Ethnopolitics

ISSN: 1744-9057 (Print) 1744-9065 (Online) Journal homepage: www.tandfonline.com/journals/reno20

Multicultural Odysseys

Will Kymlicka

To cite this article: Will Kymlicka (2007) Multicultural Odysseys, Ethnopolitics, 6:4, 585-597,
DOI: 10.1080/17449050701659789
To link to this article: https://doi.org/10.1080/17449050701659789

Published online: 19 Dec 2007.

Submit your article to this journal

Article views: 3129

View related articles

Citing articles: 42 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=reno20
Ethnopolitics, Vol. 6, No. 4, 585– 597, November 2007

SYMPOSIUM

Multicultural Odysseys

WILL KYMLICKA
Department of Philosophy, Queen’s University, Canada

The last 40 years have witnessed a revolution around the world in relations between states
and ethnocultural minorities. Older models of assimilationist and homogenizing nation
states are increasingly contested and often displaced by newer ‘multicultural’ models of
the state. In some countries, this shift is the result of endogenous domestic political pro-
cesses, negotiated in response to intense mobilization by particular minorities. But increas-
ingly this shift has an international dimension to it. International organizations are
encouraging and sometimes pressuring states to adopt a more multicultural approach.
We can distinguish two levels at which multiculturalism is being globalized. First, there
is the diffusion of the political discourse of multiculturalism: international networks of
policy makers, non-governmental organizations and scholars are circulating a set of
ideas about the importance of accommodating diversity. On any given day of the year,
somewhere in the world an international organization is sponsoring a seminar or publish-
ing a report to publicize ideas of multiculturalism. These activities often involve sharing
‘best practices’ in various countries, building transnational networks of experts and advo-
cates, creating space for the safe expression of politically sensitive topics, and training
local educators, bureaucrats, non-governmental organizations and media in the challenges
of accommodating a multi-ethnic population.
Second, there is the codification of multiculturalism in international legal norms, which
are embodied in declarations of minority rights. Efforts at developing international norms
of minority rights have proliferated in the past 15 years, at both the global and regional
levels. Globally, the United Nations (UN) adopted a Declaration on the Rights of
Persons Belonging to National or Ethnic, Religious and Linguistic Minorities in 1992
and is debating a Declaration on the Rights of Indigenous Peoples in 2007. The United
Nations Educational, Scientific and Cultural Organization (UNESCO), the International
Labour Organization and the World Bank have also developed norms on minority or

Correspondence Address: Will Kymlicka. Department of Philosophy, Queen’s University, Kingston, Ontario
K7L 3N6, Canada. Email:kymlicka@queensu.ca

1744-9057 Print/1744-9065 Online/07/040585–13 # 2007 The Editor of Ethnopolitics

DOI: 10.1080/17449050701659789
586 W. Kymlicka

indigenous rights. Declarations have also been drafted by regional organizations, such as
the Council of Europe’s 1995 Framework Convention for the Protection of National Min-
orities or the Organization of American States’ 1997 draft Declaration on the Rights of
Indigenous Peoples.
The aim of my new book is to identify some of the dilemmas raised by the global diffu-
sion of multiculturalism. I start from the premise that the underlying ideals being promoted
are attractive ones. To be sure, ideas of multiculturalism and minority rights can be
invoked in illiberal or undemocratic ways in order to perpetuate gender and caste inequal-
ities or to legitimize unjust cultural practices. However, international organizations have
clearly repudiated such illiberal conceptions of multiculturalism. In the words of
UNESCO’s Universal Declaration on Cultural Diversity, ‘No one may invoke cultural
diversity to infringe upon human rights guaranteed by international law, nor to limit
their scope’. Multiculturalism is defended as a supplement to and extension of universal
human rights, not as a justification for limiting universal rights in the name of cultural rela-
tivism or cultural tradition.
Indeed, the conception of multiculturalism being advanced by international organiz-
ations is fundamentally liberal – democratic in character. It is formulated as an evolution
of liberal norms of universal human rights and operates within the constraints of those
norms. As such, it is broadly consistent with the theories of ‘liberal multiculturalism’
developed by recent Western political theorists, in which multiculturalism is understood
as a concept that is both guided and constrained by a foundational commitment to prin-
ciples of individual freedom and equality.
Having defended this ideal of liberal multiculturalism in my own earlier work, I would
like to believe that its global diffusion is feasible and desirable. And there are some
success stories. International support for indigenous rights in Latin America, for
example, has helped to ameliorate long-standing injustices and to build more just and
inclusive democratic regimes.
However, these successes are the exception not the rule. Around the world, the project
of internationalizing minority rights has run into difficulty. Indeed, there are signs of an
impending retreat from the international commitment to multiculturalism. If this commit-
ment is to be sustained, we need to rethink the underlying concepts and strategies. My
book is intended to contribute to this rethinking.
In this short summary, I will attempt to highlight some of the key dilemmas facing
efforts to diffuse multiculturalism and suggest how this process can be reconceived.

The Shifting International Context

After 40 years of neglect, minority rights re-emerged in 1990 as a key issue for the inter-
national community. Why this sudden interest? To oversimplify, it resulted from the con-
vergence of two factors: a fear of the spread of ethnic violence after the collapse of
communism and a hope for the possibility of a viable liberal – democratic form of multi-
culturalism. Let me start with the fear. With the collapse of communism, there was
initially great optimism that liberal democracy would emerge around the world. Instead,
what emerged in many post-communist countries was violent ethnic conflict, particularly
in the Balkans and Caucuses. Moreover, there were fears that violent ethnic conflict would
spread from Yugoslavia to Central Europe (with its restive Hungarian minorities) and to
the Baltics and central Asia (with their sizeable Russian minorities), engulfing virtually the
 Multicultural Odysseys 587

entire post-communist world. The collapse of Somalia and Sudan and the genocide in
Rwanda made clear that this problem was not limited to post-communist Europe, but
affected much of the developing world. As a result, ethnic conflict came to be seen as a
serious threat to international peace and security and, indeed, as the main threat, now
that the danger of war between the superpowers had receded. There was a strong
feeling that the international community needed to ‘do something’ to help states
manage this risk.
So the first factor was a pessimistic vision of ethnic politics as a threat to peace, democ-
racy and development. The second factor was the reverse, namely the emergence within
Western democracies of new and seemingly benign forms of ethnic politics. The West
had undergone its own ‘ethnic revival’ starting in the 1960s, with a dramatic upsurge in
political mobilization by a range of different groups, including immigrant groups, substate
nationalist groups (such as the Scots, Catalans and Québécois) and indigenous peoples. As
in post-communist Europe, this Western ethnic revival was initially seen as destabilizing.
By the early 1990s, however, there was a growing sense of optimism that these fears were
overstated and that Western countries had found a way to contain ethnic political mobil-
ization safely within the boundaries of peaceful liberal –democratic politics.
Indeed, many people argued that ethnic mobilization in the West was not just con-
strained by the rules of liberal democracy and human rights, but was in fact inspired by
liberal values and human rights ideals. Ethnic politics in the West was not a threat to
democracy, but was itself a profoundly democratic phenomenon, as historically disadvan-
taged groups challenged their cultural stigmatization or invisibility in the pursuit of greater
freedom and equality. The emergence and institutionalization of ethnic politics, like the
rise of feminist and gay movements, was a manifestation of a broader process of liberal-
ization and democratization, not a threat to it.
This optimistic view was articulated in the new theories of liberal multiculturalism that
emerged in the early 1990s. According to liberal multiculturalists, the sorts of group-dif-
ferentiated minority policies that had emerged in the West since the 1960s were worthy of
support and, indeed, of celebration. This was true of the regional autonomy and language
rights adopted for substate national groups, of the multicultural accommodation policies
adopted for immigrants and of the land claims and self-government rights adopted for indi-
genous peoples. All were helping to build fairer and more inclusive democratic societies
and should be seen as part and parcel of a free and democratic society.
Viewed this way, minority rights contribute to ‘citizenization’. Historically, ethnocul-
tural and religious diversity in the West has been characterized by a range of illiberal
and undemocratic relations, including relations of conqueror and conquered, colonizer
and colonized, settler and indigenous, racialized and unmarked, normalized and
deviant, orthodox and heretic, civilized and backward, ally and enemy, and master and
slave. The task for all liberal democracies has been to turn this catalogue of uncivil
relations into relationships of liberal – democratic citizenship, both in terms of the vertical
relationship between the members of minorities and the state and the horizontal relation-
ships amongst the members of different groups. In the past, it was assumed that the only
way to engage in this process of citizenization was to impose a single undifferentiated
model of citizenship on all individuals. But the emerging theories of minority rights
start from the assumption that this complex history inevitably and appropriately generates
group-differentiated ethnopolitical claims. The key to citizenization is not to suppress
these differential claims, but to filter and frame them through the language of human
588 W. Kymlicka

rights, civil liberties and democratic accountability. And this is what liberal multicultur-
alism aims to do.
So the early 1990s witnessed both a profound pessimism about ethnic politics in the
post-communist and post-colonial world and an equally profound optimism about
ethnic politics in the West. The activities of the international community reflect both of
these perspectives. The apocalyptic fear of spiraling ethnic conflict provided the
impetus for international organizations to get involved: the hopeful belief in a liberal
and democratic form of multiculturalism provided the inspiring goal.

Dilemmas of Diffusing Multiculturalism

The convergence of these factors explains why international organizations took an active
interest in multiculturalism. But in the headlong rush to embrace multiculturalism, several
difficult issues were overlooked and these are now casting a shadow over international
efforts. In the book, I highlight two difficulties: the conditions of liberal multiculturalism
and the categories of liberal multiculturalism.
Let me start with the conditions. The theory and practice of liberal multiculturalism
suggests that multiculturalism can contribute to citizenization, but the historical record
suggests that certain conditions must be in place for it to have its intended effects. Multi-
culturalism cannot be built (or imposed) out of thin air: certain sources and preconditions
must be present.
I discuss a number of these in the book, but let me mention two: the de-securitization of
state – minority relations and the existence of a human rights consensus.

De-securitization
Where states feel insecure in geopolitical terms, fearful of neighbouring enemies, they are
unlikely to treat their own minorities fairly. More specifically, states are unlikely to accord
powers and resources to minorities that they view as potential collaborators with neigh-
bouring enemies.
In the past, this has been an issue in the West. For example, prior to the Second World
War, Italy, Denmark and Belgium feared that their German-speaking minorities were
more loyal to Germany than to their own country and would support attempts by
Germany to invade and annex areas of ethnic German concentration. These countries
worried that Germany might invade in the name of liberating their co-ethnic Germans
and that the German minority would collaborate with such an invasion.
Today, this is a non-issue throughout the established Western democracies with respect
to historic national minorities and indigenous peoples, although it remains an issue with
respect to certain immigrant groups. It is difficult to think of a single Western democracy
where the state fears that a national minority would collaborate with a neighbouring
enemy and potential aggressor. This is partly because Western states do not have neigh-
bouring enemies who might invade them. The North Atlantic Treaty Organization has
removed the possibility of one Western country invading its neighbours. As a result,
the question of whether national minorities and indigenous peoples would be loyal in
the event of invasion by a neighbouring state is moot.
Of course, Western democracies do have long-distance potential enemies, such as
Soviet communism in the past, Islamic jihadism today and perhaps China in some
 Multicultural Odysseys 589

future scenario. But in relation to these long-distance threats, national minorities and indi-
genous peoples are on the same side as the state. If Quebec gains increased powers or even
independence, no one in the rest of Canada worries that Quebec will start collaborating
with Al Qaeda or China to overthrow the Canadian state. An autonomous or independent
Quebec would be an ally of Canada, not an enemy.
In most parts of the world, however, minority groups are still seen as fifth columns
collaborating with neighbouring enemies. This is particularly true where the minority is
related to a neighbouring state by ethnicity or religion or where a minority is found on
both sides of an international border, so that the neighbouring state claims the right to
protect ‘its’ minority. Under these conditions, ethnic relations become ‘securitized’.
Relations between states and minorities are seen not as a matter of normal democratic
debate and negotiation, but as a matter of state security, in which the state has to limit
the democratic process to protect itself. Under conditions of securitization, minority pol-
itical mobilization may be banned, and even if minority demands can be voiced, they will
be rejected by the larger society and the state. After all, how can groups that are disloyal
have legitimate claims against the state? So the securitization of ethnic relations erodes
both the democratic space to voice minority demands and the likelihood that those
demands will be accepted.
In most Western countries, however, ethnic politics have been ‘de-securitized’. Ethnic
politics is just that: normal, day-to-day politics. Relations between the state and minority
groups have been taken out of the ‘security’ box and put in the ‘democratic politics’ box.
This is one essential precondition for liberal multiculturalism to emerge and take root.

Liberal – Democratic Consensus

A second precondition concerns the security, not of the state, but of individuals who would
be subject to self-governing minority institutions. States are unlikely to accept minority
self-government if they fear it will lead to islands of local tyranny within a broader demo-
cratic state.
This too has been a worry in the past in the West, where some long-standing minorities
were seen as carriers of illiberal political cultures. And this fear persists in relation to some
recent immigrant groups. But at least in relation to older minorities, such as national min-
orities and indigenous peoples, it is now widely assumed that there is a deep consensus
across ethnic lines on basic values of liberal democracy and human rights. As a result,
it is assumed that any self-government powers granted to national minorities or indigenous
peoples will be exercised in accordance with shared standards of democracy and human
rights. Everyone accepts that minority self-government will operate within the constraints
of liberal –democratic constitutionalism, which firmly upholds individual rights. Where
minorities have gained autonomy in the West, their self-governing institutions are
subject to the same constitutional constraints as the central government and so have no
legal capacity to restrict individual freedoms in the name of cultural authenticity, religious
orthodoxy or racial purity. Not only is it legally impossible for national minorities in the
West to establish illiberal regimes, but they have no wish to do so. On the contrary, all of
the evidence suggests that members of national minorities are at least as strongly com-
mitted to liberal – democratic values as members of dominant groups, if not more so.
This removes one of the central fears that dominant groups have about minority auton-
omy. In many parts of the world, there is the fear that, once national minorities or
590 W. Kymlicka

indigenous peoples acquire self-governing power, they will use it to persecute, dispossess,
expel or kill anyone who does not belong to the minority group. In Western democracies,
this is a non-issue. Where there is a strong consensus on liberal – democratic values, people
feel confident that, however issues of multiculturalism are settled, their own civil and pol-
itical rights will be respected. No matter how the claims of ethnonational and indigenous
groups are resolved – no matter what language rights, self-government rights, land rights
or multiculturalism policies are adopted – people can rest assured that they will not be
stripped of their citizenship, subjected to ethnic cleansing, jailed without a fair trial or
denied their rights to free speech, association and worship. Put simply, the consensus
on liberal – democratic values ensures that debates over accommodating diversity are
not a matter of life and death. As a result, dominant groups will not fight to the death
to resist minority claims. This, too, is a precondition for the successful adoption of
liberal multiculturalism.
There are other factors that underpin the rise of minority rights in the West, but these
two are pivotal. They also help explain the partial retreat from multiculturalism in some
countries in relation to recent Muslim immigrants, who are often seen as both disloyal
and illiberal. This shows that as perceptions of the underlying conditions change, the
fate of liberal multiculturalism also ebbs and flows.
If this analysis is correct, it suggests obvious difficulties in diffusing liberal multicultur-
alism since these two factors are absent in much of the world. Indeed, it suggests that pro-
moting multiculturalism will not only be difficult, but perhaps even harmful. Where
minorities are potential pawns in unstable regional geopolitics, and/or where human
rights guarantees are weak or absent, attempts to transplant Western models of multicul-
turalism may exacerbate pre-existing relations of enmity and exclusion, rather than con-
tribute to citizenization.
This does not mean that we should abandon the promotion of multiculturalism or defer it
to some utopian future where all the world has become consolidated liberal democracies
united in common geopolitical security arrangements. On the contrary, it is precisely in
conditions of regional insecurity and political non-freedom that minorities are most vulner-
able and most in need of international protection. It does mean, however, that we need to
distinguish what is feasible in the short-term from what is desirable in the long-term. Inter-
national organizations have singularly failed in this task. At present, international organiz-
ations waver between a naive idealism and a harsh pragmatism. They sometimes naively
promote Western models of multiculturalism without attending to the underlying precondi-
tions and sometimes abandon minorities to their fate on the grounds that multiculturalism is
too risky under the difficult circumstances of many non-Western states.
If the international promotion of multiculturalism is to be effective, we need to rethink
how different forms of multiculturalism fit into larger sequences of political reform. We
need to distinguish the minimal standards that can reasonably be expected of all countries,
even under difficult circumstances, from the higher standards appropriate to countries in
propitious circumstances. In this respect, we might draw upon the ideas of ‘progressive
implementation’ developed in the broader human rights field. It is widely recognized
that some of the social rights listed in the International Covenant on Economic, Social
and Cultural Rights cannot immediately be implemented by some of the poorer countries
(e.g. access to free university education). So it is common to distinguish those social
rights that should be immediately and universally applied from those that we seek to
achieve over time as the facilitating conditions are put in place. We need a comparable
 Multicultural Odysseys 591

theory of the progressive implementation of liberal multiculturalism, with different

minority rights kicking in as the underlying conditions are established.
International organizations have sometimes acknowledged the need for such an
account, but issues of sequencing remain poorly understood at the international level.
Without a plausible account of conditions and sequencing, international efforts at diffusing
multiculturalism are doomed to futility.

The Problem of Categories

The second dilemma concerns the categories of liberal multiculturalism. Any attempt to
codify international norms of multiculturalism requires addressing the issue of categories,
i.e. what kinds of minorities are we trying to protect? The sorts of rights that seem realistic
and defensible will largely be determined by the prior question of which types of min-
orities we have in mind as the intended beneficiaries.
There are two strategies for categorizing minority rights. One approach is to formulate
what I call ‘generic minority rights’, which apply to all ethnocultural minorities. The
other is to formulate ‘targeted minority rights’, which apply to particular types of min-
orities, such as indigenous peoples, national minorities, immigrants, the Roma/Gypsies
and so on.
If we look at the actual practice of liberal multiculturalism in the West, minority rights
are highly targeted or group differentiated. Liberal multiculturalism guarantees certain
generic minority rights to the members of all ethnocultural groups, but it also elaborates
a number of targeted categories of minority rights. The precise categories differ from
country to country, but the most common distinction is between ‘old’ minorities who
were settled on their territory prior to it becoming incorporated into a larger country
and ‘new’ minorities who were admitted to a country as immigrants. Within the category
of old minorities, a further distinction is typically drawn between ‘indigenous peoples’ and
other historic minorities, often called ‘national minorities’ or ‘nationalities’.
We see this threefold pattern in several Western democracies. Finland, for example,
accords different rights to the Sami, as an indigenous people, than to the Swedes, who
are a traditional cross-border national minority and both of these old minorities have a
different set of minority rights from those of new minorities established through recent
migration. Similarly, Denmark distinguishes the rights of the indigenous Inuit in Green-
land from the rights of the (cross-border) German national minority and distinguishes
both of these from the rights of new immigrants. Canada distinguishes the rights of the
indigenous aboriginal peoples from those of the historic French colonial settlers of New
France (Quebec) whose presence predated the British colonization of North America,
and distinguishes both of these old minorities from the multicultural rights accorded to
ethnic groups formed by immigration.
Not all countries have all three types of ethnocultural diversity. Australia and New
Zealand have indigenous peoples and immigrant ethnic groups, but no substate national
minorities. Belgium and the UK have national minorities and immigrant groups, but no
indigenous peoples. In all of these cases, however, the framework of liberal multicultur-
alism remains group differentiated. The UK, for example, distinguishes the rights of
its historic substate ‘nations’ (the Scots, Welsh and Irish) from its immigrant-origin
‘minorities’. Indeed, I am not aware of any Western democracy that relies solely on
592 W. Kymlicka

generic minority rights to govern its ethnocultural diversity, without any element of
targeted rights.
A crucial question, therefore, is how international norms deal with the targeted nature of
liberal multiculturalism. To date, international organizations have had difficulty doing so.
They have preferred to rely on generic minority rights and have only articulated targeted
norms in an ad hoc way.
A clear example of the generic minority rights strategy is Article 27 of the UN’s 1966
International Covenant on Civil and Political Rights, which states that

In those States in which ethnic, religious or linguistic minorities exist, persons

belonging to such minorities shall not be denied the right, in community with the
other members of their group, to enjoy their own culture, to profess and practise
their own religion, or to use their own language.

This UN has interpreted this article as applying to all ethnocultural minorities, no matter
how large or small, recent or historic, territorially concentrated or dispersed. Indeed, the
Human Rights Committee has recently declared that Article 27 applies even to visitors
within a country. Article 27, in short, can be seen as a truly generic cultural right – a
right that can be claimed by any individual and carried with them as they move around
the world.
But just for this reason, it suffers from inevitable limitations. If we try to identify cul-
tural rights that apply to all groups, no matter how large or small, new or old, concentrated
or dispersed, it is difficult to think of many substantive examples. And for a long period of
time, Article 27’s ‘right to enjoy one’s culture’ was understood primarily as a guarantee of
negative rights of non-interference, rather than positive rights to assistance, funding,
autonomy or official language status. In effect, it simply reaffirmed that individual
members of minorities must be free to associate with other members of their group to exer-
cise their standard rights of freedom of speech, assembly and conscience.
Understood in this way, the generic Article 27 approach is insufficient to deal with the
issues at stake in most ethnic conflicts around the world, such as those in post-communist
Europe. These conflicts centre on positive claims, such as the right to use a minority
language in courts or local administration, the funding of minority schools, universities
and media, the extent of local or regional autonomy, the guaranteeing of political represen-
tation for minorities, or the prohibition on settlement policies designed to swamp min-
orities in their historic homelands with settlers from the dominant group. Article 27 has
nothing to say about such claims. It protects certain civil rights relating to cultural
expression, but it does not prohibit states from rescinding funding to minority-language
schools and universities, abolishing local autonomy, gerrymandering electoral rules or
constituency boundaries, or encouraging settlers to swamp minority homelands.
The problem here is not the particular wording of Article 27: it is inherent in any strat-
egy that relies exclusively on generic minority rights. The commitment to identifying
generic cultural rights limits the sorts of claims that can be recognized. In particular, it pre-
cludes claims that flow from facts of historic settlement or territorial concentration. Since
Article 27 articulates a universal and portable cultural right that applies to all individuals,
even migrants and visitors, it does not articulate rights that are tied to the fact that a group
is living on (what it views as) its historic homeland. Yet it is precisely claims relating to
residence on a historic homeland that are at stake in all of the violent ethnic conflicts in
 Multicultural Odysseys 593

post-communist Europe, e.g. in Bosnia, Kosovo, Chechnya and Ngorno-Karabakh.

Indeed, homeland claims are at the heart of most violent ethnic conflicts in the West as
well (e.g. the Basque Country, Cyprus, Corsica and Northern Ireland) and in the post-colo-
nial world (e.g. Kashmir, Sri Lanka, Sudan, Palestine, Kurdistan and East Timor). His-
toric/territorial claims are also at the heart of the conflicts between states and
indigenous peoples around the world.
In all of these conflicts, national minorities and indigenous peoples claim the right to
govern themselves in what they view as their historic homeland, including the right to
use their language in public institutions within their traditional territory and to have
their language, history and culture celebrated in the public sphere (e.g. in the naming of
streets, the choice of holidays and state symbols). None of these claims can plausibly
be seen as generic or portable: they only apply to minorities with a particular sort of
history and territory.
It quickly became clear, therefore, that if international organizations wanted to ‘do
something’ about the urgent problem of ethnic conflict, they needed to go beyond
generic minority rights and articulate targeted rights. Several efforts have been made
since 1990 in developing targeted minority rights and specifically to develop norms that
can address claims related to history and territory. Unfortunately, these efforts are
mired in inconsistencies.
In the book, I explore the two major examples of international targeted norms: the
European experiment in formulating targeted norms for ‘national minorities’ and the
UN experiment in formulating targeted norms for ‘indigenous peoples’. (While some com-
mentators have called for the formulation of targeted international norms for immigrants,
these pleas to date have fallen on deaf ears.)

National Minorities
European efforts to develop minority rights norms were initially intended to deal with
violent ethnic conflict in post-communist Europe, and so focused on the particular type of
group involved in such conflicts, namely ‘national minorities’. Whereas Article 27 lumps
together ‘national, ethnic, religious and linguistic’ minorities, the Council of Europe’s
Framework Convention for the Protection of National Minorities applies only to ‘national
minorities’. While there is no universally agreed-upon definition of ‘national minorities’,
the term has a long history in European diplomacy, where it has typically referred to
historically settled minorities living on or near what they view as their national homeland.
Most European countries have explicitly stated that immigrant groups are not national min-
orities and some have also excluded the Roma from the category on the grounds that they are
a non-territorial minority. These exclusions are increasingly contested, as I explain below,
with the result that the traditional understanding of ‘national minority’ co-exists alongside
broader definitions that are less tied to history and territory. But originally at least, European
organizations were targeting their efforts at historically settled substate national minorities,
of the sort involved in violent conflicts in post-communist Europe.
The challenge facing European organizations, then, was to formulate targeted norms
that provide a principled basis for responding to the claims of national minorities, particu-
larly claims based on historical settlement and territorial concentration. In addressing this
challenge, European international organizations naturally looked to the practices of
Western democracies to see what approaches appeared to be working. And here a clear
594 W. Kymlicka

pattern emerges: most sizeable national minorities within Western democracies have
demanded and been accorded territorial autonomy, as well as official language status, at
least within their autonomous territory (Catalonia, Flanders, Tyrol, Quebec, etc.). More-
over, this combination of territorial autonomy and official language status is widely
seen as successful. Indeed it is often cited in the literature as a key illustration of how
liberal multiculturalism can deepen relations of liberal – democratic citizenship, thereby
reducing intergroup hierarchies while protecting individual rights and freedoms.
One option, therefore, was to formulate a principle of autonomy or self-government for
national minorities of the sort that underpins the ‘best practices’ of Western democracies.
And, indeed, such a principle was implicitly endorsed by all the major European organiz-
ations in the early 1990s: by the Organization for Security and Cooperation in Europe in
Article 35 of its Copenhagen Declaration of 1990, by the EU’s European Commission in
Article 5 of Chapter II of its Yugoslavia Draft Convention of 1991 and by the Council of
Europe’s Parliamentary Assembly in Article 11 of its Recommendation 1201 of 1993.
However, these early statements of support for autonomy for national minorities were
quickly abandoned and were deliberately excluded from what is now the dominant
legal framework of European minority rights, namely the 1995 Framework Convention
for the Protection of National Minorities.
Why were principles of autonomy dropped between 1993 and 1995? The answer relates
back to the issue of conditions. The two preconditions that facilitated the adoption of
autonomy for national minorities in the West were not present in post-communist
countries in the 1990s. These countries inhabited an unstable region, surrounded by his-
toric enemies, and had reason to fear that their minorities would serve as fifth columns
for predatory neighbours. (And, indeed, some ethnic Serbs in Bosnia and Croatia did
serve as fifth columns for Serbian aggression against their own country.) These countries
also lacked firm protections of human rights and so had reason to fear that, if national min-
orities achieved self-government, they would suppress the rights of anyone who did not
belong to their group. (And, indeed, when minorities seized territory and established
self-government, they often engaged in intimidation and ethnic cleansing, e.g. in
Abkhazia, Kosovo, Republica Srbska, etc.)
Under these circumstances, it was felt that imposing autonomy on post-communist
countries was inappropriate and potentially destabilizing. Indeed, it was felt that virtually
any minority claims relating to history and territory were potentially destabilizing. If the
international community legitimated the idea that a minority had special claims over its
historic homeland, this would escalate fear and invite minority belligerence and kin-state
intervention. These negative effects could arise, not only with respect to claims for
autonomy, but even for much more modest claims, such as bilingual street signs in areas
of traditional settlement. As a result, when the 1995 Framework Convention for the
Protection of National Minorities was being formulated, virtually all minority rights that
were premised on facts of historical settlement or territorial concentration were rejected.
Instead, the Framework Convention for the Protection of National Minorities essentially
restated Article 27’s generic rights to non-discrimination and cultural liberty.
This retreat from substantive targeted norms for national minorities is understandable,
given the missing preconditions for liberal multiculturalism. But the result is unstable. At
present, the Framework Convention for the Protection of National Minorities is formally
targeted at one type of group, namely historic/territorial national minorities, but its sub-
stantive content is essentially generic. If the Framework Convention for the Protection of
 Multicultural Odysseys 595

National Minorities does not recognize any claims based on history and territory, why
should it only apply to historically settled groups? If the Framework Convention for the
Protection of National Minorities only recognizes generic rights, why should it not
apply generically?
And, indeed, we see a push to expand the scope of the Framework Convention for the
Protection of National Minorities to cover groups that are not traditionally understood as
national minorities, such as the Roma, ethnic Russian settlers in the Baltics or post-war
immigrants. The very term ‘national minority’ is now being redefined as an umbrella
term that covers all ethnocultural groups residing in a country, rather than marking out
a particular type of historic/territorial minority. This push to broaden the scope of the
Framework Convention for the Protection of National Minorities remains contested,
but, if it is accepted, it would complete the retreat from its original goals. The initial motiv-
ation for the Framework Convention for the Protection of National Minorities was the per-
ception that the generic strategy was insufficient to deal with the issues of history and
territory underpinning post-communist ethnic conflicts, and that targeted norms were
therefore required. However, it proved impossible to reach agreement on how to deal
with such claims, given the difficult circumstances of post-communist countries, and so
European organizations have retreated towards a generic approach, turning the Framework
Convention for the Protection of National Minorities into a European version of the UN’s
Article 27. As a result, the very idea of targeted norms for national minorities remains sus-
pended in mid-air, its future unclear.

Indigenous Peoples
The second experiment in targeted norms is more promising. There is a broad consensus at
the UN that targeted norms are needed for indigenous peoples and that these norms should
address claims related to history and territory. Indigenous peoples are increasingly
acknowledged to have a wide range of rights relating to their areas of traditional settle-
ment, including rights to political self-government, control over natural resources,
language rights and autonomous legal, cultural and educational institutions.
Why has the UN been able to develop targeted norms for indigenous peoples when
European organizations failed to do so for national minorities? One reason is that indigen-
ous peoples are not seen as posing the same threat to the state as national minorities.
Indigenous peoples are seen as weak, remote and vulnerable groups, incapable of challen-
ging the state and isolated from larger geopolitical struggles. National minorities, by
contrast, are seen as powerful in their own right and as having powerful allies in neigh-
bouring states. As a result, many states have taken the position that autonomy rights
should be accepted for indigenous peoples, but rejected for national minorities, who are
only entitled to generic Article 27 rights.
This position presupposes that we can draw a sharp line between indigenous peoples
and national minorities, and that we can accord substantive rights to the former while
only according generic rights to the latter. I call this the ‘firewall’ strategy: the attempt
to maintain an unbridgeable gulf between the status of indigenous peoples and national
minorities in international law. This strategy was probably necessary to gain initial
support for the international indigenous rights movement, but I do not believe it is sustain-
able. It is morally inconsistent, conceptually unstable and politically unsustainable. What-
ever reasons we have for protecting the interests that indigenous peoples have in their
596 W. Kymlicka

historic territories will also apply to many national minorities. Indeed, in many parts of
Asia and Africa, there is no non-arbitrary way to distinguish indigenous peoples from
national minorities. And, as a result, an increasing number of groups around the world
that used to be called national or ethnic minorities are redefining themselves as ‘indigen-
ous peoples’ in order to benefit from international protection. The firewall is disintegrat-
ing, leaving it unclear which groups count as ‘indigenous peoples’. And this is generating
growing unease about international indigenous rights norms. States will not endorse strong
international norms of indigenous rights if they think this will create opportunities for
national minorities to make comparable claims and there is no longer any firewall to
prevent this. The difficulty in securing adoption by the UN General Assembly of the
draft Declaration On Indigenous Rights is one of many signs that the indigenous rights
track may now be in retreat and may only survive in a weakened form.
In short, the fate of targeted minority rights norms remains unsettled both within Europe
and the UN. Many international organizations would prefer to stick to a purely generic
approach that does not require putting ethnocultural groups into different categories for
the purposes of international law. But in both the European and UN contexts, a purely
generic approach has proven insufficient. It is part of the nature of ethnic conflict and
part of the logic of liberal multiculturalism that it raises group-differentiated issues,
often related to the role of history and territory. This suggests that some degree of targeting
is essential if international norms are to play a constructive role.
Yet we are far from having a coherent account of the kinds of targeting that are appro-
priate in international law. To date, targeted norms have emerged in an ad hoc fashion and
are often presented as unique exceptions to the rule of generic minority rights. But this sort
of ad hoc ‘mono-targeting’ – isolating one particular type of group for distinctive legal
rights, while according all other groups only generic minority rights – is unlikely to be
stable. If we start down the road of targeted norms, we need to do so in a more systematic
fashion. It may be that the only way to develop a sustainable set of targeted norms for indi-
genous peoples, for example, is to make it part of a broader strategy of multi-targeting,
operating in conjunction with separate targeted norms for national minorities or other
homeland groups, for immigrant groups, or for other types of groups with unique histories
and needs, such as the Roma or Afro-Latinos. Otherwise, any set of mono-targeted norms
will face intense pressure to expand or redefine the category in ways that are unsustainable,
as we are seeing both in Europe and internationally. Unfortunately, there have been few
attempts to step back and ask how to structure a multi-targeted system of minority
rights. Without a more plausible approach to the issue of categories, international
efforts at diffusing liberal multiculturalism are unlikely to succeed.

Conclusion
The future of the global diffusion of liberal multiculturalism depends on finding better
answers to these dilemmas of conditions and categories. In the book, I speculate about
what a more effective system of international minority rights might look like. Ideally,
such a system would be both multi-targeted and sequenced, containing targeted norms
for a wider range of groups, all of which would be articulated in a sequenced way,
linking the implementation of rights to the presence of underlying conditions.
However, to be honest, I am not optimistic about the likelihood of a breakthrough in the
foreseeable future. The political will to tackle issues of minority rights has dissipated since
 Multicultural Odysseys 597

the mid-1990s. Both the profound pessimism about ethnic conflict and the profound opti-
mism about liberal multiculturalism have faded, reducing the sense of urgency and hope
that inspired the earlier wave of international efforts. We may need to accept that this is as
good as it gets.
If so, there are certainly grounds for supporting the current framework. It has clearly
helped to highlight the plight of many minorities around the world and to make space
at the table for ethnic political actors to voice their concerns peacefully and democrati-
cally, while simultaneously encouraging them to formulate their claims within a human
rights framework.
These are significant accomplishments. My concern, however, is that the status quo is
not only imperfect, but in fact unstable. Unless we can think of intellectually compelling
and politically viable ways of dealing with the problems of conditions and categories, the
likely outcome will be a retreat from the more progressive aspects of the current system.
Indeed there is already evidence that this is taking place. And, if so, the long-term prog-
nosis for the peaceful global diffusion of liberal multiculturalism is poor indeed.