IHS Political Science Series

Working Paper 40
February 1997

Education for Citizenship

Will Kymlicka
Impressum

Author(s):
Will Kymlicka

Title:
Education for Citizenship

ISSN: Unspecified

1997 Institut für Höhere Studien - Institute for Advanced Studies (IHS)
Josefstädter Straße 39, A-1080 Wien
E-Mail: offi[email protected]
Web: www.ihs.ac.at

All IHS Working Papers are available online: http://irihs.ihs.ac.at/view/ihs_series/

This paper is available for download without charge at: https://irihs.ihs.ac.at/id/eprint/964/
Institut für Höhere Studien (IHS), Wien
Institute for Advanced Studies, Vienna
Reihe Politikwissenschaft / Political Science Series No. 40

Education for Citizenship

Will Kymlicka
2 — Will Kymlicka / Education for Citizenship — I H S

Education for Citizenship

Will Kymlicka

Reihe Politikwissenschaft / Political Science Series No. 40

February 1997

Will Kymlicka
Philosophy Department
University of Ottawa
70 Laurier, 2nd Floor
Ottawa K1N 6N5 Canada
e-mail: [email protected]

Institut für Höhere Studien (IHS), Wien

Institute for Advanced Studies, Vienna
The Political Science Series is published by the Department of Political Science of the
Institute for Advanced Studies (IHS), Vienna. The purpose of this series of papers is to
encourage debate in a wider professional audience. The authors are fully responsible for the
content. Contributions from guests are also noted.

All rights are reserved.

4 — Will Kymlicka / Education for Citizenship — I H S

Abstract

Although it is widely accepted that a basic task of schooling is to prepare each new generation
for their responsibilities as citizens, the appropriate form and content of citizenship education
is often controversial. This paper discusses some of these controversies. I begin by arguing
that citizenship is more complicated than is often realized, and that even ‘minimal’ conceptions
of citizenship impose significant obligations and constraints on individual and group behaviour. I
then consider three inter-related areas of debate: whether citizenship education requires
common schooling; whether promoting responsible citizenship requires promoting personal
autonomy; and whether promoting a shared civic identity requires teaching not only shared
political values or principles but also promoting particular national or cultural identities. These
three issues help illustrate the centrality of education for citizenship to both political theory and
educational philosophy.

Keywords

citizenship, education, democracy, liberalism

Note
“Education for Citizenship” will be published in Terence McLaughlin and Mark Halstead (eds.),
Foundation for Moral Education.
Will Kymlicka was Visiting Professor at the Department of Political Science of the Institute for
Advanced Studies in January 1997.
6 — Will Kymlicka / Education for Citizenship — I H S
Contents

1. The Nature and Importance of Citizenship 2

2. The Need for Citizenship Education in Schools 8

3. Citizenship and Separate Schools 11

4. Citizenship and Personal Autonomy 13

5. Citizenship and National Identity 17

Bibliography 21
 I H S — Will Kymlicka / Education for Citizenship — 1

It is widely accepted that a basic task of schooling is to prepare each new generation for their
responsibilities as citizens. Indeed, the need to create a knowledgeable and responsible
citizenry was one of the major reasons for establishing a public school system, and for making
education mandatory. Education for citizenship includes, but also goes far beyond, classes in
‘civics’. Citizenship education is not just a matter of learning the basic facts about the
institutions and procedures of political life; it also involves acquiring a range of dispositions,
virtues, and loyalties which are intimately bound up with the practice of democratic citizenship.
Children acquire these virtues and loyalties not just (or even primarily) in civics classes. Rather,
they are inculcated throughout the educational system. The aim of educating citizens affects
what subjects are taught, how they are taught, and in what sorts of classrooms. In this sense,
education for citizenship is not an isolated subset of the curriculum, but rather is one of the
ordering goals or principles which shapes the entire curriculum.

In this paper, I will discuss some of the issues raised by citizenship education. I will begin by
considering what citizenship means in modern democratic societies, and what sorts of
capacities and dispositions it requires (section 1). I hope to show that liberal democratic
citizenship is more complicated than is often realized, and that even ‘minimal’ conceptions of
citizenship impose significant obligations and constraints on individual and group behaviour. I
will then discuss why schools must play a role in educating children for citizenship (section 2).
It would be unrealistic to expect schools by themselves to develop the skills and virtues
needed for democratic citizenship. People learn to be responsible citizens not only in schools,
but also in the family, neighbourhood, churches, and manifold other groups and forums in civil
society. Schools are not the only, or perhaps even the primary, forum for learning citizenship,
but they are, I believe, indispensable. These other institutions supplement, but cannot replace,
the provision of citizenship education in schools.

The rest of the paper will then consider three inter-related areas of controversy: whether
citizenship education requires common schooling (section 3); whether promoting responsible
citizenship requires promoting personal autonomy (section 4); and whether promoting a shared
civic identity requires teaching not only shared political values or principles but also promoting
particular national or cultural identities (section 5). These three issues are by no means
exhaustive of the range of controversies that arise, but they suggest the centrality of education
for citizenship to both political theory and educational philosophy.
I H S — Will Kymlicka / Education for Citizenship — 2

1. The Nature and Importance of Citizenship

There has been an explosion of interest in the concept of citizenship amongst political
theorists. In 1978, it could be confidently stated that “the concept of citizenship has gone out
of fashion among political thinkers” (van Gunsteren 1978:9). Fifteen years later, citizenship has
become the “buzz word” amongst thinkers on all points of the political spectrum (Heater
1990:293).

Interest in citizenship has been sparked by a number of recent political events and trends
throughout the world – increasing voter apathy and long-term welfare dependency in the United
States, the resurgence of nationalist movements in Eastern Europe, the stresses created by
an increasingly multicultural and multiracial population in Western Europe, the backlash
against the welfare state in Thatcher’s England, the failure of environmental policies that rely on
voluntary citizen cooperation, etc.

These events have made clear that the health and stability of a modern democracy depends,
not only on the justice of its ‘basic structure’,1 but also on the qualities and attitudes of its
citizens: e.g., their sense of identity, and how they view potentially competing forms of
national, regional, ethnic or religious identities; their ability to tolerate and work together with
others who are different from themselves; their desire to participate in the political process in
order to promote the public good and hold political authorities accountable; their willingness to
show self-restraint and exercise personal responsibility in their economic demands, and in
personal choices which affect their health and the environment. Without citizens who possess
these qualities, democracies become difficult to govern, even unstable. 2

Many classical liberals believed that a liberal democracy could function effectively even in the
absence of an especially virtuous citizenry, by creating checks and balances. Institutional and
procedural devices such as the separation of powers, a bicameral legislature and federalism
would all serve to block would-be oppressors. Even if each person pursued her own self-
interest, without regard for the common good, one set of private interests would check another
set of private interests. Kant, for example, thought that the problem of good government “can
be solved even for a race of devils” (quoted in Galston 1991:215). However, it has become clear
that procedural-institutional mechanisms to balance self-interest are not enough, and that
some level of civic virtue and public-spiritedness is required.

Consider the many ways that public policy relies on responsible personal lifestyle decisions:
the state will be unable to provide adequate health care if citizens do not act responsibly with

1. Rawls says that the “basic structure” of society is the primary subject of a theory of justice (1993:257–89).
2. This may account for the recent interest in citizenship promotion amongst governments (eg., Britain’s
Commission on Citizenship, Encouraging Citizenship 1990; Senate of Australia, Active Citizenship Revisited 1991;
Senate of Canada, Canadian Citizenship: Sharing the Responsibility 1993).
 I H S — Will Kymlicka / Education for Citizenship — 3

respect to their own health, in terms of maintaining a healthy diet, exercising regularly, and
limiting their consumption of liquor and tobacco; the state will be unable to meet the needs of
children, the elderly or the disabled if citizens do not agree to share this responsibility by
providing some care for their relatives; the state cannot protect the environment if citizens are
unwilling to reduce, reuse and recycle in their own consumer choices; the ability of the
government to regulate the economy can be undermined if citizens borrow immoderate
amounts or demand excessive wage increases; attempts to create a fairer society will flounder
if citizens are chronically intolerant of difference and generally lacking in a sense of justice.
Without cooperation and self-restraint in these areas, “the ability of liberal societies to function
successfully progressively diminishes” (Galston 1991:220). 3

In short, we need “a fuller, richer and yet more subtle understanding and practice of
citizenship”, because “what the state needs from the citizenry cannot be secured by coercion,
but only cooperation and self-restraint in the exercise of private power” (Cairns and Williams
1985:43). Yet there is growing fear that the civility and public-spiritedness of citizens of liberal
democracies may be in serious decline (Walzer 1992:90). 4

Certain virtues are needed in virtually any political order, whether it is liberal and democratic or
not. These would include general virtues, such as courage and law-abidingness, as well as
economic virtues, such as the capacity to delay self-gratification or to adapt to economic and
technological change. 5 But there are also certain virtues which are distinctive to a liberal
democracy, relating to the basic principles of a liberal regime, and to the political role citizens
occupy within it, and it is these which I wish to focus on.

I will consider four such virtues:

(a) public-spiritedness, including the ability to evaluate the performance of those in office, and
the willingness to engage in public discourse;

(b) a sense of justice, and the capacity to discern and respect the rights of others, and to
moderate one’s own claims accordingly;

(c) civility and tolerance;

3. Hence recent theories of citizenship emphasize that citizenship requires a balance of rights and
responsibilities. For a survey of recent work on citizenship theory, which I am drawing on in this section, see
Kymlicka and Norman 1994. For a useful collection of recent articles, see Beiner 1995. For a more historical
survey of citizenship theory, see Walzer 1989, and the readings collected in Clarke 1994.
4. According to a recent survey, only 12% of American teenagers said voting was important to being a good
citizen. Moreover, this apathy is not just a function of youth – comparisons with similar surveys from the previous
50 years suggest that “the current cohort knows less, cares less, votes less, and is less critical of its leaders
and institutions than young people have been at any time over the past five decades” (Glendon 1991:129). The
evidence from Great Britain is similar (Heater 1990:215).
5. For a helpful discussion and typology, see Galston 1991:221–4.
I H S — Will Kymlicka / Education for Citizenship — 4

(d) a shared sense of solidarity or loyalty.

Many commentators argue that the fourth virtue is inapplicable to liberal democracies, or
perhaps more accurately, is redundant, since it supervenes on the first three. On this view,
whatever sense of shared loyalty is required in a liberal democracy simply involves loyalty to
principles of tolerance, justice and democracy. Shared commitment to these basic political
principles is a sufficient foundation for a shared political identity or loyalty. I think that is
mistaken, and will return to this point in section 5 below.

For the moment, however, I want to focus on the first three, starting with ‘public-spiritedness’.
This includes the ability and willingness to engage in public discourse about matters of public
policy, and to question authority. These are perhaps the most distinctive aspects of citizenship
in a liberal democracy, since they are precisely what distinguish ‘citizens’ within a democracy
from the ‘subjects’ of an authoritarian regime.

The need to question authority arises in part from the fact that citizens in a representative
democracy elect representatives who govern in their name. Hence an important responsibility
of citizens is to monitor those officials, and judge their conduct. The need to engage in public
discourse arises from the fact that the decisions of government in a democracy should be
made publicly, through free and open discussion. But the virtue of public discourse is not just
the willingness to participate in politics, or to make one’s views known. Rather, as William
Galston notes, it “includes the willingness to listen seriously to a range of views which, given
the diversity of liberal societies, will include ideas the listener is bound to find strange and even
obnoxious. The virtue of political discourse also includes the willingness to set forth one’s own
views intelligibly and candidly as the basis for a politics of persuasion rather than manipulation
or coercion” (Galston 1991:227).

Stephen Macedo calls this the virtue of “public reasonableness”. Liberal citizens must give
reasons for their political demands, not just state preferences or make threats. Moreover, these
reasons must be “public” reasons, in the sense that they are capable of persuading people of
different faiths and nationalities. Hence it is not enough to invoke Scripture or tradition. 6 Liberal
citizens must justify their political demands in terms that fellow citizens can understand and
accept as consistent with their status as free and equal citizens. It requires a conscientious
effort to distinguish those beliefs which are matters of private faith from those which are
capable of public defense, and to see how issues look from the point of view of those with
differing religious commitments and cultural backgrounds. As I discuss below, this is a
stringent requirement that many religious groups find difficult to accept.

6. See also Audi 1989; Strike 1994.

 I H S — Will Kymlicka / Education for Citizenship — 5

The virtue of public reasonableness is less relevant for citizens who do not wish to participate
in political affairs, and there will always be a portion of the population who have little or no
desire to be politically active. Some people will find their greatest joys and projects in other
areas of life, including the family, or the arts, or religion. A liberal democracy must respect such
diverse conceptions of the good life, and should not compel people to adopt a conception of the
good life which privileges political participation as the source of meaning or satisfaction. 7 For
these more or less apolitical people, the virtue of public reasonableness may be less
important.

Some commentators would argue that most people in contemporary democracies will fall into
this apolitical camp – that meaningful political participation is almost inevitably confined to
elites. According to T.H. McLaughlin, this is one of the important points of division between
‘minimal’ and ‘maximal’ conceptions of citizenship. On the minimal view, citizenship for most
people primarily involves passive respect for laws, not the active exercise of political rights. By
contrast, maximal conceptions of democracy insist that a true democracy, or that political
justice, must aim for more widespread participation (McLaughlin 1992a).

Justice clearly requires that everyone have the opportunity to become active citizens, if they so
choose, which means eliminating any economic or social barriers to the participation of
disadvantaged groups, such as women, the poor, racial and ethnic minorities, etc. But whether
we should encourage all individuals to choose to be active political participants is another
matter. Whether active citizenship should be encouraged depends, I think, on the second virtue
listed above – namely, a sense of justice. To have a sense of justice does not simply mean
that we do not actively harm or exploit others. It also involves the duty to prevent injustice, by
creating and upholding just institutions. So if there are serious injustices in our society which
can only be rectified by political action, then citizens should recognize an obligation to protest
against that injustice. Or if our political institutions are no longer functioning, perhaps due to
excessive levels of apathy, or to the abuse of power, then citizens have an obligation to protect
these institutions from being undermined. To sit passively by while injustices are committed, or
democratic institutions collapse, in the hope that others will step in, is to be a free rider.
Everyone should do their fair share to create and uphold just institutions.

The extent of injustice, and the health of political institutions, will vary from society to society.
In some times and places, though perhaps only in rare and fortunate circumstances, our

7. This is why liberals cannot endorse a strong version of ‘civic republicanism’. In one sense, civic republicanism
refers to any view which highlights the importance of civic virtues, and the extent to which the functioning of a
democracy requires certain virtues and identities amongst its citizens. In this sense, as I have argued, liberals
must be republicans. But in another stronger sense, civic republicanism refers to the view that the best life – the
most truly human life – is one which privileges political participation over other spheres of human endeavour. This
sort of position is defended by Oldfield (1990b), Pocock (1992), Beiner (1992), Skinner (1992), amongst others.
However, it is inconsistent with liberalism’s commitment to pluralism, and in any event is implausible as general
account of the good life for all persons. See Kymlicka and Norman 1994:361–2.
I H S — Will Kymlicka / Education for Citizenship — 6

natural duty of justice will not require us to participate actively. Where a society is basically
well-ordered, and its institutions healthy, then individuals should be free to follow their own
conceptions of the good, even if these give little or no weight to political participation.

So there will be times and places where minimal citizenship is all that we can or should
require. And for minimal citizens, the stringent demands of ‘public reasonableness’ will be less
significant. But even here, the requirements of liberal citizenship are by no means trivial. The
obligations of minimal citizenship are often described in purely negative terms – ie, the
obligation not to break the law, and not to harm others, or restrict their rights and liberties.
Minimal citizenship, in short, is often seen as simply requiring non-interference with others.

But that ignores one of the most basic requirements of liberal citizenship, albeit one that is
often neglected in theoretical discussions. This is the virtue of ‘civility’ or ‘decency’, and it is a
virtue that even the most minimal citizen must learn, since it applies not only to political
activity, but also – indeed, primarily – to our actions in everyday life, on the street, in
neighbourhood shops, and in the diverse institutions and forums of civil society.

Civility refers to the way we treat non-intimates with whom we come into face-to-face contact.
To understand civility, it is helpful to compare it with the related requirement of non-
discrimination. The legal prohibition on discrimination initially only applied to government
actions. Government laws and policies which discriminated against people on the basis of race
or gender have gradually been struck down in Western democracies, since they violate the
basic liberal commitment to equality of opportunity. But it has become clear that whether
individuals have genuinely equal opportunity depends not only government actions, but also on
the actions of institutions within civil society – corporations, schools, stores, landlords etc. If
people are discriminated against by prejudiced shop-owners or real estate agents, they will be
denied equal citizenship, even if the state itself does not discriminate. Hence legal
requirements of non-discrimination have increasingly been applied to ‘private’ firms and
associations.

This extension of non-discrimination from government to civil society is not just a shift in the
scale of liberal norms, it also involves a radical extension in the obligations of liberal
citizenship. For the obligation to treat people as equal citizens now applies to the most
common everyday decisions of individuals. It is no longer permissible for businesses to refuse
to hire black employees, or to serve black customers, or to segregate their black employees or
customers. But not just that. The norms of non-discrimination entail that it is impermissible for
businesses to ignore their black customers, or treat them rudely, although it is not always
possible to legally enforce this. Businesses must in effect make blacks feel welcome, just as if
they were whites. Blacks must, in short, be treated with civility. The same applies to the way
citizens treat each other in schools or recreational associations, even in private clubs.
 I H S — Will Kymlicka / Education for Citizenship — 7

This sort of civility is the logical extension of non-discrimination, since it is needed to ensure
that all citizens have the same opportunity to participate within civil society. But it now extends
into the very hearts and minds of citizens. Liberal citizens must learn to interact in everyday
settings on an equal basis with people for whom they might harbour prejudice.

The extent to which this requirement of civility can (or should) be legally enforced is limited. It
is easier to compel businesses to be non-discriminatory in hiring than to compel them to treat
black customers with civility. But the recent spread of laws and regulations against sexual and
racial harassment, both in society generally and within schools and businesses, can be seen
as an attempt to ensure a level of civility, since they include forms of offensive speech as well
as physical intimidation, And while it is obviously impossible to compel civility between citizens
in less formal settings – e.g, whether whites smile or scowl at an Asian family in the
neighbourhood park – liberal citizenship nonetheless requires this sort of civility.

It is easy to trivialize this requirement of civility as being simply ‘good manners’. Philip Rieff, for
example, dismisses the insistence on civility as a superficial facade that simply hides a deeper
indifference to the needs of others. As he puts it, “We have long known what ‘equality’ means
in American culture: it means... a smile fixed to the face, demanding you return a smile”
(quoted in Cuddihy 1978:6). John Murray Cuddihy views civility as the imposition of a
Protestant (and bourgeois) sense of ‘good taste’ on other religious groups. He argues that
Catholics and Jews (and now Muslims) have had to abandon their conception of true faith,
which required the public expression of contempt for other religions, to conform to this ‘religion
of civility’.

It is true that liberal societies have reinforced, and thereby partially conflated, the moral
obligation of civility with an aesthetic conception of ‘good manners’. For example, the
expectation of civility is sometimes used to discourage the sort of forceful protest that may be
needed for an oppressed group to be heard. For a disadvantaged group to ‘make a scene’ is
often seen as ‘in bad taste’. This sort of exaggerated emphasis on good manners can be used
to promote servility. True civility does not mean smiling at others no matter how badly they
treat you, as if oppressed groups should be nice to their oppressors. Rather, it means treating
others as equals on the condition that they extend the same recognition to you. While there is
some overlap between civility and a more general politeness, they are nonetheless distinct –
civility involves upholding norms of equality within the public life of a society, including civil
society, and thereby upholding essential liberal values.8

8. My discussion here draws extensively on Jeff Spinner’s account of civility (1994:chap. 3). It also draws on
Patricia White’s account of civility, or what she calls ‘decency’ (1992), although I disagree in part with her
emphasis. She seems primarily concerned with improving the overall level of ‘decency’ in society, rather than with
eliminating glaring instances of incivility aimed at identifiable groups. For example, she compares the smiling and
cooperative waiters in a Canadian cafe with the surly and uncooperative waiters in a Polish cafe (1992:208), and
argues that we should educate children to be friendly with strangers rather than surly. While I agree that it’s a
I H S — Will Kymlicka / Education for Citizenship — 8

2. The Need for Citizenship Education in Schools

Even the most minimal conception of liberal citizenship, therefore, requires a significant range
of civic virtues. But are schools the appropriate arena to teach these virtues, given that this
would involve inculcating substantive (and controversial) moral beliefs? I believe the schools
have an unavoidable role, in part because no other social institution can take their place.

To be sure, other institutions can play a supplementary role in promoting civic virtue. For
example, theorists of the ‘New Right’ often praise the market as a school of virtue. Many
Thatcher/Reagan reforms of the 1980s aimed to extend the scope of markets in people’s lives
– through freer trade, deregulation, tax cuts, the weakening of trade unions, and reducing
welfare benefits – in part in order to teach people the virtues of initiative and self-reliance.
Moreover, markets are said to encourage civility, since companies which refuse to hire black
employees, or serve black customers, will be at a competitive disadvantage.

However, the limits of the market as a school of civic virtue are clear. Many market
deregulations arguably made possible an era of unprecedented greed and economic
irresponsibility, as evidenced by the savings-and-loan and junk bond scandals in America.
Markets teach initiative, but not a sense of justice or social responsibility (Mulgan 1991:39).
And so long as a sizeable portion of the population harbours prejudices towards certain groups,
then businesses will have an economic incentive to serve that market, by creating goods and
services that exclude these groups.9 In any event, how the market cannot teach those civic
virtues specific to political participation and dialogue – eg., the virtue of public reasonableness.

Following Rousseau and J.S. Mill, many ‘participatory democrats’ assume that political
participation itself will teach people responsibility and toleration. As Adrian Oldfield notes, they
place their faith in the activity of participation “as the means whereby individuals may become

good thing for people to display this sort of decency, and that a minimal level of it is a precondition of a functioning
democracy, I do not think this is the fundamental problem for citizenship education. From my point of view, waiters
who are only minimally cheerful to all their customers are morally preferable to waiters who are generally very
cheerful but who are surly to black customers. The latter may display more decency overall, but their behaviour
towards an identifiable group threaten the most basic norms of liberal citizenship. However, I agree with White
that it is important to be sensitive to the cultural variations in norms of civility (White 1992:215). Iris Young makes a
similar point about cultural variations in norms of public reasonableness (Young 1993).
9. For example, real-estate agents have an economic incentive to maintain segregated housing. In any event,
New Right reforms arguably violated the requirements of liberal justice, since cutting welfare benefits, far from
getting the disadvantaged back on their feet, has expanded the underclass. Class inequalities have been
exacerbated, and the working poor and unemployed have been effectively disenfranchised, unable to participate
in the social and political life of the country (Fierlbeck 1991:579). So even if the market taught civic virtue, laissez-
faire capitalism violates the principle that all members of society have an equal opportunity to be active citizens.
 I H S — Will Kymlicka / Education for Citizenship — 9

accustomed to perform the duties of citizenship. Political participation enlarges the minds of
individuals, familiarizes them with interests which lie beyond the immediacy of personal
circumstance and environment, and encourages them to acknowledge that public concerns are
the proper ones to which they should pay attention” (Oldfield 1990a:184).

Unfortunately, this faith in the educative function of participation seems overly optimistic.
Emphasizing participation does not yet explain how to ensure that citizens participate
responsibly – i.e., in a public-spirited, rather than self-interested or prejudiced, way (Mulgan
1991:40–1). Empowered citizens may use their power irresponsibly by pushing for benefits and
entitlements they cannot ultimately afford; or by voting themselves tax breaks and slashing
assistance to the needy; or by “seeking scapegoats in the indolence of the poor, the
strangeness of ethnic minorities, or the insolence and irresponsibility of modern women”
(Fierlbeck 1991:592). Successful political participation requires the ability to create coalitions,
which encourages a partial development of the virtues of justice and public reasonableness. No
one can hope to succeed in political life if they make no effort to listen to or accommodate the
needs and views of others. But in many cases, a winning coalition can be built while ignoring
the claims of marginalized groups. Indeed, if a significant portion of the population is
prejudiced, then ignoring or attacking such group may be the best route to political success.

‘Civil-society theorists’ emphasize the necessity of civility and self-restraint to a healthy

democracy, but deny that either the market or political participation is sufficient to teach these
virtues. Instead, it is in the voluntary organizations of civil society – churches, families, unions,
ethnic associations, cooperatives, environmental groups, neighbourhood associations, support
groups, charities – that we learn the virtues of mutual obligation. As Michael Walzer puts it,
“The civility that makes democratic politics possible can only be learned in the associational
networks” of civil society (Walzer 1992:104).

Because these groups are voluntary, failure to live up to the responsibilities that come with
them is usually met simply with disapproval, rather than legal punishment. Yet because the
disapproval comes from family, friends, colleagues or comrades, it is in many ways a more
powerful incentive to act responsibly than punishment by an impersonal state. It is here that
“human character, competence, and capacity for citizenship are formed”, for it is here that we
internalize the idea of personal responsibility and mutual obligation, and learn the voluntary
self-restraint which is essential to truly responsible citizenship.

The claim that civil society is the “seedbed of civic virtue” (Glendon 1991:109) is essentially an
empirical claim, for which there is little hard evidence one way or the other. 10 It is an old and
venerable view, but it is not obviously true. It may be in the neighbourhood that we learn to be

10. But see Putnam (1993), who provides strong evidence that the reason why some Italian regional
governments function better than others is the number and vitality of civic associations in each region.
I H S — Will Kymlicka / Education for Citizenship — 10

good neighbours, but neighbourhood associations also teach people to operate on the “NIMBY”
(not in my backyard) principle when it comes to the location of group homes or public works.
Similarly, the family is often “a school of despotism” that teaches male dominance over women
(Okin 1992:65); churches often teach deference to authority and intolerance of other faiths;
ethnic groups often teach prejudice against other races, and so on.

Walzer recognizes that most people are “trapped in one or another subordinate relationship,
where the ‘civility’ they learned was deferential rather than independent and active”. In these
circumstances, he says, we have to “reconstruct” the associational network “under new
conditions of freedom and equality”. Similarly, when the activities of some associations “are
narrowly conceived, partial and particularist”, then “they need political correction”. Walzer calls
his view “critical associationalism” to signify that the associations of civil society may need to
be reformed in the light of principles of citizenship (Walzer 1992:106–7).

But this may go too far in the other direction. Rather than supporting voluntary associations,
this approach may unintentionally license wholesale intervention in them. Governments must of
course intervene to protect the rights of people inside and outside the group, if these rights are
threatened. But do we want governments to reconstruct churches, for example, to make them
more internally democratic, or to make sure that their members learn to be critical rather than
deferential? And, in any event, wouldn’t reconstructing churches, families or unions to make
them more internally democratic start to undermine their essentially uncoerced and voluntary
character, which is what supposedly made them the seedbeds of civic virtue?

Indeed, it would be unreasonable to expect churches to teach the virtue of public

reasonableness. Public reasonableness is essential in political debate, but is unnecessary and
sometimes undesirable in the private sphere. It would be absurd to ask church-goers to abstain
from appealing to Scripture in deciding how to run their church.

Civil-society theorists demand too much of voluntary associations in expecting them to be the
main school for, or a small-scale replica of, democratic citizenship. While these associations
may teach civic virtue, that is not their raison d’être. The reason why people join churches,
families, or ethnic organizations is not to learn civic virtue. It is rather to honour certain values,
and enjoy certain human goods, and these motives may have little to do with the promotion of
citizenship. To expect parents or priests to organize the internal life of their groups so as to
maximally promote citizenship is to ignore why these groups exist in the first place. (Some
associations, like the Boy Scouts, are designed to promote citizenship, but they are the
exception not the rule).

It seems then that we cannot rely on the market, the family, or the associations of civil society
to teach civic virtue. People will not automatically learn to engage in public discourse, or to
 I H S — Will Kymlicka / Education for Citizenship — 11

question authority, in any of these spheres, since these spheres are often held together by
private discourse and respect for authority.

This suggests that schools must teach children how to engage in the kind of critical reasoning
and moral perspective that defines public reasonableness. And indeed, as I noted earlier,
promoting these sorts of virtues was one of the fundamental justifications for mandatory
education. But using schools to promote civic virtue raises many controversies, of which I will
briefly examine three – the role of separate schools, the teaching of autonomy, and the
relationship between civic and cultural identities.

3. Citizenship and Separate Schools

The need for citizenship education raises questions about the role of separate schools in a
liberal democracy, particularly religious schools. Various religious groups have sought to
establish separate schools, partly in order to teach their religious doctrine, but also to reduce
the exposure of their children to the members of other religious groups. Most liberal states
have accepted this demand, as a way of respecting parental rights and religious freedom, but
have insisted that such schools teach a core curriculum, including citizenship education.

It is not clear, however, that this compromise position – separate schools with a common
curriculum – provides the appropriate sort of citizenship education. Such schools are obviously
capable of teaching basic facts about government. But as I noted earlier, citizenship education
is not simply a matter of knowledge of political institutions and constitutional principles. It is
also a matter of how we think about and behave towards others, particularly those who differ
from us in their race, religion, class, etc. Liberal citizenship requires cultivating the habit of
civility, and the capacity for public reasonableness, in our interaction with others. Indeed, it is
precisely these habits and capacities which most need to be learned in schools, for they are
unlikely to be learned in smaller groups or associations, like the family, neighbourhood, or
church, which tend to be homogenous in their ethnocultural backgrounds and religious beliefs.

It is not clear that separate religious schools can provide an adequate education in either
civility or public reasonableness. For these virtues are not only, or even primarily, learned
through the explicit curriculum. For example, common schools teach civility not just by telling
students to be nice, but also by insisting that students sit beside students of different races
and religions, and cooperate with them on school projects or sports teams (Gutmann 1987:53).

Similarly, common schools teach public reasonableness not only by telling students that there
are a plurality of religious views in the world, and that reasonable people disagree on the merits
of these views. They also create the social circumstances whereby students can see the
reasonableness of these disagreements. It is not enough to simply tell students that the
majority of the people in the world do not share their religion. So long as one is surrounded by
I H S — Will Kymlicka / Education for Citizenship — 12

people who share one’s faith, one may still succumb to the temptation to think that everyone
who rejects one’s religion is somehow illogical or depraved. To learn public reasonableness,
students must come to know and understand people who are reasonable and decent and
humane, but who do not share their religion. Only in this way can students learn how personal
faith differs from public reasonableness, and where to draw that line. This sort of learning
requires the presence within a classroom of people with varying ethnocultural and religious
backgrounds (Callan 1995).

In these ways, religious schools are limited in their capacity to provide an adequate citizenship
education. Of course, it is important not to idealize common schools, which suffer their own
deficiencies. For example, while common schools in North America typically contain a diversity
of religions, they are more segregated than religious schools by class, race and academic
talent (Gutmann 1987: 115–17). Yet divisions of class and race are equally important obstacles
to civility and public reasonableness as religious divisions. Indeed, one could argue that the
greatest failure of liberal citizenship in the United States is not the division between religious
groups, but the increasing desire of middle-class whites to distance themselves (both
physically and emotionally) from inner-city blacks, or the poor more generally. In terms of
teaching students how to have a public dialogue with the disadvantaged, religious schools may
well do better than a common school in the suburbs full of well-off (but religiously diverse)
whites.

Moreover, it is important to distinguish temporary or transitional separate schooling from

permanent separation. The requirements of liberal citizenship suggest that common schooling
is necessary – or at least highly desirable – at some point in the educational process. But
there is no reason why the entire process should be integrated. Indeed, there are good reasons
for thinking that some children may do best by having their early schooling in separate
schools, beside others who share their background, before moving into a common school later
in the process. For example, this may be true of historically disadvantaged groups (girls,
blacks) who can best develop their self-esteem in an environment free of prejudice (McLaughlin
1992b:122). More generally, schooling within a particular ethnocultural or religious setting may
provide virtues unavailable within the common schools. If common schools do a better job
promoting a shared sense of justice, separate schools may do better at providing children with
a clear sense of what it is to have a stable sense of the good. They may provide a better
environment for developing the capacity for in-depth engagement with a particular cultural
tradition, and for loyalty and commitment to particular projects and relationships (Callan
1995:22–23; McLaughlin 1992b:123–4). There is more than one starting point from which
children can learn liberal citizenship. 11

11. For more detailed explorations of this theme, see Callan 1995; McLaughlin 1992b.
 I H S — Will Kymlicka / Education for Citizenship — 13

The requirement of common schools – even if limited to the later stages of children’s education
– will be rejected by some religious groups, who insist on keeping their children separate and
apart from the rest of society. Should a liberal state impose integrated common schools, in the
name of citizenship education? In answering this, it is worth distinguishing two kinds of
religious groups that might seek exemption from common schooling. Some groups, like the
Amish, voluntarily isolate themselves from the larger society, and avoid participating in either
politics or the mainstream institutions of civil society. They do not vote, or hire employees, or
attempt to influence public policy (except where a proposed policy would jeopardize their
isolation), and seek only to be left alone. Since they do not participate in either politics or civil
society, it is less urgent that they learn the virtues of civility and public reasonableness. Jeff
Spinner calls the Amish “partial citizens”, and he argues that because they have relinquished
the right to participate, they can also be absolved of the responsibilities which accompany that
right, including the responsibility to learn and practice civility and public reasonableness
(Spinner 1994:98). Hence he supports their right to withdraw their children from school at the
age of 14, before they would have to learn about the larger society, or interact with non-Amish
children. Assuming that such groups are small, and sincerely committed to their self-imposed
isolation, they pose no threat to the practice of liberal citizenship in society generally. Such
groups should not be encouraged, since they accept no responsibility to work together with
other citizens to solve the country’s injustices and problems. They are free riders, in a sense,
benefitting from a stable liberal order that they do nothing to help maintain. 12 But a liberal state
can afford a few such free riders.13

By contrast, other religious groups seeking exemption from integrated schools are active
participants in both civil society and politics, and seek to influence public policy generally. This
would include fundamentalist Christians in the United States, or Muslims in Britain. In these
cases, one could argue that, having chosen to exercise their rights as full citizens, they must
accept the sort of education needed to promote responsible citizenship, including the
obligation to attend common schools at some point in the educational process.

4. Citizenship and Personal Autonomy

A related question is whether schools, be they separate or common, should promote the
capacity for individual autonomy. ‘Autonomy’ means different things to different people. I am

12. I am here disagreeing with those who defend the exemption for the Amish by arguing that their separate
schools provide adequate citizenship education. This was the view of the American Supreme Court, which said
that the Amish education system prepared Amish children to be good citizens, since they became productive and
peaceful members of the Amish community (Wisconsin v Yoder 406 US 205 (1972)). However, as I noted earlier,
liberal citizenship requires more than being law -abiding and economically self-sufficient. For a critique of Yoder’s
account of civic responsibilities, see Shapiro and Arneson 1995.
13. As Spinner notes, there are unlikely to be many such groups, since the price of ‘partial citizenship’ is to cut
oneself off from the opportunities and resources of the mainstream society (Spinner 1994: chap. 5).
I H S — Will Kymlicka / Education for Citizenship — 14

using the term to refer to the capacity to rationally reflect on, and potentially revise, our
conceptions of the good life. An autonomous person is capable of reflecting on her current
ends, and assessing whether they are worthy of her continued allegiance. Autonomy, on this
view, is consistent with people endorsing their inherited way of life, if they reflectively prefer it to
the alternatives. But it is inconsistent with an uncritical attitude towards inherited traditions, or
with an unquestioning acceptance of the pronouncements of parents, priests or community
leaders regarding the worth of different ways of life. 14

I did not include autonomy in my list of the basic virtues of liberal citizenship in section 1, and I
do not think that autonomy, in and of itself, is necessary for the practice of democratic
citizenship. However, there are good reasons to think that autonomy will be indirectly promoted
by citizenship education, since it is closely associated, both conceptually and
developmentally, to various civic virtues.

For example, responsible citizenship involves the willingness to hold political authorities
accountable. Hence schools should teach children to be sceptical of the political authorities
who govern in our name, and to be cognizant of the dangers of the abuse of power. As Amy
Gutmann puts it, children at school “must learn not just to behave in accordance with authority
but to think critically about authority if they are to live up to the democratic ideal of sharing
political sovereignty as citizens”. People who “are ruled only by habit and authority... are
incapable of constituting a society of sovereign citizens” (Gutmann 1987:51).

This democratic virtue is exercised in public life, and promoting it does not entail or require
encouraging children to question parental or religious authority in private life. As Galston puts
it, the need to teach children how to evaluate political leaders “does not warrant the conclusion
that the state must (or may) structure public education to foster in children sceptical reflection
on ways of life inherited from parents or local communities” (Galston 1991:253). But there will
likely be some spillover effect. Indeed, there is strong evidence that adolescents’ attitudes
towards authority tend “to be uniform across all the authority figures they encounter”, so that
encouraging scepticism of political authority will likely encourage questioning of familial or
religious authority (Emler and Reicher 1987). Galston himself admits that it is not easy for
schools to promote a child’s willingness to question political authority without undermining her
“unswerving belief in the correctness” of her parents’ way of life.

14. I mean to distinguish this account of autonomy from two other interpretations. On one (Kantian) view, the
exercise of choice is intrinsically valuable, since it is the most distinctly human attribute. On another (Millian) view,
the exercise of choice is valuable insofar as it leads to greater ‘individuality’ – that is, insofar as it leads
individuals to reject traditional ways of life, and construct their own unique way of life. People who reject these
views may nonetheless accept the more modest idea that informed choice is valuable because our current
beliefs about the good may be mistaken, and so it is important for people to be able to assess the value of
alternative ways of life. On this, see Kymlicka 1989:chap. 2.
 I H S — Will Kymlicka / Education for Citizenship — 15

Citizenship education not only involves promoting a certain sort of critical attitude towards
authority, it also involves developing habits of civility and the capacity for public
reasonableness. Both of these indirectly promote autonomy, since they encourage children to
interact with the members of other groups, to understand the reasonableness of other ways of
life, and to distance themselves from their own cultural traditions.

Consider civility. In section 1, I emphasized that norms of civility and non-discrimination protect
ethnic and religious groups from prejudice and discrimination. This means that groups wishing
to maintain their group identity and cultural practices will face fewer legal barriers or social
stigmas. But civility also increases the interaction between the members of different groups,
and hence the likelihood that individuals will learn and adopt new ways of life. Historically,
cultural boundaries have often been maintained by the visible expression of prejudice towards
outsiders; people stayed within their group because they were not welcome elsewhere. The
spread of civility in social institutions (including schools) means that these boundaries tend to
break down. Members of one group are more likely to cooperate with and befriend children of
other groups, and so learn about other ways of life, and possibly adopt new identities and
practices.

Simply by teaching and practising civility, schools make this sort of mingling and fraternizing
between the members of different groups more likely, and hence make the breakdown of
cultural barriers more likely. In some cases, adopting other ways of life may be done in an
unreflective way, simply imitating one’s peers, and hence does not count as the exercise of
autonomy. But schools also promote a more reflective process, by teaching the virtue of public
reasonableness. Because reasonable people disagree about the merits of different religions
and conceptions of the good life, children must learn to distinguish reasons based on private
faith from reasons that can be publicly accepted in a diverse society. To develop this capacity,
children must not only learn how to distance themselves from beliefs that are taken for granted
in their private life, but they must also learn to put themselves in other people’s shoes, in order
to see what sorts of reasons might be acceptable to people from other backgrounds. The virtue
of public reasonableness does not require that children come to admire or cherish other ways
of life. But it does require that children be exposed to competing ways of life, and be
encouraged to view them as the expressions of coherent conceptions of value which have been
sincerely affirmed by other reasonable people. Learning to view other ways of life in this way
does not inevitably lead to the questioning of one’s own way of life, but it surely makes it more
likely, since it requires a sort of broad-mindedness which is hard to combine with an
unreflective deference to traditional practices or authorities.

For all these reasons, education for democratic citizenship will almost unavoidably, albeit
indirectly, promote autonomy. Through citizenship education children both become aware of
alternative ways of life, and are given the intellectual skills needed to understand and
appreciate them. As Gutmann puts it, citizenship education involves “equipping children with
I H S — Will Kymlicka / Education for Citizenship — 16

the intellectual skills necessary to evaluate ways of life different from that of their parents”,
because “many if not all of the capacities necessary for choice among good lives are also
necessary for choice among good societies” (Gutmann 1987:30,40). Democratic citizenship
and personal autonomy, while distinct, are interconnected at various levels.

As a result, those groups which rely heavily on an uncritical acceptance of tradition and
authority, while not strictly ruled out, are bound to be discouraged by the critical and tolerant
attitudes which civic education encourages (Macedo 1990:53–4). This indeed is why religious
groups often seek to establish separate schools, even when they have to teach a common
curriculum. They fear that if their children attend common schools, they will be more likely to
question traditional practices, even if the school curriculum does not directly promote this sort
of autonomous attitude. To preserve an uncritical deference to communal traditions, children
can only be exposed to a minimal level of citizenship education, one that teaches facts about
government, but not civility, public reasonableness, or critical attitudes to political authority.

I should note two qualifications here. First, citizenship education historically has often
discouraged, rather than encouraged, autonomy. The aim of citizenship education, in the past,
was to promote an unreflective patriotism, one which glorifies the past history and current
political system of the country, and which vilifies opponents of that political system, whether
they be internal dissidents or external enemies (Nelson 1980). This sort of civic education,
needless to say, promoted passivity and deference, not a critical attitude towards political
authority or a broad-mindedness towards cultural differences. Today, however, educational
theorists and policy-makers increasingly reject this model of civic education, in favour of one
that promotes more active and reflective forms of citizenship. 15 The earlier form of civic
education can still be found, of course, and some people continue to defend it (see Galston
1991: 244; AASA 1987:26). However, if our aim is to produce self-governing democratic
citizens, rather than passive subjects of an authoritarian government, a different sort of civic
education is required, one which is much more likely to promote autonomy.

Second, I have suggested that the promotion of personal autonomy should be seen as the
indirect consequence of civic education, not as its direct or explicit purpose. I do not mean to
deny, however, that there might be other reasons for directly promoting personal autonomy.
Indeed, a strong case could be made that promoting autonomy is an integral part of an
adequate education for modern life. While autonomy may not be needed to fulfil the social role
of citizen, it may be needed if children are to enjoy life to the greatest extent possible. If so,
then children may have a right to an autonomy-promoting education, even where their parents

15. We can mark this shift by comparing two accounts of the relationship between civic education and moral
reasoning. Writing in 1980, Jack Nelson objected that contemporary accounts of civic education promoted
passive deference, and so conflicted with the sort of autonomy which he felt was required by true moral agents.
By 1991, however, William Galston was arguing that contemporary accounts of civic education excessively
promote critical reflectiveness, and so undermined the sort of moral identity and moral commitment underlying
many religious groups.
 I H S — Will Kymlicka / Education for Citizenship — 17

resist it. To pursue this question however would raise issues that go far beyond this paper.
While I am myself attracted to the view that schools should promote autonomy, it would be
misleading to defend this as a precondition of democratic citizenship. Autonomy, I think, is
valuable not because it makes people better citizens, but because it enables people to lead
more fulfilling lives, quite independently of their role as citizens.16

5. Citizenship and National Identity

Finally, l want to briefly address the issue of identity. As I noted earlier, many commentators
argue that social unity in a liberal democracy rests not on a shared identity, but rather on
shared allegiance to political principles. As Rawls puts it, “although a well-ordered society is
divided and pluralistic... public agreement on questions of political and social justice supports
ties of civic friendship and secures the bonds of association” (Rawls 1980: 540; Strike 1994:8).
On this view, by teaching certain common political principles – like principles of justice,
tolerance and civility – citizenship education provides the foundation for national unity as well.

I think this is a mistake. Shared political principles obviously are helpful to maintain social
unity, and indeed deep conflict over basic principles can lead to civil war. But shared principles
are not sufficient. Consider the case of Canada. As a result of the rapid liberalization of
Quebecois society since the 1960s, there has been a pronounced convergence in political
principles between English- and French-speaking Canadians over the last 30 years, so that it
would now be “difficult to identify consistent differences in attitudes on issues such as moral
values, prestige ranking of professions, role of the government, workers’ rights, aboriginal
rights, equality between the sexes and races, and conception of authority” (Dion 1991:301). If
the ‘shared principles’ approach were correct, we should have witnessed a decline in support
for Quebec secession over this period, yet nationalist sentiment has in fact grown consistently.

The fact that anglophones and francophones in Canada share the same principles of justice is
not a strong reason to remain together, since the Québécois rightly assume that their own
national state could respect the same principles. Deciding to secede would not require them to
abandon their political principles, since they could implement the same principles in their own
state.

16. I am skipping lightly over a very deep division within liberal political philosophy. There is an important debate
between “political” or “pragmatic” liberals and “comprehensive” or “ethical” liberals over the role of autonomy with
liberal theory. Political liberals, like John Rawls and Charles Larmore, argue that because many groups within
society do not value autonomy, liberals must look for a way of justifying liberal institutions that does not appeal to
such a ‘sectarian’ value (Rawls 1993; Larmore 1987). Comprehensive liberals, like Joseph Raz, argue that liberal
institutions can only be defended by appealing to the value of autonomy (Raz 1986). I discuss this debate, and
defend the comprehensive liberal option, in Kymlicka 1995: chap. 8. See also Callan (1994b), who argues that the
distinction between political and comprehensive liberalism cannot be sustained in the educational context.
However, for a critique of the emphasis on autonomy, and a defense of Muslim demands for a separate school
system that restricts the development of autonomy, see Halstead 1990; 1991.
I H S — Will Kymlicka / Education for Citizenship — 18

The fact that increased Quebec nationalism has gone hand in hand with increased
convergence on political principles is often seen as a ‘paradox’. But it reflects a very general
trend. Consider the Flemish in Belgium, or the Basques in Spain. Throughout the West, an
increasing convergence on liberal values has gone hand in hand with continued, even
increasing, demands for self-government by ethnonational minorities.

This suggests that shared political principles are not sufficient for social unity. The fact that two
groups share the same principles of justice does not necessarily give them a strong reason to
remain together, rather than splitting into two separate countries.17 If two groups want to live
together under a single state, then sharing political principles makes it easier to do so. But
shared political principles is not, in and of itself, a reason why two groups should want to live
together.

Social unity, then, requires not only shared principles, but also a sense of shared membership.
Citizens must have a sense of belonging to the same community, and a shared desire to
continue to live together. Social unity, in short, requires that citizens identify their fellow
citizens as one of ‘us’. This sense of shared identity helps sustain the relationships of trust
and solidarity needed for citizens to accept the results of democratic decisions, and the
obligations of liberal justice (Miller 1995).

What underlies this shared national identity? In non-liberal states, shared identity is typically
based on a common ethnic descent, religious faith, or conception of the good. However, these
cannot provide the basis for social unity in a liberal state, since none of them are shared in
modern pluralist states.

What then makes citizens in a liberal state feel that they belong together, that they are
members of the same nation? The answer typically involves a sense of shared history, and a
common language. Citizens share a sense of belonging to a particular historical society
because they share a language and history; they participate in common social and political
institutions which are based on this shared language, and which manifest and perpetuate this
shared history; and they see their life-choices as bound up with the survival of this society and
its institutions into the indefinite future. Citizens can share a national identity in this sense, and
yet share very little in terms of ethnicity, religion, or conceptions of the good. 18

17. For a more detailed development of this argument, see Norman 1995. For a related critique of the idea that
shared principles underlie social unity, see Paris 1991.
18. This is a thumbnail sketch of the nature of national identity in a liberal state, and its role in promoting political
stability and relationships of trust and solidarity. For accounts of liberal nationalism, see Tamir 1993; Spinner
1994:chap. 7; Miller 1995; and Kymlicka 1995.
 I H S — Will Kymlicka / Education for Citizenship — 19

The need for this sort of common national identity raises many questions for citizenship
education. I will focus on two, regarding the teaching of languages, and the teaching of history,
both of which are fundamental to the construction of a national identity.

First, what should be the language of the school system? This is a remarkably neglected
question in liberal theory. 19 The need for a common national identity suggests that states
should inculcate a common language. And indeed the definition, standardization and teaching
of an official language has been one of the first tasks of ‘nation-building’ throughout the world
(Gellner 1983).

But whether imposing a common language promotes social unity depends on the
circumstances. The historical evidence suggests that voluntary immigrant groups are willing to
adopt the language of the mainstream society. They have already uprooted themselves from
their original homeland, and know that the success of their decision to emigrate depends on
some measure of integration into their host society. Insofar as they demand education in their
mother-tongue, it is in addition to, or as a means of facilitating, learning the common language,
not as a substitute. Much of the opposition to bilingual education for immigrant groups is, I
think, misguided, but liberal states have a legitimate interest in ensuring that these programs
do ultimately lead to competence in the language of the mainstream society.

The case of territorially concentrated language groups whose homeland has been incorporated
into larger states – like the Québécois, Puerto Ricans, or Flemish, or indigenous peoples
around the world – is very different. They have strongly – even violently – resisted the attempt
to have the majority language imposed on them. This reflects the fact that they typically view
themselves as forming their own ‘nation’ or ‘people’, and so have their own sense of national
identity, with their own language, history, and encompassing social institutions.

States with such groups are not nation-states, but multi-nation states, and attempts to impose
a single national identity on these ‘national minorities’ are likely to undermine rather than
promote social unity. Multination states are most stable if they are seen as a federation of
peoples, each with their own historic territories, language rights, and powers of self-
government.20

In multination states, then, citizenship education typically has a dual function – it promotes a
national identity within each constituent national group, defined by a common language and
history, but it also seeks to promote some sort of transnational identity which can bind

19. As Brian Weinstein put it, political theorists have had a lot to say about “the language of politics” – that is, the
symbols, metaphors and rhetorical devices of political discourse – but have had virtually nothing to say about “the
politics of language” – that is, the decisions about which languages to use in political, legal and educational
forums (Weinstein 1983:7–13).
20. For evidence of this, see Gurr 1993; Hannum 1990.
I H S — Will Kymlicka / Education for Citizenship — 20

together the various national groups within the state. Unfortunately, recent developments in
multination states – eg. the breakdown of Yugoslavia and Czechoslovakia, the constitutional
crises in Belgium and Canada – suggest that it is very difficult to construct and maintain this
transnational identity. And indeed schools have little idea how to go about promoting this
identity.

This points to an important gap in political and educational theory. Most liberal accounts of
civic identity argue that shared political principles are the basis of civic identity. Implicitly,
however, they typically assume that citizens share not only principles, but also a common
language and sense of membership in a national community. The problem is that neither the
explicit emphasis on principles, nor the implicit emphasis on shared language and history, can
explain social unity in multination states. If schools are to fulfil their responsibilities regarding
citizenship education, we need an entirely new account of the basis of shared identity in
multination states.

Insofar as a common national identity rests on identifying with a shared history, as well as a
common language, this raises important questions about the teaching of history. One way – a
particularly effective way – to promote identification with a group’s history is to deliberately
misrepresent that history. As William Galston puts it, in reference to the United States,
“rigorous historical research will almost certainly vindicate complex ‘revisionist’ accounts of key
figures in American history. Civic education, however, requires a nobler, moralizing history: a
pantheon of heroes who confer legitimacy on central institutions and are worthy of emulation”
(Galston 1991:244). Similarly, Andrew Oldenquist argues that information about the American
nation and government

should be taught so as to provide grounds for developing pride and affection... If instead
we start nine-year-olds with a litany of evils and injustices, they will be likely to learn
cynicism and alienation. A teacher may respond, ‘But I teach about problems and
injustices because I want to make my country better; if I did not have concern and
affection for it I would not care about reforming it’. Precisely. The teacher did not
acquire affection for our country by being told that we exterminated Indians, lynched
Blacks, and slaughtered Vietnamese. The teacher’s concern and affection survived this
knowledge because of prior training and experience, and the pupils, like the teacher,
need to acquire a basis for good citizenship before they are plunged into what is ugly”
(AASA:1987:26).

This raises a number of troubling questions about citizenship education. For one thing, this
way of promoting a national identity may undermine another goal of citizenship education – i.e.
the development of the capacity for independent and critical thought about society and its
problems. Moreover, the proper development of civic virtue may require an honest appreciation
of how those virtues were lacking in our history. It seems unlikely that children can learn the
 I H S — Will Kymlicka / Education for Citizenship — 21

true meaning of civility and public reasonableness when historical figures who were in fact
insensitive to great injustices are held up as exemplars of civic virtue (Callan 1994a).

Also, it seems clear that the sanitized version of history that Galston and Oldenquist defend
can itself be a cause of disunity. An account of history that focuses on the “pantheon of
heroes”, while ignoring the historical mistreatment of women, blacks, Indians, Jews etc., is
essentially an account of the history of upper-class white men. And it is precisely this view of
history which many minorities find so offensive. They are insulted by the way their struggles
are rendered invisible in school books.

For these reasons, schools should, I think, teach history truthfully. But that doesn’t mean that
history should not play a special role in the curriculum. There is, I think, a legitimate role for
schools to promote an emotional identification with our history. Students should view the
nation’s history as their history, and hence take pride in its accomplishments, as well as
shame in its injustices. This sense of identification with the nation’s history is one of the few
means available to maintain social unity in a pluralistic state, and may be needed if citizens
are to embrace their responsibilities for upholding just institutions, and rectifying historical
injustices.21

This shows, yet again, that citizenship education is not simply a matter of teaching the basic
facts about governmental institutions or constitutional principles. It is also a matter of
inculcating particular habits, virtues and identities.

Bibliography

American Association of School Administrators (1987) Citizenship: goal of education (AASA

Publications, Arlington).

Arneson, Richard and Ian Shapiro (1996) “Democracy and Religious Freedom: A Critique of
Wisconsin v Yoder”, in NOMOS 38: Political Order (New York University Press, New York,
forthcoming).

Audi, Robert (1989) “The Separation of church and state and the obligations of citizenship”
Philosophy & Public Affairs 18/3, pp. 259–96.

Beiner, Ronald (1992). “Citizenship”, in What’s the Matter with Liberalism? (University of
California Press, Berkeley).

Beiner, Ronald, ed. (1995) Theorizing Citizenship (State University of New York Press, Albany).

21. For a sensitive exploration of this issue, see Callan 1994a.

I H S — Will Kymlicka / Education for Citizenship — 22

Cairns, Alan and Cynthia Williams, eds. (1985) Constitutionalism, Citizenship and Society in
Canada (University of Toronto Press, Toronto).

Callan, Eamonn (1994a) “Beyond Sentimental Civic Education”, American Journal of Education
102

Callan, Eamonn (1994b) “Political Liberalism and Political Education” (unpublished).

Callan, Eamonn (1995) “Common Schools for Common Education”, forthcoming in Canadian
Journal of Education.

Clarke, Paul Barry, ed., (1994) Citizenship (Pluto Press, London).

Cuddihy, John Murray (1978) No Offense: Civil Religion and Protestant Taste (Seabury Press,
New York).

Dion, Stéphane (1991) “Le Nationalisme dans la Convergence Culturelle”, in L’Engagement

Intellectuel: Melanges en l’honneur de Léon Dion, eds. R. Hudon and R. Pelletier (Les
Presses de l’Université Laval, Sainte-Foy).

Emler, N. and Reicher S. (1987) “Orientations to Institutional Authority in Adolescents”, Journal

of Moral Education 16/2, pp. 108–116.

Fierlbeck, Katherine (1991) “Redefining responsibilities: the politics of citizenship in the United
Kingdom”, Canadian Journal of Political Science 24/3, pp. 575–83.

Galston, William (1991) Liberal Purposes: goods, virtues, and duties in the liberal state
(Cambridge University Press, Cambridge).

Gellner, Ernest (1983) Nations and Nationalism (Blackwell, Oxford).

Glendon, Mary Ann (1991). Rights Talk: the impoverishment of political discourse (Free Press,
New York).

Gurr, Ted (1993) Minorities at Risk: A Global View of Ethnopolitical Conflict (Institute of Peace
Press, Washington).

Gutmann, Amy (1987) Democratic Education (Princeton University Press, Princeton).

Halstead, J.M. (1990) “Muslim Schools and the Ideal of Autonomy”, Ethics in Education 9/4,
pp. 4–6.
 I H S — Will Kymlicka / Education for Citizenship — 23

Halstead, Mark (1991) “Radical Feminism, Islam and the Single-Sex School Debate”, Gender
and Education 3/1, pp. 263–78.

Hannum, Hurst (1990) Autonomy, Sovereignty, and Self-Determination: The Adjudication of

Conflicting Rights (University of Pennsylvania Press, Philadelphia).

Heater, Derek (1990) Citizenship: the civic ideal in world history, politics and education
(Longman, London).

Kymlicka, Will (1989) Liberalism, Community and Culture (Oxford University Press, Oxford).

Kymlicka, Will (1995) Multicultural Citizenship: A Liberal Theory of Minority Rights (Oxford
University Press, Oxford).

Kymlicka, Will and Wayne Norman (1994) “Return of the Citizen: A Survey of Recent Work on
Citizenship Theory”, Ethics, Vol. 104/2, 1994, pp. 352–81. Reprinted in Beiner 1995.

Larmore, Charles (1987) Patterns of Moral Complexity (Cambridge University Press,

Cambridge).

Macedo, Stephen (1990) Liberal Virtues: citizenship, virtue and community (Oxford University
Press, Oxford).

McLaughlin, T.H. (1992a) “Citizenship, Diversity and Education”, Journal of Moral Education
21/3, pp. 235–50.

McLaughlin, T.H. (1992b) “The Ethics of Separate Schools”, in Mal Leicester and Monica
Taylor, eds., Ethics, Ethnicity and Education (Kogan Page: London), pp. 114–136.

Miller, David (1995) On Nationality (Oxford University Press, Oxford, forthcoming).

Mulgan, Geoff (1991) “Citizens and Responsibilities”, in Geoff Andrews, ed., Citizenship
(Lawrence and Wishart, London), pp. 37–49.

Nelson, Jack (1980) “The Uncomfortable Relationship between Moral Education and
Citizenship Instruction”, in Richard Wilson and Gordon Schochet, eds., Moral
Development and Politics (Praeger), pp. 256–85.

Norman, Wayne (1995) “The Ideology of Shared Values”, in Joseph Carens, ed., Is Quebec
Nationalism Just? (McGill-Queen’s University Press, Montreal).
I H S — Will Kymlicka / Education for Citizenship — 24

Okin, Susan (1992) “Women, Equality and Citizenship”, Queen’s Quarterly 99/1, pp. 56–71.

Oldfield, Adrian (1990a) “Citizenship: An Unnatural Practice?” Political Quarterly 61, pp. 177–
87.

Oldfield, Adrian (1990b) Citizenship and Community: civic republicanism and the modern world
(Routledge, London).

Paris, David (1991) “Moral Education and the ‘Tie that Binds’ in Liberal Political Theory”,
American Political Science Review 85/3, pp. 875–901.

Pocock, J.G.A. (1992) “The Ideal of Citizenship Since Classical Times”, Queen’s Quarterly
99/1, pp. 33–55.

Putnam, Robert (1993) Making Democracy Work: Civic Traditions in Modern Italy (Princeton
University Press, Princeton).

Rawls, John (1980) “Kantian Constructivism in Moral Theory”, Journal of Philosophy 77/9, pp.

Rawls, John (1993) Political Liberalism (Columbia University Press, New York).

Raz, Joseph (1986) The Morality of Freedom (Oxford University Press, Oxford).

Skinner, Quentin (1992). “On Justice, the Common Good and the Priority of Liberty”, in Chantal
Mouffe, ed., Dimensions of Radical Democracy: pluralism, citizenship and community
(Routledge, London), pp. 211–24.

Spinner, Jeff (1994) The Boundaries of Citizenship: Race, Ethnicity and Nationality in the
Liberal State (Johns Hopkins University Press, Baltimore).

Strike, Kenneth (1994) “On the Construction of Public Speech: Pluralism and Public Reason”,
Educational Theory 44/1, pp. 1–26.

Tamir, Yael (1993) Liberal Nationalism (Princeton University Press, Princeton).

van Gunsteren, Herman (1978), “Notes Towards a Theory of Citizenship”, in F. Dallmayr, ed.,
From Contract to Community (Marcel Decker, New York).

Walzer, Michael (1989) “Citizenship”, in Political Innovation and Conceptual Change, ed. T. Ball
and J. Farr (Cambridge University Press, Cambridge), pp. 211–19.
 I H S — Will Kymlicka / Education for Citizenship — 25

Walzer, Michael (1992) “The Civil Society Argument”, in Chantal Mouffe, ed., Dimensions of
Radical Democracy: pluralism, citizenship and community (Routledge, London), pp. 89–
107.

Weinstein, Brian (1983) The Civic Tongue: Political Consequences of Language Choice
(Longman, New York).

White, Patricia (1992) “Decency and Education for Citizenship”, Journal of Moral Education,
Vol. 21/3: 207–16.

Young, Iris Marion (1993) “Justice and Communicative Democracy”, in Roger Gottlieb, ed.,
Radical Philosophy: Tradition, Counter-Tradition, Politics (Temple University Press,
Philadelphia), pp. 123–43.