multicultural citizenship poses for
liberal democracies and feminist attempts to reconceptualize citizenship
in the light of the politics of difference. Address: Department of Politics,
University of Stirling, Stirling FK9 5SG, Scotland. [email:]